[{"page":"246-259","quality_controlled":"1","series_title":"LNCS","file_date_updated":"2022-08-12T07:27:58Z","publisher":"Springer","_id":"9210","scopus_import":"1","author":[{"full_name":"Volhejn, Vaclav","first_name":"Vaclav","last_name":"Volhejn","id":"d5235fb4-7a6d-11eb-b254-f25d12d631a8"},{"last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"publication_status":"published","article_processing_charge":"No","department":[{"_id":"ChLa"}],"date_created":"2021-03-01T09:01:16Z","title":"Does SGD implicitly optimize for smoothness?","intvolume":"     12544","volume":12544,"ddc":["510"],"date_updated":"2022-08-12T07:28:47Z","year":"2021","citation":{"short":"V. Volhejn, C. Lampert, in:, 42nd German Conference on Pattern Recognition, Springer, 2021, pp. 246–259.","mla":"Volhejn, Vaclav, and Christoph Lampert. “Does SGD Implicitly Optimize for Smoothness?” <i>42nd German Conference on Pattern Recognition</i>, vol. 12544, Springer, 2021, pp. 246–59, doi:<a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">10.1007/978-3-030-71278-5_18</a>.","ista":"Volhejn V, Lampert C. 2021. Does SGD implicitly optimize for smoothness? 42nd German Conference on Pattern Recognition. DAGM GCPR: German Conference on Pattern Recognition LNCS vol. 12544, 246–259.","ama":"Volhejn V, Lampert C. Does SGD implicitly optimize for smoothness? In: <i>42nd German Conference on Pattern Recognition</i>. Vol 12544. LNCS. Springer; 2021:246-259. doi:<a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">10.1007/978-3-030-71278-5_18</a>","apa":"Volhejn, V., &#38; Lampert, C. (2021). Does SGD implicitly optimize for smoothness? In <i>42nd German Conference on Pattern Recognition</i> (Vol. 12544, pp. 246–259). Tübingen, Germany: Springer. <a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">https://doi.org/10.1007/978-3-030-71278-5_18</a>","ieee":"V. Volhejn and C. Lampert, “Does SGD implicitly optimize for smoothness?,” in <i>42nd German Conference on Pattern Recognition</i>, Tübingen, Germany, 2021, vol. 12544, pp. 246–259.","chicago":"Volhejn, Vaclav, and Christoph Lampert. “Does SGD Implicitly Optimize for Smoothness?” In <i>42nd German Conference on Pattern Recognition</i>, 12544:246–59. LNCS. Springer, 2021. <a href=\"https://doi.org/10.1007/978-3-030-71278-5_18\">https://doi.org/10.1007/978-3-030-71278-5_18</a>."},"doi":"10.1007/978-3-030-71278-5_18","day":"17","abstract":[{"lang":"eng","text":"Modern neural networks can easily fit their training set perfectly. Surprisingly, despite being “overfit” in this way, they tend to generalize well to future data, thereby defying the classic bias–variance trade-off of machine learning theory. Of the many possible explanations, a prevalent one is that training by stochastic gradient descent (SGD) imposes an implicit bias that leads it to learn simple functions, and these simple functions generalize well. However, the specifics of this implicit bias are not well understood.\r\nIn this work, we explore the smoothness conjecture which states that SGD is implicitly biased towards learning functions that are smooth. We propose several measures to formalize the intuitive notion of smoothness, and we conduct experiments to determine whether SGD indeed implicitly optimizes for these measures. Our findings rule out the possibility that smoothness measures based on first-order derivatives are being implicitly enforced. They are supportive, though, of the smoothness conjecture for measures based on second-order derivatives."}],"language":[{"iso":"eng"}],"conference":{"start_date":"2020-09-28","name":"DAGM GCPR: German Conference on Pattern Recognition ","location":"Tübingen, Germany","end_date":"2020-10-01"},"publication":"42nd German Conference on Pattern Recognition","has_accepted_license":"1","oa_version":"Submitted Version","month":"03","file":[{"date_updated":"2022-08-12T07:27:58Z","file_name":"2020_GCPR_submitted_Volhejn.pdf","content_type":"application/pdf","date_created":"2022-08-12T07:27:58Z","file_size":420234,"checksum":"3e3628ab1cf658d82524963f808004ea","file_id":"11820","creator":"dernst","success":1,"access_level":"open_access","relation":"main_file"}],"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2021-03-17T00:00:00Z","type":"conference","publication_identifier":{"isbn":["9783030712778"],"eissn":["1611-3349"],"issn":["0302-9743"]},"oa":1},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"9418"}]},"ddc":["000"],"status":"public","file":[{"date_created":"2021-05-24T11:15:57Z","checksum":"f34ff17017527db5ba6927f817bdd125","file_size":502356,"date_updated":"2021-05-24T11:15:57Z","file_name":"iclr2021_conference.pdf","content_type":"application/pdf","relation":"main_file","access_level":"open_access","file_id":"9417","creator":"bphuong"}],"main_file_link":[{"url":"https://openreview.net/pdf?id=krz7T0xU9Z_","open_access":"1"}],"abstract":[{"lang":"eng","text":"We study the inductive bias of two-layer ReLU networks trained by gradient flow. We identify a class of easy-to-learn (`orthogonally separable') datasets, and characterise the solution that ReLU networks trained on such datasets converge to. Irrespective of network width, the solution turns out to be a combination of two max-margin classifiers: one corresponding to the positive data subset and one corresponding to the negative data subset. The proof is based on the recently introduced concept of extremal sectors, for which we prove a number of properties in the context of orthogonal separability. In particular, we prove stationarity of activation patterns from some time  onwards, which enables a reduction of the ReLU network to an ensemble of linear subnetworks."}],"oa":1,"day":"01","date_published":"2021-05-01T00:00:00Z","type":"conference","date_updated":"2023-09-07T13:29:50Z","citation":{"ama":"Phuong M, Lampert C. The inductive bias of ReLU networks on orthogonally separable data. In: <i>9th International Conference on Learning Representations</i>. ; 2021.","apa":"Phuong, M., &#38; Lampert, C. (2021). The inductive bias of ReLU networks on orthogonally separable data. In <i>9th International Conference on Learning Representations</i>. Virtual.","chicago":"Phuong, Mary, and Christoph Lampert. “The Inductive Bias of ReLU Networks on Orthogonally Separable Data.” In <i>9th International Conference on Learning Representations</i>, 2021.","ieee":"M. Phuong and C. Lampert, “The inductive bias of ReLU networks on orthogonally separable data,” in <i>9th International Conference on Learning Representations</i>, Virtual, 2021.","short":"M. Phuong, C. Lampert, in:, 9th International Conference on Learning Representations, 2021.","mla":"Phuong, Mary, and Christoph Lampert. “The Inductive Bias of ReLU Networks on Orthogonally Separable Data.” <i>9th International Conference on Learning Representations</i>, 2021.","ista":"Phuong M, Lampert C. 2021. The inductive bias of ReLU networks on orthogonally separable data. 9th International Conference on Learning Representations.  ICLR: International Conference on Learning Representations."},"year":"2021","conference":{"location":"Virtual","end_date":"2021-05-07","name":" ICLR: International Conference on Learning Representations","start_date":"2021-05-03"},"language":[{"iso":"eng"}],"file_date_updated":"2021-05-24T11:15:57Z","quality_controlled":"1","title":"The inductive bias of ReLU networks on orthogonally separable data","month":"05","oa_version":"Published Version","publication_status":"published","article_processing_charge":"No","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"date_created":"2021-05-24T11:16:46Z","author":[{"last_name":"Bui Thi Mai","first_name":"Phuong","full_name":"Bui Thi Mai, Phuong","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert"}],"publication":"9th International Conference on Learning Representations","_id":"9416","has_accepted_license":"1","scopus_import":"1"},{"page":"125","file_date_updated":"2021-05-24T11:56:02Z","publisher":"Institute of Science and Technology Austria","_id":"9418","author":[{"id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","full_name":"Bui Thi Mai, Phuong","first_name":"Phuong","last_name":"Bui Thi Mai"}],"department":[{"_id":"GradSch"},{"_id":"ChLa"}],"date_created":"2021-05-24T13:06:23Z","article_processing_charge":"No","publication_status":"published","alternative_title":["ISTA Thesis"],"title":"Underspecification in deep learning","ddc":["000"],"citation":{"mla":"Phuong, Mary. <i>Underspecification in Deep Learning</i>. Institute of Science and Technology Austria, 2021, doi:<a href=\"https://doi.org/10.15479/AT:ISTA:9418\">10.15479/AT:ISTA:9418</a>.","short":"M. Phuong, Underspecification in Deep Learning, Institute of Science and Technology Austria, 2021.","ista":"Phuong M. 2021. Underspecification in deep learning. Institute of Science and Technology Austria.","ama":"Phuong M. Underspecification in deep learning. 2021. doi:<a href=\"https://doi.org/10.15479/AT:ISTA:9418\">10.15479/AT:ISTA:9418</a>","apa":"Phuong, M. (2021). <i>Underspecification in deep learning</i>. Institute of Science and Technology Austria. <a href=\"https://doi.org/10.15479/AT:ISTA:9418\">https://doi.org/10.15479/AT:ISTA:9418</a>","chicago":"Phuong, Mary. “Underspecification in Deep Learning.” Institute of Science and Technology Austria, 2021. <a href=\"https://doi.org/10.15479/AT:ISTA:9418\">https://doi.org/10.15479/AT:ISTA:9418</a>.","ieee":"M. Phuong, “Underspecification in deep learning,” Institute of Science and Technology Austria, 2021."},"year":"2021","date_updated":"2023-09-08T11:11:12Z","day":"30","doi":"10.15479/AT:ISTA:9418","degree_awarded":"PhD","abstract":[{"lang":"eng","text":"Deep learning is best known for its empirical success across a wide range of applications\r\nspanning computer vision, natural language processing and speech. Of equal significance,\r\nthough perhaps less known, are its ramifications for learning theory: deep networks have\r\nbeen observed to perform surprisingly well in the high-capacity regime, aka the overfitting\r\nor underspecified regime. Classically, this regime on the far right of the bias-variance curve\r\nis associated with poor generalisation; however, recent experiments with deep networks\r\nchallenge this view.\r\n\r\nThis thesis is devoted to investigating various aspects of underspecification in deep learning.\r\nFirst, we argue that deep learning models are underspecified on two levels: a) any given\r\ntraining dataset can be fit by many different functions, and b) any given function can be\r\nexpressed by many different parameter configurations. We refer to the second kind of\r\nunderspecification as parameterisation redundancy and we precisely characterise its extent.\r\nSecond, we characterise the implicit criteria (the inductive bias) that guide learning in the\r\nunderspecified regime. Specifically, we consider a nonlinear but tractable classification\r\nsetting, and show that given the choice, neural networks learn classifiers with a large margin.\r\nThird, we consider learning scenarios where the inductive bias is not by itself sufficient to\r\ndeal with underspecification. We then study different ways of ‘tightening the specification’: i)\r\nIn the setting of representation learning with variational autoencoders, we propose a hand-\r\ncrafted regulariser based on mutual information. ii) In the setting of binary classification, we\r\nconsider soft-label (real-valued) supervision. We derive a generalisation bound for linear\r\nnetworks supervised in this way and verify that soft labels facilitate fast learning. Finally, we\r\nexplore an application of soft-label supervision to the training of multi-exit models."}],"language":[{"iso":"eng"}],"has_accepted_license":"1","oa_version":"Published Version","acknowledged_ssus":[{"_id":"ScienComp"},{"_id":"CampIT"},{"_id":"E-Lib"}],"month":"05","file":[{"creator":"bphuong","file_id":"9419","relation":"main_file","success":1,"access_level":"open_access","file_name":"mph-thesis-v519-pdfimages.pdf","content_type":"application/pdf","date_updated":"2021-05-24T11:22:29Z","checksum":"4f0abe64114cfed264f9d36e8d1197e3","file_size":2673905,"date_created":"2021-05-24T11:22:29Z"},{"file_id":"9420","creator":"bphuong","access_level":"closed","relation":"source_file","date_updated":"2021-05-24T11:56:02Z","content_type":"application/zip","file_name":"thesis.zip","date_created":"2021-05-24T11:56:02Z","checksum":"f5699e876bc770a9b0df8345a77720a2","file_size":92995100}],"status":"public","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","related_material":{"record":[{"status":"deleted","relation":"part_of_dissertation","id":"7435"},{"id":"7481","relation":"part_of_dissertation","status":"public"},{"status":"public","relation":"part_of_dissertation","id":"9416"},{"relation":"part_of_dissertation","id":"7479","status":"public"}]},"type":"dissertation","date_published":"2021-05-30T00:00:00Z","publication_identifier":{"issn":["2663-337X"]},"oa":1,"supervisor":[{"first_name":"Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}]},{"external_id":{"arxiv":["2004.12623"]},"citation":{"apa":"Royer, A., &#38; Lampert, C. (2020). Localizing grouped instances for efficient detection in low-resource scenarios. In <i>IEEE Winter Conference on Applications of Computer Vision</i>.  Snowmass Village, CO, United States: IEEE. <a href=\"https://doi.org/10.1109/WACV45572.2020.9093288\">https://doi.org/10.1109/WACV45572.2020.9093288</a>","ama":"Royer A, Lampert C. Localizing grouped instances for efficient detection in low-resource scenarios. In: <i>IEEE Winter Conference on Applications of Computer Vision</i>. IEEE; 2020. doi:<a href=\"https://doi.org/10.1109/WACV45572.2020.9093288\">10.1109/WACV45572.2020.9093288</a>","ieee":"A. Royer and C. Lampert, “Localizing grouped instances for efficient detection in low-resource scenarios,” in <i>IEEE Winter Conference on Applications of Computer Vision</i>,  Snowmass Village, CO, United States, 2020.","chicago":"Royer, Amélie, and Christoph Lampert. “Localizing Grouped Instances for Efficient Detection in Low-Resource Scenarios.” In <i>IEEE Winter Conference on Applications of Computer Vision</i>. IEEE, 2020. <a href=\"https://doi.org/10.1109/WACV45572.2020.9093288\">https://doi.org/10.1109/WACV45572.2020.9093288</a>.","mla":"Royer, Amélie, and Christoph Lampert. “Localizing Grouped Instances for Efficient Detection in Low-Resource Scenarios.” <i>IEEE Winter Conference on Applications of Computer Vision</i>, 1716–1725, IEEE, 2020, doi:<a href=\"https://doi.org/10.1109/WACV45572.2020.9093288\">10.1109/WACV45572.2020.9093288</a>.","short":"A. Royer, C. Lampert, in:, IEEE Winter Conference on Applications of Computer Vision, IEEE, 2020.","ista":"Royer A, Lampert C. 2020. Localizing grouped instances for efficient detection in low-resource scenarios. IEEE Winter Conference on Applications of Computer Vision. WACV: Winter Conference on Applications of Computer Vision, 1716–1725."},"year":"2020","date_updated":"2023-09-07T13:16:17Z","abstract":[{"lang":"eng","text":"State-of-the-art detection systems are generally evaluated on their ability to exhaustively retrieve objects densely distributed in the image, across a wide variety of appearances and semantic categories. Orthogonal to this, many real-life object detection applications, for example in remote sensing, instead require dealing with large images that contain only a few small objects of a single class, scattered heterogeneously across the space. In addition, they are often subject to strict computational constraints, such as limited battery capacity and computing power.To tackle these more practical scenarios, we propose a novel flexible detection scheme that efficiently adapts to variable object sizes and densities: We rely on a sequence of detection stages, each of which has the ability to predict groups of objects as well as individuals. Similar to a detection cascade, this multi-stage architecture spares computational effort by discarding large irrelevant regions of the image early during the detection process. The ability to group objects provides further computational and memory savings, as it allows working with lower image resolutions in early stages, where groups are more easily detected than individuals, as they are more salient. We report experimental results on two aerial image datasets, and show that the proposed method is as accurate yet computationally more efficient than standard single-shot detectors, consistently across three different backbone architectures."}],"day":"01","arxiv":1,"doi":"10.1109/WACV45572.2020.9093288","author":[{"first_name":"Amélie","last_name":"Royer","orcid":"0000-0002-8407-0705","full_name":"Royer, Amélie","id":"3811D890-F248-11E8-B48F-1D18A9856A87"},{"orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"scopus_import":1,"_id":"7936","title":"Localizing grouped instances for efficient detection in low-resource scenarios","article_processing_charge":"No","department":[{"_id":"ChLa"}],"date_created":"2020-06-07T22:00:53Z","publication_status":"published","quality_controlled":"1","publisher":"IEEE","type":"conference","date_published":"2020-03-01T00:00:00Z","oa":1,"publication_identifier":{"isbn":["9781728165530"]},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","related_material":{"record":[{"status":"deleted","id":"8331","relation":"dissertation_contains"},{"status":"public","relation":"dissertation_contains","id":"8390"}]},"status":"public","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2004.12623"}],"publication":"IEEE Winter Conference on Applications of Computer Vision","article_number":"1716-1725","month":"03","oa_version":"Preprint","language":[{"iso":"eng"}],"conference":{"end_date":"2020-03-05","location":" Snowmass Village, CO, United States","name":"WACV: Winter Conference on Applications of Computer Vision","start_date":"2020-03-01"}},{"author":[{"id":"3811D890-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8407-0705","full_name":"Royer, Amélie","first_name":"Amélie","last_name":"Royer"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887"}],"_id":"7937","scopus_import":"1","title":"A flexible selection scheme for minimum-effort transfer learning","publication_status":"published","department":[{"_id":"ChLa"}],"article_processing_charge":"No","date_created":"2020-06-07T22:00:53Z","quality_controlled":"1","publisher":"IEEE","external_id":{"arxiv":["2008.11995"]},"date_updated":"2023-09-07T13:16:17Z","citation":{"chicago":"Royer, Amélie, and Christoph Lampert. “A Flexible Selection Scheme for Minimum-Effort Transfer Learning.” In <i>2020 IEEE Winter Conference on Applications of Computer Vision</i>. IEEE, 2020. <a href=\"https://doi.org/10.1109/WACV45572.2020.9093635\">https://doi.org/10.1109/WACV45572.2020.9093635</a>.","ieee":"A. Royer and C. Lampert, “A flexible selection scheme for minimum-effort transfer learning,” in <i>2020 IEEE Winter Conference on Applications of Computer Vision</i>, Snowmass Village, CO, United States, 2020.","apa":"Royer, A., &#38; Lampert, C. (2020). A flexible selection scheme for minimum-effort transfer learning. In <i>2020 IEEE Winter Conference on Applications of Computer Vision</i>. Snowmass Village, CO, United States: IEEE. <a href=\"https://doi.org/10.1109/WACV45572.2020.9093635\">https://doi.org/10.1109/WACV45572.2020.9093635</a>","ama":"Royer A, Lampert C. A flexible selection scheme for minimum-effort transfer learning. In: <i>2020 IEEE Winter Conference on Applications of Computer Vision</i>. IEEE; 2020. doi:<a href=\"https://doi.org/10.1109/WACV45572.2020.9093635\">10.1109/WACV45572.2020.9093635</a>","ista":"Royer A, Lampert C. 2020. A flexible selection scheme for minimum-effort transfer learning. 2020 IEEE Winter Conference on Applications of Computer Vision. WACV: Winter Conference on Applications of Computer Vision, 2180–2189.","short":"A. Royer, C. Lampert, in:, 2020 IEEE Winter Conference on Applications of Computer Vision, IEEE, 2020.","mla":"Royer, Amélie, and Christoph Lampert. “A Flexible Selection Scheme for Minimum-Effort Transfer Learning.” <i>2020 IEEE Winter Conference on Applications of Computer Vision</i>, 2180–2189, IEEE, 2020, doi:<a href=\"https://doi.org/10.1109/WACV45572.2020.9093635\">10.1109/WACV45572.2020.9093635</a>."},"year":"2020","abstract":[{"text":"Fine-tuning is a popular way of exploiting knowledge contained in a pre-trained convolutional network for a new visual recognition task. However, the orthogonal setting of transferring knowledge from a pretrained network to a visually different yet semantically close source is rarely considered: This commonly happens with real-life data, which is not necessarily as clean as the training source (noise, geometric transformations, different modalities, etc.).To tackle such scenarios, we introduce a new, generalized form of fine-tuning, called flex-tuning, in which any individual unit (e.g. layer) of a network can be tuned, and the most promising one is chosen automatically. In order to make the method appealing for practical use, we propose two lightweight and faster selection procedures that prove to be good approximations in practice. We study these selection criteria empirically across a variety of domain shifts and data scarcity scenarios, and show that fine-tuning individual units, despite its simplicity, yields very good results as an adaptation technique. As it turns out, in contrast to common practice, rather than the last fully-connected unit it is best to tune an intermediate or early one in many domain- shift scenarios, which is accurately detected by flex-tuning.","lang":"eng"}],"doi":"10.1109/WACV45572.2020.9093635","arxiv":1,"day":"01","publication":"2020 IEEE Winter Conference on Applications of Computer Vision","month":"03","article_number":"2180-2189","oa_version":"Preprint","language":[{"iso":"eng"}],"conference":{"name":"WACV: Winter Conference on Applications of Computer Vision","start_date":"2020-03-01","location":"Snowmass Village, CO, United States","end_date":"2020-03-05"},"date_published":"2020-03-01T00:00:00Z","type":"conference","oa":1,"publication_identifier":{"isbn":["9781728165530"]},"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","related_material":{"record":[{"status":"deleted","relation":"dissertation_contains","id":"8331"},{"status":"public","id":"8390","relation":"dissertation_contains"}]},"main_file_link":[{"open_access":"1","url":"http://arxiv.org/abs/2008.11995"}]},{"ddc":["004"],"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2004.00642"}],"abstract":[{"text":"We present a generative model of images that explicitly reasons over the set\r\nof objects they show. Our model learns a structured latent representation that\r\nseparates objects from each other and from the background; unlike prior works,\r\nit explicitly represents the 2D position and depth of each object, as well as\r\nan embedding of its segmentation mask and appearance. The model can be trained\r\nfrom images alone in a purely unsupervised fashion without the need for object\r\nmasks or depth information. Moreover, it always generates complete objects,\r\neven though a significant fraction of training images contain occlusions.\r\nFinally, we show that our model can infer decompositions of novel images into\r\ntheir constituent objects, including accurate prediction of depth ordering and\r\nsegmentation of occluded parts.","lang":"eng"}],"oa":1,"arxiv":1,"day":"01","date_published":"2020-04-01T00:00:00Z","type":"preprint","external_id":{"arxiv":["2004.00642"]},"date_updated":"2021-01-12T08:16:44Z","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","short":"CC BY-SA (4.0)","name":"Creative Commons Attribution-ShareAlike 4.0 International Public License (CC BY-SA 4.0)","image":"/images/cc_by_sa.png"},"year":"2020","citation":{"mla":"Anciukevicius, Titas, et al. “Object-Centric Image Generation with Factored Depths, Locations, and Appearances.” <i>ArXiv</i>, 2004.00642.","short":"T. Anciukevicius, C. Lampert, P.M. Henderson, ArXiv (n.d.).","ista":"Anciukevicius T, Lampert C, Henderson PM. Object-centric image generation with factored depths, locations, and appearances. arXiv, 2004.00642.","apa":"Anciukevicius, T., Lampert, C., &#38; Henderson, P. M. (n.d.). Object-centric image generation with factored depths, locations, and appearances. <i>arXiv</i>.","ama":"Anciukevicius T, Lampert C, Henderson PM. Object-centric image generation with factored depths, locations, and appearances. <i>arXiv</i>.","chicago":"Anciukevicius, Titas, Christoph Lampert, and Paul M Henderson. “Object-Centric Image Generation with Factored Depths, Locations, and Appearances.” <i>ArXiv</i>, n.d.","ieee":"T. Anciukevicius, C. Lampert, and P. M. Henderson, “Object-centric image generation with factored depths, locations, and appearances,” <i>arXiv</i>. ."},"language":[{"iso":"eng"}],"title":"Object-centric image generation with factored depths, locations, and appearances","month":"04","article_number":"2004.00642","oa_version":"Preprint","publication_status":"submitted","article_processing_charge":"No","date_created":"2020-06-29T23:55:23Z","department":[{"_id":"ChLa"}],"author":[{"last_name":"Anciukevicius","first_name":"Titas","full_name":"Anciukevicius, Titas"},{"first_name":"Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Henderson, Paul M","orcid":"0000-0002-5198-7445","last_name":"Henderson","first_name":"Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425"}],"publication":"arXiv","_id":"8063","license":"https://creativecommons.org/licenses/by-sa/4.0/"},{"language":[{"iso":"eng"}],"oa_version":"Preprint","month":"01","publication":"Domain Adaptation for Visual Understanding","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1711.05139"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","status":"public","related_material":{"record":[{"relation":"dissertation_contains","id":"8331","status":"deleted"},{"id":"8390","relation":"dissertation_contains","status":"public"}]},"publication_identifier":{"isbn":["9783030306717"]},"oa":1,"date_published":"2020-01-08T00:00:00Z","type":"book_chapter","publisher":"Springer Nature","editor":[{"full_name":"Singh, Richa","last_name":"Singh","first_name":"Richa"},{"full_name":"Vatsa, Mayank","first_name":"Mayank","last_name":"Vatsa"},{"last_name":"Patel","first_name":"Vishal M.","full_name":"Patel, Vishal M."},{"last_name":"Ratha","first_name":"Nalini","full_name":"Ratha, Nalini"}],"page":"33-49","quality_controlled":"1","publication_status":"published","article_processing_charge":"No","date_created":"2020-07-05T22:00:46Z","department":[{"_id":"ChLa"}],"title":"XGAN: Unsupervised image-to-image translation for many-to-many mappings","_id":"8092","scopus_import":"1","author":[{"id":"3811D890-F248-11E8-B48F-1D18A9856A87","full_name":"Royer, Amélie","orcid":"0000-0002-8407-0705","last_name":"Royer","first_name":"Amélie"},{"first_name":"Konstantinos","last_name":"Bousmalis","full_name":"Bousmalis, Konstantinos"},{"full_name":"Gouws, Stephan","last_name":"Gouws","first_name":"Stephan"},{"full_name":"Bertsch, Fred","last_name":"Bertsch","first_name":"Fred"},{"full_name":"Mosseri, Inbar","first_name":"Inbar","last_name":"Mosseri"},{"full_name":"Cole, Forrester","first_name":"Forrester","last_name":"Cole"},{"full_name":"Murphy, Kevin","last_name":"Murphy","first_name":"Kevin"}],"doi":"10.1007/978-3-030-30671-7_3","arxiv":1,"day":"08","abstract":[{"lang":"eng","text":"Image translation refers to the task of mapping images from a visual domain to another. Given two unpaired collections of images, we aim to learn a mapping between the corpus-level style of each collection, while preserving semantic content shared across the two domains. We introduce xgan, a dual adversarial auto-encoder, which captures a shared representation of the common domain semantic content in an unsupervised way, while jointly learning the domain-to-domain image translations in both directions. We exploit ideas from the domain adaptation literature and define a semantic consistency loss which encourages the learned embedding to preserve semantics shared across domains. We report promising qualitative results for the task of face-to-cartoon translation. The cartoon dataset we collected for this purpose, “CartoonSet”, is also publicly available as a new benchmark for semantic style transfer at https://google.github.io/cartoonset/index.html."}],"date_updated":"2023-09-07T13:16:18Z","year":"2020","citation":{"chicago":"Royer, Amélie, Konstantinos Bousmalis, Stephan Gouws, Fred Bertsch, Inbar Mosseri, Forrester Cole, and Kevin Murphy. “XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings.” In <i>Domain Adaptation for Visual Understanding</i>, edited by Richa Singh, Mayank Vatsa, Vishal M. Patel, and Nalini Ratha, 33–49. Springer Nature, 2020. <a href=\"https://doi.org/10.1007/978-3-030-30671-7_3\">https://doi.org/10.1007/978-3-030-30671-7_3</a>.","ieee":"A. Royer <i>et al.</i>, “XGAN: Unsupervised image-to-image translation for many-to-many mappings,” in <i>Domain Adaptation for Visual Understanding</i>, R. Singh, M. Vatsa, V. M. Patel, and N. Ratha, Eds. Springer Nature, 2020, pp. 33–49.","ama":"Royer A, Bousmalis K, Gouws S, et al. XGAN: Unsupervised image-to-image translation for many-to-many mappings. In: Singh R, Vatsa M, Patel VM, Ratha N, eds. <i>Domain Adaptation for Visual Understanding</i>. Springer Nature; 2020:33-49. doi:<a href=\"https://doi.org/10.1007/978-3-030-30671-7_3\">10.1007/978-3-030-30671-7_3</a>","apa":"Royer, A., Bousmalis, K., Gouws, S., Bertsch, F., Mosseri, I., Cole, F., &#38; Murphy, K. (2020). XGAN: Unsupervised image-to-image translation for many-to-many mappings. In R. Singh, M. Vatsa, V. M. Patel, &#38; N. Ratha (Eds.), <i>Domain Adaptation for Visual Understanding</i> (pp. 33–49). Springer Nature. <a href=\"https://doi.org/10.1007/978-3-030-30671-7_3\">https://doi.org/10.1007/978-3-030-30671-7_3</a>","ista":"Royer A, Bousmalis K, Gouws S, Bertsch F, Mosseri I, Cole F, Murphy K. 2020.XGAN: Unsupervised image-to-image translation for many-to-many mappings. In: Domain Adaptation for Visual Understanding. , 33–49.","mla":"Royer, Amélie, et al. “XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings.” <i>Domain Adaptation for Visual Understanding</i>, edited by Richa Singh et al., Springer Nature, 2020, pp. 33–49, doi:<a href=\"https://doi.org/10.1007/978-3-030-30671-7_3\">10.1007/978-3-030-30671-7_3</a>.","short":"A. Royer, K. Bousmalis, S. Gouws, F. Bertsch, I. Mosseri, F. Cole, K. Murphy, in:, R. Singh, M. Vatsa, V.M. Patel, N. Ratha (Eds.), Domain Adaptation for Visual Understanding, Springer Nature, 2020, pp. 33–49."},"external_id":{"arxiv":["1711.05139"]}},{"file":[{"file_name":"paper.pdf","content_type":"application/pdf","date_updated":"2020-07-31T16:57:12Z","file_size":10262773,"date_created":"2020-07-31T16:57:12Z","creator":"phenders","file_id":"8187","success":1,"relation":"main_file","access_level":"open_access"}],"main_file_link":[{"open_access":"1","url":"https://openaccess.thecvf.com/content_CVPR_2020/papers/Henderson_Leveraging_2D_Data_to_Learn_Textured_3D_Mesh_Generation_CVPR_2020_paper.pdf"}],"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication_identifier":{"eissn":["2575-7075"],"eisbn":["9781728171685"]},"oa":1,"date_published":"2020-07-01T00:00:00Z","type":"conference","conference":{"start_date":"2020-06-14","name":"CVPR: Conference on Computer Vision and Pattern Recognition","location":"Virtual","end_date":"2020-06-19"},"language":[{"iso":"eng"}],"oa_version":"Submitted Version","month":"07","publication":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition","has_accepted_license":"1","ddc":["004"],"doi":"10.1109/CVPR42600.2020.00752","arxiv":1,"day":"01","abstract":[{"lang":"eng","text":"Numerous methods have been proposed for probabilistic generative modelling of\r\n3D objects. However, none of these is able to produce textured objects, which\r\nrenders them of limited use for practical tasks. In this work, we present the\r\nfirst generative model of textured 3D meshes. Training such a model would\r\ntraditionally require a large dataset of textured meshes, but unfortunately,\r\nexisting datasets of meshes lack detailed textures. We instead propose a new\r\ntraining methodology that allows learning from collections of 2D images without\r\nany 3D information. To do so, we train our model to explain a distribution of\r\nimages by modelling each image as a 3D foreground object placed in front of a\r\n2D background. Thus, it learns to generate meshes that when rendered, produce\r\nimages similar to those in its training set.\r\n  A well-known problem when generating meshes with deep networks is the\r\nemergence of self-intersections, which are problematic for many use-cases. As a\r\nsecond contribution we therefore introduce a new generation process for 3D\r\nmeshes that guarantees no self-intersections arise, based on the physical\r\nintuition that faces should push one another out of the way as they move.\r\n  We conduct extensive experiments on our approach, reporting quantitative and\r\nqualitative results on both synthetic data and natural images. These show our\r\nmethod successfully learns to generate plausible and diverse textured 3D\r\nsamples for five challenging object classes."}],"date_updated":"2023-10-17T07:37:11Z","year":"2020","citation":{"ista":"Henderson PM, Tsiminaki V, Lampert C. 2020. Leveraging 2D data to learn textured 3D mesh generation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Conference on Computer Vision and Pattern Recognition, 7498–7507.","short":"P.M. Henderson, V. Tsiminaki, C. Lampert, in:, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2020, pp. 7498–7507.","mla":"Henderson, Paul M., et al. “Leveraging 2D Data to Learn Textured 3D Mesh Generation.” <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, IEEE, 2020, pp. 7498–507, doi:<a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">10.1109/CVPR42600.2020.00752</a>.","chicago":"Henderson, Paul M, Vagia Tsiminaki, and Christoph Lampert. “Leveraging 2D Data to Learn Textured 3D Mesh Generation.” In <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, 7498–7507. IEEE, 2020. <a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">https://doi.org/10.1109/CVPR42600.2020.00752</a>.","ieee":"P. M. Henderson, V. Tsiminaki, and C. Lampert, “Leveraging 2D data to learn textured 3D mesh generation,” in <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, Virtual, 2020, pp. 7498–7507.","ama":"Henderson PM, Tsiminaki V, Lampert C. Leveraging 2D data to learn textured 3D mesh generation. In: <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>. IEEE; 2020:7498-7507. doi:<a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">10.1109/CVPR42600.2020.00752</a>","apa":"Henderson, P. M., Tsiminaki, V., &#38; Lampert, C. (2020). Leveraging 2D data to learn textured 3D mesh generation. In <i>Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition</i> (pp. 7498–7507). Virtual: IEEE. <a href=\"https://doi.org/10.1109/CVPR42600.2020.00752\">https://doi.org/10.1109/CVPR42600.2020.00752</a>"},"external_id":{"arxiv":["2004.04180"]},"publisher":"IEEE","page":"7498-7507","quality_controlled":"1","file_date_updated":"2020-07-31T16:57:12Z","publication_status":"published","article_processing_charge":"No","date_created":"2020-07-31T16:53:49Z","department":[{"_id":"ChLa"}],"title":"Leveraging 2D data to learn textured 3D mesh generation","_id":"8186","scopus_import":"1","author":[{"full_name":"Henderson, Paul M","orcid":"0000-0002-5198-7445","last_name":"Henderson","first_name":"Paul M","id":"13C09E74-18D9-11E9-8878-32CFE5697425"},{"last_name":"Tsiminaki","first_name":"Vagia","full_name":"Tsiminaki, Vagia"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}]},{"language":[{"iso":"eng"}],"conference":{"name":"NeurIPS: Neural Information Processing Systems","start_date":"2020-12-06","end_date":"2020-12-12","location":"Vancouver, Canada"},"publication":"34th Conference on Neural Information Processing Systems","month":"07","oa_version":"Preprint","acknowledged_ssus":[{"_id":"ScienComp"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","status":"public","main_file_link":[{"url":"https://arxiv.org/abs/2007.06705","open_access":"1"}],"type":"conference","date_published":"2020-07-07T00:00:00Z","oa":1,"publication_identifier":{"isbn":["9781713829546"]},"quality_controlled":"1","page":"3106–3117","publisher":"Curran Associates","author":[{"last_name":"Henderson","first_name":"Paul M","full_name":"Henderson, Paul M","orcid":"0000-0002-5198-7445","id":"13C09E74-18D9-11E9-8878-32CFE5697425"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"_id":"8188","intvolume":"        33","title":"Unsupervised object-centric video generation and decomposition in 3D","article_processing_charge":"No","department":[{"_id":"ChLa"}],"date_created":"2020-07-31T16:59:19Z","publication_status":"published","volume":33,"acknowledgement":"This research was supported by the Scientific Service Units (SSU) of IST Austria through resources\r\nprovided by Scientific Computing (SciComp). PH is employed part-time by Blackford Analysis, but\r\nthey did not support this project in any way.","external_id":{"arxiv":["2007.06705"]},"citation":{"ieee":"P. M. Henderson and C. Lampert, “Unsupervised object-centric video generation and decomposition in 3D,” in <i>34th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2020, vol. 33, pp. 3106–3117.","chicago":"Henderson, Paul M, and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” In <i>34th Conference on Neural Information Processing Systems</i>, 33:3106–3117. Curran Associates, 2020.","ama":"Henderson PM, Lampert C. Unsupervised object-centric video generation and decomposition in 3D. In: <i>34th Conference on Neural Information Processing Systems</i>. Vol 33. Curran Associates; 2020:3106–3117.","apa":"Henderson, P. M., &#38; Lampert, C. (2020). Unsupervised object-centric video generation and decomposition in 3D. In <i>34th Conference on Neural Information Processing Systems</i> (Vol. 33, pp. 3106–3117). Vancouver, Canada: Curran Associates.","ista":"Henderson PM, Lampert C. 2020. Unsupervised object-centric video generation and decomposition in 3D. 34th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 33, 3106–3117.","short":"P.M. Henderson, C. Lampert, in:, 34th Conference on Neural Information Processing Systems, Curran Associates, 2020, pp. 3106–3117.","mla":"Henderson, Paul M., and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” <i>34th Conference on Neural Information Processing Systems</i>, vol. 33, Curran Associates, 2020, pp. 3106–3117."},"year":"2020","date_updated":"2023-04-25T09:49:58Z","abstract":[{"text":"A natural approach to generative modeling of videos is to represent them as a composition of moving objects. Recent works model a set of 2D sprites over a slowly-varying background, but without considering the underlying 3D scene that\r\ngives rise to them. We instead propose to model a video as the view seen while moving through a scene with multiple 3D objects and a 3D background. Our model is trained from monocular videos without any supervision, yet learns to\r\ngenerate coherent 3D scenes containing several moving objects. We conduct detailed experiments on two datasets, going beyond the visual complexity supported by state-of-the-art generative approaches. We evaluate our method on\r\ndepth-prediction and 3D object detection---tasks which cannot be addressed by those earlier works---and show it out-performs them even on 2D instance segmentation and tracking.","lang":"eng"}],"day":"07","arxiv":1},{"ddc":["000"],"acknowledgement":"Last but not least, I would like to acknowledge the support of the IST IT and scientific computing team for helping provide a great work environment.","citation":{"ista":"Royer A. 2020. Leveraging structure in Computer Vision tasks for flexible Deep Learning models. Institute of Science and Technology Austria.","mla":"Royer, Amélie. <i>Leveraging Structure in Computer Vision Tasks for Flexible Deep Learning Models</i>. Institute of Science and Technology Austria, 2020, doi:<a href=\"https://doi.org/10.15479/AT:ISTA:8390\">10.15479/AT:ISTA:8390</a>.","short":"A. Royer, Leveraging Structure in Computer Vision Tasks for Flexible Deep Learning Models, Institute of Science and Technology Austria, 2020.","ieee":"A. Royer, “Leveraging structure in Computer Vision tasks for flexible Deep Learning models,” Institute of Science and Technology Austria, 2020.","chicago":"Royer, Amélie. “Leveraging Structure in Computer Vision Tasks for Flexible Deep Learning Models.” Institute of Science and Technology Austria, 2020. <a href=\"https://doi.org/10.15479/AT:ISTA:8390\">https://doi.org/10.15479/AT:ISTA:8390</a>.","ama":"Royer A. Leveraging structure in Computer Vision tasks for flexible Deep Learning models. 2020. doi:<a href=\"https://doi.org/10.15479/AT:ISTA:8390\">10.15479/AT:ISTA:8390</a>","apa":"Royer, A. (2020). <i>Leveraging structure in Computer Vision tasks for flexible Deep Learning models</i>. Institute of Science and Technology Austria. <a href=\"https://doi.org/10.15479/AT:ISTA:8390\">https://doi.org/10.15479/AT:ISTA:8390</a>"},"year":"2020","date_updated":"2023-10-16T10:04:02Z","abstract":[{"text":"Deep neural networks have established a new standard for data-dependent feature extraction pipelines in the Computer Vision literature. Despite their remarkable performance in the standard supervised learning scenario, i.e. when models are trained with labeled data and tested on samples that follow a similar distribution, neural networks have been shown to struggle with more advanced generalization abilities, such as transferring knowledge across visually different domains, or generalizing to new unseen combinations of known concepts. In this thesis we argue that, in contrast to the usual black-box behavior of neural networks, leveraging more structured internal representations is a promising direction\r\nfor tackling such problems. In particular, we focus on two forms of structure. First, we tackle modularity: We show that (i) compositional architectures are a natural tool for modeling reasoning tasks, in that they efficiently capture their combinatorial nature, which is key for generalizing beyond the compositions seen during training. We investigate how to to learn such models, both formally and experimentally, for the task of abstract visual reasoning. Then, we show that (ii) in some settings, modularity allows us to efficiently break down complex tasks into smaller, easier, modules, thereby improving computational efficiency; We study this behavior in the context of generative models for colorization, as well as for small objects detection. Secondly, we investigate the inherently layered structure of representations learned by neural networks, and analyze its role in the context of transfer learning and domain adaptation across visually\r\ndissimilar domains. ","lang":"eng"}],"day":"14","degree_awarded":"PhD","doi":"10.15479/AT:ISTA:8390","file_date_updated":"2020-09-14T13:39:17Z","page":"197","publisher":"Institute of Science and Technology Austria","author":[{"first_name":"Amélie","last_name":"Royer","orcid":"0000-0002-8407-0705","full_name":"Royer, Amélie","id":"3811D890-F248-11E8-B48F-1D18A9856A87"}],"_id":"8390","title":"Leveraging structure in Computer Vision tasks for flexible Deep Learning models","alternative_title":["ISTA Thesis"],"article_processing_charge":"No","date_created":"2020-09-14T13:42:09Z","department":[{"_id":"ChLa"}],"publication_status":"published","related_material":{"record":[{"id":"7936","relation":"part_of_dissertation","status":"public"},{"status":"public","id":"7937","relation":"part_of_dissertation"},{"status":"public","relation":"part_of_dissertation","id":"8193"},{"relation":"part_of_dissertation","id":"8092","status":"public"},{"status":"public","relation":"part_of_dissertation","id":"911"}]},"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","status":"public","file":[{"creator":"dernst","file_id":"8391","success":1,"access_level":"open_access","relation":"main_file","file_name":"2020_Thesis_Royer.pdf","content_type":"application/pdf","date_updated":"2020-09-14T13:39:14Z","checksum":"c914d2f88846032f3d8507734861b6ee","file_size":30224591,"date_created":"2020-09-14T13:39:14Z"},{"file_name":"thesis_sources.zip","content_type":"application/x-zip-compressed","date_updated":"2020-09-14T13:39:17Z","checksum":"ae98fb35d912cff84a89035ae5794d3c","file_size":74227627,"date_created":"2020-09-14T13:39:17Z","creator":"dernst","file_id":"8392","relation":"main_file","access_level":"closed"}],"type":"dissertation","date_published":"2020-09-14T00:00:00Z","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode","name":"Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)","image":"/images/cc_by_nc_sa.png","short":"CC BY-NC-SA (4.0)"},"oa":1,"supervisor":[{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert"}],"publication_identifier":{"isbn":["978-3-99078-007-7"],"issn":["2663-337X"]},"language":[{"iso":"eng"}],"has_accepted_license":"1","month":"09","acknowledged_ssus":[{"_id":"CampIT"},{"_id":"ScienComp"}],"oa_version":"Published Version"},{"project":[{"name":"Elastic Coordination for Scalable Machine Learning","grant_number":"805223","_id":"268A44D6-B435-11E9-9278-68D0E5697425","call_identifier":"H2020"}],"acknowledged_ssus":[{"_id":"ScienComp"}],"oa_version":"Published Version","month":"07","has_accepted_license":"1","publication":"Proceedings of the 37th International Conference on Machine Learning","conference":{"end_date":"2020-07-18","location":"Online","start_date":"2020-07-12","name":"ICML: International Conference on Machine Learning"},"language":[{"iso":"eng"}],"publication_identifier":{"issn":["2640-3498"]},"oa":1,"type":"conference","date_published":"2020-07-12T00:00:00Z","file":[{"access_level":"open_access","success":1,"relation":"main_file","file_id":"9120","creator":"dernst","date_created":"2021-02-15T09:00:01Z","file_size":281286,"checksum":"cc755d0054bc4b2be778ea7aa7884d2f","date_updated":"2021-02-15T09:00:01Z","content_type":"application/pdf","file_name":"2020_PMLR_Konstantinov.pdf"}],"user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","status":"public","related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"10799"}],"link":[{"url":"http://proceedings.mlr.press/v119/konstantinov20a/konstantinov20a-supp.pdf","relation":"supplementary_material"}]},"department":[{"_id":"DaAl"},{"_id":"ChLa"}],"date_created":"2020-11-05T15:25:58Z","article_processing_charge":"No","publication_status":"published","intvolume":"       119","title":"On the sample complexity of adversarial multi-source PAC learning","scopus_import":"1","_id":"8724","author":[{"id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H","first_name":"Nikola H","last_name":"Konstantinov"},{"first_name":"Elias","last_name":"Frantar","full_name":"Frantar, Elias","id":"09a8f98d-ec99-11ea-ae11-c063a7b7fe5f"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","last_name":"Alistarh","first_name":"Dan-Adrian","full_name":"Alistarh, Dan-Adrian","orcid":"0000-0003-3650-940X"},{"orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"publisher":"ML Research Press","quality_controlled":"1","ec_funded":1,"page":"5416-5425","file_date_updated":"2021-02-15T09:00:01Z","day":"12","arxiv":1,"abstract":[{"lang":"eng","text":"We study the problem of learning from multiple untrusted data sources, a scenario of increasing practical relevance given the recent emergence of crowdsourcing and collaborative learning paradigms. Specifically, we analyze the situation in which a learning system obtains datasets from multiple sources, some of which might be biased or even adversarially perturbed. It is\r\nknown that in the single-source case, an adversary with the power to corrupt a fixed fraction of the training data can prevent PAC-learnability, that is, even in the limit of infinitely much training data, no learning system can approach the optimal test error. In this work we show that, surprisingly, the same is not true in the multi-source setting, where the adversary can arbitrarily\r\ncorrupt a fixed fraction of the data sources. Our main results are a generalization bound that provides finite-sample guarantees for this learning setting, as well as corresponding lower bounds. Besides establishing PAC-learnability our results also show that in a cooperative learning setting sharing data with other parties has provable benefits, even if some\r\nparticipants are malicious. "}],"citation":{"short":"N.H. Konstantinov, E. Frantar, D.-A. Alistarh, C. Lampert, in:, Proceedings of the 37th International Conference on Machine Learning, ML Research Press, 2020, pp. 5416–5425.","mla":"Konstantinov, Nikola H., et al. “On the Sample Complexity of Adversarial Multi-Source PAC Learning.” <i>Proceedings of the 37th International Conference on Machine Learning</i>, vol. 119, ML Research Press, 2020, pp. 5416–25.","ista":"Konstantinov NH, Frantar E, Alistarh D-A, Lampert C. 2020. On the sample complexity of adversarial multi-source PAC learning. Proceedings of the 37th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 119, 5416–5425.","apa":"Konstantinov, N. H., Frantar, E., Alistarh, D.-A., &#38; Lampert, C. (2020). On the sample complexity of adversarial multi-source PAC learning. In <i>Proceedings of the 37th International Conference on Machine Learning</i> (Vol. 119, pp. 5416–5425). Online: ML Research Press.","ama":"Konstantinov NH, Frantar E, Alistarh D-A, Lampert C. On the sample complexity of adversarial multi-source PAC learning. In: <i>Proceedings of the 37th International Conference on Machine Learning</i>. Vol 119. ML Research Press; 2020:5416-5425.","chicago":"Konstantinov, Nikola H, Elias Frantar, Dan-Adrian Alistarh, and Christoph Lampert. “On the Sample Complexity of Adversarial Multi-Source PAC Learning.” In <i>Proceedings of the 37th International Conference on Machine Learning</i>, 119:5416–25. ML Research Press, 2020.","ieee":"N. H. Konstantinov, E. Frantar, D.-A. Alistarh, and C. Lampert, “On the sample complexity of adversarial multi-source PAC learning,” in <i>Proceedings of the 37th International Conference on Machine Learning</i>, Online, 2020, vol. 119, pp. 5416–5425."},"year":"2020","date_updated":"2023-09-07T13:42:08Z","external_id":{"arxiv":["2002.10384"]},"acknowledgement":"Dan Alistarh is supported in part by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML). This research was supported by the Scientific Service Units (SSU) of IST Austria through resources provided by Scientific Computing (SciComp).","volume":119,"ddc":["000"]},{"volume":128,"ddc":["004"],"year":"2020","citation":{"apa":"Sun, R., &#38; Lampert, C. (2020). KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. <i>International Journal of Computer Vision</i>. Springer Nature. <a href=\"https://doi.org/10.1007/s11263-019-01232-x\">https://doi.org/10.1007/s11263-019-01232-x</a>","ama":"Sun R, Lampert C. KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. <i>International Journal of Computer Vision</i>. 2020;128(4):970-995. doi:<a href=\"https://doi.org/10.1007/s11263-019-01232-x\">10.1007/s11263-019-01232-x</a>","chicago":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a Multiclass Classifier Operates Outside of Its Specifications.” <i>International Journal of Computer Vision</i>. Springer Nature, 2020. <a href=\"https://doi.org/10.1007/s11263-019-01232-x\">https://doi.org/10.1007/s11263-019-01232-x</a>.","ieee":"R. Sun and C. Lampert, “KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications,” <i>International Journal of Computer Vision</i>, vol. 128, no. 4. Springer Nature, pp. 970–995, 2020.","short":"R. Sun, C. Lampert, International Journal of Computer Vision 128 (2020) 970–995.","mla":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a Multiclass Classifier Operates Outside of Its Specifications.” <i>International Journal of Computer Vision</i>, vol. 128, no. 4, Springer Nature, 2020, pp. 970–95, doi:<a href=\"https://doi.org/10.1007/s11263-019-01232-x\">10.1007/s11263-019-01232-x</a>.","ista":"Sun R, Lampert C. 2020. KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. International Journal of Computer Vision. 128(4), 970–995."},"date_updated":"2024-02-22T14:57:30Z","external_id":{"isi":["000494406800001"]},"isi":1,"day":"01","doi":"10.1007/s11263-019-01232-x","abstract":[{"text":"We study the problem of automatically detecting if a given multi-class classifier operates outside of its specifications (out-of-specs), i.e. on input data from a different distribution than what it was trained for. This is an important problem to solve on the road towards creating reliable computer vision systems for real-world applications, because the quality of a classifier’s predictions cannot be guaranteed if it operates out-of-specs. Previously proposed methods for out-of-specs detection make decisions on the level of single inputs. This, however, is insufficient to achieve low false positive rate and high false negative rates at the same time. In this work, we describe a new procedure named KS(conf), based on statistical reasoning. Its main component is a classical Kolmogorov–Smirnov test that is applied to the set of predicted confidence values for batches of samples. Working with batches instead of single samples allows increasing the true positive rate without negatively affecting the false positive rate, thereby overcoming a crucial limitation of single sample tests. We show by extensive experiments using a variety of convolutional network architectures and datasets that KS(conf) reliably detects out-of-specs situations even under conditions where other tests fail. It furthermore has a number of properties that make it an excellent candidate for practical deployment: it is easy to implement, adds almost no overhead to the system, works with any classifier that outputs confidence scores, and requires no a priori knowledge about how the data distribution could change.","lang":"eng"}],"ec_funded":1,"quality_controlled":"1","page":"970-995","file_date_updated":"2020-07-14T12:47:45Z","publisher":"Springer Nature","article_type":"original","scopus_import":"1","_id":"6944","issue":"4","author":[{"full_name":"Sun, Rémy","first_name":"Rémy","last_name":"Sun"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","last_name":"Lampert","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2019-10-14T09:14:28Z","article_processing_charge":"Yes (via OA deal)","department":[{"_id":"ChLa"}],"publication_status":"published","intvolume":"       128","title":"KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications","file":[{"file_id":"7110","creator":"dernst","relation":"main_file","access_level":"open_access","date_updated":"2020-07-14T12:47:45Z","content_type":"application/pdf","file_name":"2019_IJCV_Sun.pdf","date_created":"2019-11-26T10:30:02Z","checksum":"155e63edf664dcacb3bdc1c2223e606f","file_size":1715072}],"user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","related_material":{"record":[{"id":"6482","relation":"earlier_version","status":"public"}],"link":[{"relation":"erratum","url":"https://doi.org/10.1007/s11263-019-01262-5"}]},"status":"public","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"type":"journal_article","date_published":"2020-04-01T00:00:00Z","publication_identifier":{"eissn":["1573-1405"],"issn":["0920-5691"]},"oa":1,"language":[{"iso":"eng"}],"has_accepted_license":"1","publication":"International Journal of Computer Vision","project":[{"name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036","call_identifier":"FP7","_id":"2532554C-B435-11E9-9278-68D0E5697425"},{"name":"IST Austria Open Access Fund","_id":"B67AFEDC-15C9-11EA-A837-991A96BB2854"}],"oa_version":"Published Version","month":"04"},{"type":"journal_article","date_published":"2020-04-01T00:00:00Z","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"oa":1,"publication_identifier":{"issn":["0920-5691"],"eissn":["1573-1405"]},"status":"public","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","file":[{"file_id":"6973","creator":"dernst","relation":"main_file","access_level":"open_access","date_updated":"2020-07-14T12:47:46Z","content_type":"application/pdf","file_name":"2019_CompVision_Henderson.pdf","date_created":"2019-10-25T10:28:29Z","file_size":2243134,"checksum":"a0f05dd4f5f64e4f713d8d9d4b5b1e3f"}],"has_accepted_license":"1","publication":"International Journal of Computer Vision","month":"04","project":[{"_id":"B67AFEDC-15C9-11EA-A837-991A96BB2854","name":"IST Austria Open Access Fund"}],"oa_version":"Published Version","language":[{"iso":"eng"}],"external_id":{"isi":["000491042100002"],"arxiv":["1901.06447"]},"isi":1,"citation":{"chicago":"Henderson, Paul M, and Vittorio Ferrari. “Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading.” <i>International Journal of Computer Vision</i>. Springer Nature, 2020. <a href=\"https://doi.org/10.1007/s11263-019-01219-8\">https://doi.org/10.1007/s11263-019-01219-8</a>.","ieee":"P. M. Henderson and V. Ferrari, “Learning single-image 3D reconstruction by generative modelling of shape, pose and shading,” <i>International Journal of Computer Vision</i>, vol. 128. Springer Nature, pp. 835–854, 2020.","ama":"Henderson PM, Ferrari V. Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. <i>International Journal of Computer Vision</i>. 2020;128:835-854. doi:<a href=\"https://doi.org/10.1007/s11263-019-01219-8\">10.1007/s11263-019-01219-8</a>","apa":"Henderson, P. M., &#38; Ferrari, V. (2020). Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. <i>International Journal of Computer Vision</i>. Springer Nature. <a href=\"https://doi.org/10.1007/s11263-019-01219-8\">https://doi.org/10.1007/s11263-019-01219-8</a>","ista":"Henderson PM, Ferrari V. 2020. Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. International Journal of Computer Vision. 128, 835–854.","short":"P.M. Henderson, V. Ferrari, International Journal of Computer Vision 128 (2020) 835–854.","mla":"Henderson, Paul M., and Vittorio Ferrari. “Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading.” <i>International Journal of Computer Vision</i>, vol. 128, Springer Nature, 2020, pp. 835–54, doi:<a href=\"https://doi.org/10.1007/s11263-019-01219-8\">10.1007/s11263-019-01219-8</a>."},"year":"2020","date_updated":"2023-08-17T14:01:16Z","abstract":[{"text":"We present a unified framework tackling two problems: class-specific 3D reconstruction from a single image, and generation of new 3D shape samples. These tasks have received considerable attention recently; however, most existing approaches rely on 3D supervision, annotation of 2D images with keypoints or poses, and/or training with multiple views of each object instance. Our framework is very general: it can be trained in similar settings to existing approaches, while also supporting weaker supervision. Importantly, it can be trained purely from 2D images, without pose annotations, and with only a single view per instance. We employ meshes as an output representation, instead of voxels used in most prior work. This allows us to reason over lighting parameters and exploit shading information during training, which previous 2D-supervised methods cannot. Thus, our method can learn to generate and reconstruct concave object classes. We evaluate our approach in various settings, showing that: (i) it learns to disentangle shape from pose and lighting; (ii) using shading in the loss improves performance compared to just silhouettes; (iii) when using a standard single white light, our model outperforms state-of-the-art 2D-supervised methods, both with and without pose supervision, thanks to exploiting shading cues; (iv) performance improves further when using multiple coloured lights, even approaching that of state-of-the-art 3D-supervised methods; (v) shapes produced by our model capture smooth surfaces and fine details better than voxel-based approaches; and (vi) our approach supports concave classes such as bathtubs and sofas, which methods based on silhouettes cannot learn.","lang":"eng"}],"day":"01","doi":"10.1007/s11263-019-01219-8","arxiv":1,"ddc":["004"],"volume":128,"acknowledgement":"Open access funding provided by Institute of Science and Technology (IST Austria).","author":[{"id":"13C09E74-18D9-11E9-8878-32CFE5697425","last_name":"Henderson","first_name":"Paul M","full_name":"Henderson, Paul M","orcid":"0000-0002-5198-7445"},{"first_name":"Vittorio","last_name":"Ferrari","full_name":"Ferrari, Vittorio"}],"scopus_import":"1","_id":"6952","intvolume":"       128","title":"Learning single-image 3D reconstruction by generative modelling of shape, pose and shading","department":[{"_id":"ChLa"}],"date_created":"2019-10-17T13:38:20Z","article_processing_charge":"Yes (via OA deal)","publication_status":"published","file_date_updated":"2020-07-14T12:47:46Z","quality_controlled":"1","page":"835-854","article_type":"original","publisher":"Springer Nature"},{"date_published":"2020-04-26T00:00:00Z","type":"conference","date_updated":"2023-09-07T13:29:50Z","citation":{"mla":"Phuong, Mary, and Christoph Lampert. “Functional vs. Parametric Equivalence of ReLU Networks.” <i>8th International Conference on Learning Representations</i>, 2020.","short":"M. Phuong, C. Lampert, in:, 8th International Conference on Learning Representations, 2020.","ista":"Phuong M, Lampert C. 2020. Functional vs. parametric equivalence of ReLU networks. 8th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","apa":"Phuong, M., &#38; Lampert, C. (2020). Functional vs. parametric equivalence of ReLU networks. In <i>8th International Conference on Learning Representations</i>. Online.","ama":"Phuong M, Lampert C. Functional vs. parametric equivalence of ReLU networks. In: <i>8th International Conference on Learning Representations</i>. ; 2020.","chicago":"Phuong, Mary, and Christoph Lampert. “Functional vs. Parametric Equivalence of ReLU Networks.” In <i>8th International Conference on Learning Representations</i>, 2020.","ieee":"M. Phuong and C. Lampert, “Functional vs. parametric equivalence of ReLU networks,” in <i>8th International Conference on Learning Representations</i>, Online, 2020."},"year":"2020","abstract":[{"text":"We address the following question:  How redundant is the parameterisation of ReLU networks? Specifically, we consider transformations of the weight space which leave the function implemented by the network intact.  Two such transformations are known for feed-forward architectures:  permutation of neurons within a layer, and positive scaling of all incoming weights of a neuron coupled with inverse scaling of its outgoing weights. In this work, we show for architectures with non-increasing widths that permutation and scaling are in fact the only function-preserving weight transformations.  For any eligible architecture we give an explicit construction of a neural network such that any other network that implements the same function can be obtained from the original one by the application of permutations and rescaling.  The proof relies on a geometric understanding of boundaries between linear regions of ReLU networks, and we hope the developed mathematical tools are of independent interest.","lang":"eng"}],"oa":1,"day":"26","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","related_material":{"record":[{"status":"public","id":"9418","relation":"dissertation_contains"}],"link":[{"relation":"supplementary_material","url":"https://iclr.cc/virtual_2020/poster_Bylx-TNKvH.html"}]},"ddc":["000"],"file":[{"creator":"bphuong","file_id":"7482","access_level":"open_access","relation":"main_file","content_type":"application/pdf","file_name":"main.pdf","date_updated":"2020-07-14T12:47:59Z","file_size":405469,"checksum":"8d372ea5defd8cb8fdc430111ed754a9","date_created":"2020-02-11T09:07:27Z"}],"author":[{"id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","last_name":"Bui Thi Mai","first_name":"Phuong","full_name":"Bui Thi Mai, Phuong"},{"first_name":"Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"publication":"8th International Conference on Learning Representations","_id":"7481","has_accepted_license":"1","title":"Functional vs. parametric equivalence of ReLU networks","month":"04","publication_status":"published","oa_version":"Published Version","article_processing_charge":"No","date_created":"2020-02-11T09:07:37Z","department":[{"_id":"ChLa"}],"file_date_updated":"2020-07-14T12:47:59Z","language":[{"iso":"eng"}],"quality_controlled":"1","conference":{"start_date":"2020-04-27","name":"ICLR: International Conference on Learning Representations","end_date":"2020-04-30","location":"Online"}},{"isi":1,"external_id":{"isi":["000679281300007"],"arxiv":["1906.08178"]},"date_updated":"2025-06-02T08:53:47Z","year":"2019","citation":{"mla":"Ashok, Pranav, et al. “Strategy Representation by Decision Trees with Linear Classifiers.” <i>16th International Conference on Quantitative Evaluation of Systems</i>, vol. 11785, Springer Nature, 2019, pp. 109–28, doi:<a href=\"https://doi.org/10.1007/978-3-030-30281-8_7\">10.1007/978-3-030-30281-8_7</a>.","short":"P. Ashok, T. Brázdil, K. Chatterjee, J. Křetínský, C. Lampert, V. Toman, in:, 16th International Conference on Quantitative Evaluation of Systems, Springer Nature, 2019, pp. 109–128.","ista":"Ashok P, Brázdil T, Chatterjee K, Křetínský J, Lampert C, Toman V. 2019. Strategy representation by decision trees with linear classifiers. 16th International Conference on Quantitative Evaluation of Systems. QEST: Quantitative Evaluation of Systems, LNCS, vol. 11785, 109–128.","apa":"Ashok, P., Brázdil, T., Chatterjee, K., Křetínský, J., Lampert, C., &#38; Toman, V. (2019). Strategy representation by decision trees with linear classifiers. In <i>16th International Conference on Quantitative Evaluation of Systems</i> (Vol. 11785, pp. 109–128). Glasgow, United Kingdom: Springer Nature. <a href=\"https://doi.org/10.1007/978-3-030-30281-8_7\">https://doi.org/10.1007/978-3-030-30281-8_7</a>","ama":"Ashok P, Brázdil T, Chatterjee K, Křetínský J, Lampert C, Toman V. Strategy representation by decision trees with linear classifiers. In: <i>16th International Conference on Quantitative Evaluation of Systems</i>. Vol 11785. Springer Nature; 2019:109-128. doi:<a href=\"https://doi.org/10.1007/978-3-030-30281-8_7\">10.1007/978-3-030-30281-8_7</a>","ieee":"P. Ashok, T. Brázdil, K. Chatterjee, J. Křetínský, C. Lampert, and V. Toman, “Strategy representation by decision trees with linear classifiers,” in <i>16th International Conference on Quantitative Evaluation of Systems</i>, Glasgow, United Kingdom, 2019, vol. 11785, pp. 109–128.","chicago":"Ashok, Pranav, Tomáš Brázdil, Krishnendu Chatterjee, Jan Křetínský, Christoph Lampert, and Viktor Toman. “Strategy Representation by Decision Trees with Linear Classifiers.” In <i>16th International Conference on Quantitative Evaluation of Systems</i>, 11785:109–28. Springer Nature, 2019. <a href=\"https://doi.org/10.1007/978-3-030-30281-8_7\">https://doi.org/10.1007/978-3-030-30281-8_7</a>."},"abstract":[{"text":"Graph games and Markov decision processes (MDPs) are standard models in reactive synthesis and verification of probabilistic systems with nondeterminism. The class of   𝜔 -regular winning conditions; e.g., safety, reachability, liveness, parity conditions; provides a robust and expressive specification formalism for properties that arise in analysis of reactive systems. The resolutions of nondeterminism in games and MDPs are represented as strategies, and we consider succinct representation of such strategies. The decision-tree data structure from machine learning retains the flavor of decisions of strategies and allows entropy-based minimization to obtain succinct trees. However, in contrast to traditional machine-learning problems where small errors are allowed, for winning strategies in graph games and MDPs no error is allowed, and the decision tree must represent the entire strategy. In this work we propose decision trees with linear classifiers for representation of strategies in graph games and MDPs. We have implemented strategy representation using this data structure and we present experimental results for problems on graph games and MDPs, which show that this new data structure presents a much more efficient strategy representation as compared to standard decision trees.","lang":"eng"}],"arxiv":1,"doi":"10.1007/978-3-030-30281-8_7","day":"04","volume":11785,"author":[{"first_name":"Pranav","last_name":"Ashok","full_name":"Ashok, Pranav"},{"last_name":"Brázdil","first_name":"Tomáš","full_name":"Brázdil, Tomáš"},{"orcid":"0000-0002-4561-241X","full_name":"Chatterjee, Krishnendu","first_name":"Krishnendu","last_name":"Chatterjee","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Křetínský, Jan","first_name":"Jan","last_name":"Křetínský"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"},{"full_name":"Toman, Viktor","orcid":"0000-0001-9036-063X","last_name":"Toman","first_name":"Viktor","id":"3AF3DA7C-F248-11E8-B48F-1D18A9856A87"}],"_id":"6942","scopus_import":"1","title":"Strategy representation by decision trees with linear classifiers","alternative_title":["LNCS"],"intvolume":"     11785","publication_status":"published","date_created":"2019-10-14T06:57:49Z","article_processing_charge":"No","department":[{"_id":"KrCh"},{"_id":"ChLa"}],"page":"109-128","quality_controlled":"1","publisher":"Springer Nature","date_published":"2019-09-04T00:00:00Z","type":"conference","oa":1,"publication_identifier":{"issn":["0302-9743"],"eisbn":["9783030302818"],"isbn":["9783030302801"]},"status":"public","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","main_file_link":[{"url":"https://arxiv.org/abs/1906.08178","open_access":"1"}],"publication":"16th International Conference on Quantitative Evaluation of Systems","month":"09","oa_version":"Preprint","project":[{"name":"Game Theory","grant_number":"S11407","call_identifier":"FWF","_id":"25863FF4-B435-11E9-9278-68D0E5697425"},{"grant_number":"S11402-N23","name":"Rigorous Systems Engineering","call_identifier":"FWF","_id":"25F2ACDE-B435-11E9-9278-68D0E5697425"},{"name":"Efficient Algorithms for Computer Aided Verification","grant_number":"ICT15-003","_id":"25892FC0-B435-11E9-9278-68D0E5697425"}],"language":[{"iso":"eng"}],"conference":{"location":"Glasgow, United Kingdom","end_date":"2019-09-12","start_date":"2019-09-10","name":"QEST: Quantitative Evaluation of Systems"}},{"place":"Wiesbaden","related_material":{"link":[{"description":"News on IST Website","relation":"press_release","url":"https://ist.ac.at/en/news/book-release-how-machines-learn/"}]},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","status":"public","publication_identifier":{"isbn":["978-3-658-26762-9"],"eisbn":["978-3-658-26763-6"]},"day":"30","edition":"1","doi":"10.1007/978-3-658-26763-6","abstract":[{"text":"Wissen Sie, was sich hinter künstlicher Intelligenz und maschinellem Lernen verbirgt? \r\nDieses Sachbuch erklärt Ihnen leicht verständlich und ohne komplizierte Formeln die grundlegenden Methoden und Vorgehensweisen des maschinellen Lernens. Mathematisches Vorwissen ist dafür nicht nötig. Kurzweilig und informativ illustriert Lisa, die Protagonistin des Buches, diese anhand von Alltagssituationen. \r\nEin Buch für alle, die in Diskussionen über Chancen und Risiken der aktuellen Entwicklung der künstlichen Intelligenz und des maschinellen Lernens mit Faktenwissen punkten möchten. Auch für Schülerinnen und Schüler geeignet!","lang":"ger"}],"citation":{"ieee":"K. Kersting, C. Lampert, and C. Rothkopf, Eds., <i>Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt</i>, 1st ed. Wiesbaden: Springer Nature, 2019.","chicago":"Kersting, Kristian, Christoph Lampert, and Constantin Rothkopf, eds. <i>Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt</i>. 1st ed. Wiesbaden: Springer Nature, 2019. <a href=\"https://doi.org/10.1007/978-3-658-26763-6\">https://doi.org/10.1007/978-3-658-26763-6</a>.","apa":"Kersting, K., Lampert, C., &#38; Rothkopf, C. (Eds.). (2019). <i>Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt</i> (1st ed.). Wiesbaden: Springer Nature. <a href=\"https://doi.org/10.1007/978-3-658-26763-6\">https://doi.org/10.1007/978-3-658-26763-6</a>","ama":"Kersting K, Lampert C, Rothkopf C, eds. <i>Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt</i>. 1st ed. Wiesbaden: Springer Nature; 2019. doi:<a href=\"https://doi.org/10.1007/978-3-658-26763-6\">10.1007/978-3-658-26763-6</a>","ista":"Kersting K, Lampert C, Rothkopf C eds. 2019. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt 1st ed., Wiesbaden: Springer Nature, XIV, 245p.","short":"K. Kersting, C. Lampert, C. Rothkopf, eds., Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt, 1st ed., Springer Nature, Wiesbaden, 2019.","mla":"Kersting, Kristian, et al., editors. <i>Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt</i>. 1st ed., Springer Nature, 2019, doi:<a href=\"https://doi.org/10.1007/978-3-658-26763-6\">10.1007/978-3-658-26763-6</a>."},"year":"2019","date_updated":"2021-12-22T14:40:58Z","type":"book_editor","date_published":"2019-10-30T00:00:00Z","editor":[{"full_name":"Kersting, Kristian","first_name":"Kristian","last_name":"Kersting"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert"},{"full_name":"Rothkopf, Constantin","last_name":"Rothkopf","first_name":"Constantin"}],"publisher":"Springer Nature","quality_controlled":"1","page":"XIV, 245","language":[{"iso":"ger"}],"article_processing_charge":"No","date_created":"2019-12-11T14:15:56Z","department":[{"_id":"ChLa"}],"oa_version":"None","publication_status":"published","title":"Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt","month":"10","_id":"7171"},{"has_accepted_license":"1","publication":"IEEE International Conference on Computer Vision","project":[{"grant_number":"308036","name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"oa_version":"Submitted Version","month":"10","language":[{"iso":"eng"}],"conference":{"location":"Seoul, Korea","end_date":"2019-11-02","name":"ICCV: International Conference on Computer Vision","start_date":"2019-10-27"},"type":"conference","date_published":"2019-10-01T00:00:00Z","publication_identifier":{"issn":["15505499"],"isbn":["9781728148038"]},"oa":1,"file":[{"date_updated":"2020-07-14T12:47:59Z","content_type":"application/pdf","file_name":"main.pdf","date_created":"2020-02-11T09:06:39Z","checksum":"7b77fb5c2d27c4c37a7612ba46a66117","file_size":735768,"file_id":"7480","creator":"bphuong","relation":"main_file","access_level":"open_access"}],"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","related_material":{"record":[{"status":"public","id":"9418","relation":"dissertation_contains"}]},"status":"public","scopus_import":"1","_id":"7479","author":[{"full_name":"Bui Thi Mai, Phuong","last_name":"Bui Thi Mai","first_name":"Phuong","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887"}],"article_processing_charge":"No","department":[{"_id":"ChLa"}],"date_created":"2020-02-11T09:06:57Z","publication_status":"published","title":"Distillation-based training for multi-exit architectures","quality_controlled":"1","ec_funded":1,"page":"1355-1364","file_date_updated":"2020-07-14T12:47:59Z","publisher":"IEEE","year":"2019","citation":{"ista":"Phuong M, Lampert C. 2019. Distillation-based training for multi-exit architectures. IEEE International Conference on Computer Vision. ICCV: International Conference on Computer Vision vol. 2019–October, 1355–1364.","short":"M. Phuong, C. Lampert, in:, IEEE International Conference on Computer Vision, IEEE, 2019, pp. 1355–1364.","mla":"Phuong, Mary, and Christoph Lampert. “Distillation-Based Training for Multi-Exit Architectures.” <i>IEEE International Conference on Computer Vision</i>, vol. 2019–October, IEEE, 2019, pp. 1355–64, doi:<a href=\"https://doi.org/10.1109/ICCV.2019.00144\">10.1109/ICCV.2019.00144</a>.","chicago":"Phuong, Mary, and Christoph Lampert. “Distillation-Based Training for Multi-Exit Architectures.” In <i>IEEE International Conference on Computer Vision</i>, 2019–October:1355–64. IEEE, 2019. <a href=\"https://doi.org/10.1109/ICCV.2019.00144\">https://doi.org/10.1109/ICCV.2019.00144</a>.","ieee":"M. Phuong and C. Lampert, “Distillation-based training for multi-exit architectures,” in <i>IEEE International Conference on Computer Vision</i>, Seoul, Korea, 2019, vol. 2019–October, pp. 1355–1364.","ama":"Phuong M, Lampert C. Distillation-based training for multi-exit architectures. In: <i>IEEE International Conference on Computer Vision</i>. Vol 2019-October. IEEE; 2019:1355-1364. doi:<a href=\"https://doi.org/10.1109/ICCV.2019.00144\">10.1109/ICCV.2019.00144</a>","apa":"Phuong, M., &#38; Lampert, C. (2019). Distillation-based training for multi-exit architectures. In <i>IEEE International Conference on Computer Vision</i> (Vol. 2019–October, pp. 1355–1364). Seoul, Korea: IEEE. <a href=\"https://doi.org/10.1109/ICCV.2019.00144\">https://doi.org/10.1109/ICCV.2019.00144</a>"},"date_updated":"2023-09-08T11:11:12Z","external_id":{"isi":["000531438101047"]},"isi":1,"day":"01","doi":"10.1109/ICCV.2019.00144","abstract":[{"lang":"eng","text":"Multi-exit architectures, in which a stack of processing layers is interleaved with early output layers, allow the processing of a test example to stop early and thus save computation time and/or energy.  In this work, we propose a new training procedure for multi-exit architectures based on the principle of knowledge distillation. The method encourage searly exits to mimic later, more accurate exits, by matching their output probabilities.\r\nExperiments  on  CIFAR100  and  ImageNet  show  that distillation-based training significantly improves the accuracy of early exits while maintaining state-of-the-art accuracy  for  late  ones.   The  method  is  particularly  beneficial when  training  data  is  limited  and  it  allows  a  straightforward extension to semi-supervised learning,i.e. making use of unlabeled data at training time. Moreover, it takes only afew lines to implement and incurs almost no computational overhead at training time, and none at all at test time."}],"volume":"2019-October","ddc":["000"]},{"month":"10","article_number":"1749-1753","oa_version":"Preprint","project":[{"_id":"2532554C-B435-11E9-9278-68D0E5697425","call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036"}],"publication":"Proceedings of the 2019 International Conference on Computer Vision Workshop","conference":{"location":"Seoul, South Korea","end_date":"2019-10-28","start_date":"2019-10-27","name":"ICCVW: International Conference on Computer Vision Workshop"},"language":[{"iso":"eng"}],"oa":1,"publication_identifier":{"isbn":["9781728150239"]},"date_published":"2019-10-01T00:00:00Z","type":"conference","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","status":"public","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1807.02136"}],"title":"Detecting visual relationships using box attention","publication_status":"published","article_processing_charge":"No","date_created":"2020-04-05T22:00:51Z","department":[{"_id":"ChLa"}],"author":[{"id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","last_name":"Kolesnikov","first_name":"Alexander","full_name":"Kolesnikov, Alexander"},{"last_name":"Kuznetsova","first_name":"Alina","full_name":"Kuznetsova, Alina"},{"orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Ferrari, Vittorio","last_name":"Ferrari","first_name":"Vittorio"}],"_id":"7640","scopus_import":"1","publisher":"IEEE","ec_funded":1,"quality_controlled":"1","abstract":[{"text":"We propose a new model for detecting visual relationships, such as \"person riding motorcycle\" or \"bottle on table\". This task is an important step towards comprehensive structured mage understanding, going beyond detecting individual objects. Our main novelty is a Box Attention mechanism that allows to model pairwise interactions between objects using standard object detection pipelines. The resulting model is conceptually clean, expressive and relies on well-justified training and prediction procedures. Moreover, unlike previously proposed approaches, our model does not introduce any additional complex components or hyperparameters on top of those already required by the underlying detection model. We conduct an experimental evaluation on two datasets, V-COCO and Open Images, demonstrating strong quantitative and qualitative results.","lang":"eng"}],"doi":"10.1109/ICCVW.2019.00217","arxiv":1,"day":"01","isi":1,"external_id":{"arxiv":["1807.02136"],"isi":["000554591601098"]},"date_updated":"2023-09-08T11:18:37Z","year":"2019","citation":{"short":"A. Kolesnikov, A. Kuznetsova, C. Lampert, V. Ferrari, in:, Proceedings of the 2019 International Conference on Computer Vision Workshop, IEEE, 2019.","mla":"Kolesnikov, Alexander, et al. “Detecting Visual Relationships Using Box Attention.” <i>Proceedings of the 2019 International Conference on Computer Vision Workshop</i>, 1749–1753, IEEE, 2019, doi:<a href=\"https://doi.org/10.1109/ICCVW.2019.00217\">10.1109/ICCVW.2019.00217</a>.","ista":"Kolesnikov A, Kuznetsova A, Lampert C, Ferrari V. 2019. Detecting visual relationships using box attention. Proceedings of the 2019 International Conference on Computer Vision Workshop. ICCVW: International Conference on Computer Vision Workshop, 1749–1753.","apa":"Kolesnikov, A., Kuznetsova, A., Lampert, C., &#38; Ferrari, V. (2019). Detecting visual relationships using box attention. In <i>Proceedings of the 2019 International Conference on Computer Vision Workshop</i>. Seoul, South Korea: IEEE. <a href=\"https://doi.org/10.1109/ICCVW.2019.00217\">https://doi.org/10.1109/ICCVW.2019.00217</a>","ama":"Kolesnikov A, Kuznetsova A, Lampert C, Ferrari V. Detecting visual relationships using box attention. In: <i>Proceedings of the 2019 International Conference on Computer Vision Workshop</i>. IEEE; 2019. doi:<a href=\"https://doi.org/10.1109/ICCVW.2019.00217\">10.1109/ICCVW.2019.00217</a>","chicago":"Kolesnikov, Alexander, Alina Kuznetsova, Christoph Lampert, and Vittorio Ferrari. “Detecting Visual Relationships Using Box Attention.” In <i>Proceedings of the 2019 International Conference on Computer Vision Workshop</i>. IEEE, 2019. <a href=\"https://doi.org/10.1109/ICCVW.2019.00217\">https://doi.org/10.1109/ICCVW.2019.00217</a>.","ieee":"A. Kolesnikov, A. Kuznetsova, C. Lampert, and V. Ferrari, “Detecting visual relationships using box attention,” in <i>Proceedings of the 2019 International Conference on Computer Vision Workshop</i>, Seoul, South Korea, 2019."}},{"volume":11269,"date_updated":"2024-02-22T14:57:29Z","citation":{"ieee":"R. Sun and C. Lampert, “KS(conf): A light-weight test if a ConvNet operates outside of Its specifications,” presented at the GCPR: Conference on Pattern Recognition, Stuttgart, Germany, 2019, vol. 11269, pp. 244–259.","chicago":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a ConvNet Operates Outside of Its Specifications,” 11269:244–59. Springer Nature, 2019. <a href=\"https://doi.org/10.1007/978-3-030-12939-2_18\">https://doi.org/10.1007/978-3-030-12939-2_18</a>.","apa":"Sun, R., &#38; Lampert, C. (2019). KS(conf): A light-weight test if a ConvNet operates outside of Its specifications (Vol. 11269, pp. 244–259). Presented at the GCPR: Conference on Pattern Recognition, Stuttgart, Germany: Springer Nature. <a href=\"https://doi.org/10.1007/978-3-030-12939-2_18\">https://doi.org/10.1007/978-3-030-12939-2_18</a>","ama":"Sun R, Lampert C. KS(conf): A light-weight test if a ConvNet operates outside of Its specifications. In: Vol 11269. Springer Nature; 2019:244-259. doi:<a href=\"https://doi.org/10.1007/978-3-030-12939-2_18\">10.1007/978-3-030-12939-2_18</a>","ista":"Sun R, Lampert C. 2019. KS(conf): A light-weight test if a ConvNet operates outside of Its specifications. GCPR: Conference on Pattern Recognition, LNCS, vol. 11269, 244–259.","mla":"Sun, Rémy, and Christoph Lampert. <i>KS(Conf): A Light-Weight Test If a ConvNet Operates Outside of Its Specifications</i>. Vol. 11269, Springer Nature, 2019, pp. 244–59, doi:<a href=\"https://doi.org/10.1007/978-3-030-12939-2_18\">10.1007/978-3-030-12939-2_18</a>.","short":"R. Sun, C. Lampert, in:, Springer Nature, 2019, pp. 244–259."},"year":"2019","external_id":{"arxiv":["1804.04171"]},"arxiv":1,"doi":"10.1007/978-3-030-12939-2_18","day":"14","abstract":[{"lang":"eng","text":"Computer vision systems for automatic image categorization have become accurate and reliable enough that they can run continuously for days or even years as components of real-world commercial applications. A major open problem in this context, however, is quality control. Good classification performance can only be expected if systems run under the specific conditions, in particular data distributions, that they were trained for. Surprisingly, none of the currently used deep network architectures have a built-in functionality that could detect if a network operates on data from a distribution it was not trained for, such that potentially a warning to the human users could be triggered. In this work, we describe KS(conf), a procedure for detecting such outside of specifications (out-of-specs) operation, based on statistical testing of the network outputs. We show by extensive experiments using the ImageNet, AwA2 and DAVIS datasets on a variety of ConvNets architectures that KS(conf) reliably detects out-of-specs situations. It furthermore has a number of properties that make it a promising candidate for practical deployment: it is easy to implement, adds almost no overhead to the system, works with all networks, including pretrained ones, and requires no a priori knowledge of how the data distribution could change. "}],"page":"244-259","ec_funded":1,"quality_controlled":"1","publisher":"Springer Nature","_id":"6482","scopus_import":"1","author":[{"full_name":"Sun, Rémy","last_name":"Sun","first_name":"Rémy"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","last_name":"Lampert","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"publication_status":"published","date_created":"2019-05-24T09:48:36Z","article_processing_charge":"No","department":[{"_id":"ChLa"}],"alternative_title":["LNCS"],"title":"KS(conf): A light-weight test if a ConvNet operates outside of Its specifications","intvolume":"     11269","main_file_link":[{"url":"https://arxiv.org/abs/1804.04171","open_access":"1"}],"status":"public","related_material":{"record":[{"relation":"later_version","id":"6944","status":"public"}]},"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","date_published":"2019-02-14T00:00:00Z","type":"conference","publication_identifier":{"issn":["0302-9743"],"eissn":["1611-3349"],"isbn":["9783030129385","9783030129392"]},"oa":1,"language":[{"iso":"eng"}],"conference":{"location":"Stuttgart, Germany","end_date":"2018-10-12","name":"GCPR: Conference on Pattern Recognition","start_date":"2018-10-09"},"oa_version":"Preprint","project":[{"name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036","call_identifier":"FP7","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"month":"02"},{"page":"2251 - 2265","quality_controlled":"1","article_type":"original","publisher":"Institute of Electrical and Electronics Engineers (IEEE)","author":[{"full_name":"Xian, Yongqin","first_name":"Yongqin","last_name":"Xian"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph","orcid":"0000-0002-4561-241X","last_name":"Lampert","first_name":"Christoph"},{"full_name":"Schiele, Bernt","last_name":"Schiele","first_name":"Bernt"},{"first_name":"Zeynep","last_name":"Akata","full_name":"Akata, Zeynep"}],"issue":"9","_id":"6554","scopus_import":"1","title":"Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly","intvolume":"        41","publication_status":"published","department":[{"_id":"ChLa"}],"article_processing_charge":"No","date_created":"2019-06-11T14:05:59Z","volume":41,"isi":1,"external_id":{"isi":["000480343900015"],"arxiv":["1707.00600"]},"date_updated":"2023-09-05T13:18:09Z","citation":{"ama":"Xian Y, Lampert C, Schiele B, Akata Z. Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly. <i>IEEE Transactions on Pattern Analysis and Machine Intelligence</i>. 2019;41(9):2251-2265. doi:<a href=\"https://doi.org/10.1109/tpami.2018.2857768\">10.1109/tpami.2018.2857768</a>","apa":"Xian, Y., Lampert, C., Schiele, B., &#38; Akata, Z. (2019). Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly. <i>IEEE Transactions on Pattern Analysis and Machine Intelligence</i>. Institute of Electrical and Electronics Engineers (IEEE). <a href=\"https://doi.org/10.1109/tpami.2018.2857768\">https://doi.org/10.1109/tpami.2018.2857768</a>","chicago":"Xian, Yongqin, Christoph Lampert, Bernt Schiele, and Zeynep Akata. “Zero-Shot Learning - A Comprehensive Evaluation of the Good, the Bad and the Ugly.” <i>IEEE Transactions on Pattern Analysis and Machine Intelligence</i>. Institute of Electrical and Electronics Engineers (IEEE), 2019. <a href=\"https://doi.org/10.1109/tpami.2018.2857768\">https://doi.org/10.1109/tpami.2018.2857768</a>.","ieee":"Y. Xian, C. Lampert, B. Schiele, and Z. Akata, “Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly,” <i>IEEE Transactions on Pattern Analysis and Machine Intelligence</i>, vol. 41, no. 9. Institute of Electrical and Electronics Engineers (IEEE), pp. 2251–2265, 2019.","short":"Y. Xian, C. Lampert, B. Schiele, Z. Akata, IEEE Transactions on Pattern Analysis and Machine Intelligence 41 (2019) 2251–2265.","mla":"Xian, Yongqin, et al. “Zero-Shot Learning - A Comprehensive Evaluation of the Good, the Bad and the Ugly.” <i>IEEE Transactions on Pattern Analysis and Machine Intelligence</i>, vol. 41, no. 9, Institute of Electrical and Electronics Engineers (IEEE), 2019, pp. 2251–65, doi:<a href=\"https://doi.org/10.1109/tpami.2018.2857768\">10.1109/tpami.2018.2857768</a>.","ista":"Xian Y, Lampert C, Schiele B, Akata Z. 2019. Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly. IEEE Transactions on Pattern Analysis and Machine Intelligence. 41(9), 2251–2265."},"year":"2019","abstract":[{"text":"Due to the importance of zero-shot learning, i.e. classifying images where there is a lack of labeled training data, the number of proposed approaches has recently increased steadily. We argue that it is time to take a step back and to analyze the status quo of the area. The purpose of this paper is three-fold. First, given the fact that there is no agreed upon zero-shot learning benchmark, we first define a new benchmark by unifying both the evaluation protocols and data splits of publicly available datasets used for this task. This is an important contribution as published results are often not comparable and sometimes even flawed due to, e.g. pre-training on zero-shot test classes. Moreover, we propose a new zero-shot learning dataset, the Animals with Attributes 2 (AWA2) dataset which we make publicly available both in terms of image features and the images themselves. Second, we compare and analyze a significant number of the state-of-the-art methods in depth, both in the classic zero-shot setting but also in the more realistic generalized zero-shot setting. Finally, we discuss in detail the limitations of the current status of the area which can be taken as a basis for advancing it.","lang":"eng"}],"arxiv":1,"doi":"10.1109/tpami.2018.2857768","day":"01","language":[{"iso":"eng"}],"publication":"IEEE Transactions on Pattern Analysis and Machine Intelligence","month":"09","oa_version":"Preprint","status":"public","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1707.00600"}],"date_published":"2019-09-01T00:00:00Z","type":"journal_article","oa":1,"publication_identifier":{"eissn":["1939-3539"],"issn":["0162-8828"]}}]