@article{10764, abstract = {Presynaptic glutamate replenishment is fundamental to brain function. In high activity regimes, such as epileptic episodes, this process is thought to rely on the glutamate-glutamine cycle between neurons and astrocytes. However the presence of an astroglial glutamine supply, as well as its functional relevance in vivo in the healthy brain remain controversial, partly due to a lack of tools that can directly examine glutamine transfer. Here, we generated a fluorescent probe that tracks glutamine in live cells, which provides direct visual evidence of an activity-dependent glutamine supply from astroglial networks to presynaptic structures under physiological conditions. This mobilization is mediated by connexin43, an astroglial protein with both gap-junction and hemichannel functions, and is essential for synaptic transmission and object recognition memory. Our findings uncover an indispensable recruitment of astroglial glutamine in physiological synaptic activity and memory via an unconventional pathway, thus providing an astrocyte basis for cognitive processes.}, author = {Cheung, Giselle T and Bataveljic, Danijela and Visser, Josien and Kumar, Naresh and Moulard, Julien and Dallérac, Glenn and Mozheiko, Daria and Rollenhagen, Astrid and Ezan, Pascal and Mongin, Cédric and Chever, Oana and Bemelmans, Alexis Pierre and Lübke, Joachim and Leray, Isabelle and Rouach, Nathalie}, issn = {20411723}, journal = {Nature Communications}, publisher = {Springer Nature}, title = {{Physiological synaptic activity and recognition memory require astroglial glutamine}}, doi = {10.1038/s41467-022-28331-7}, volume = {13}, year = {2022}, } @article{10765, abstract = {We establish the Hardy-Littlewood property (à la Borovoi-Rudnick) for Zariski open subsets in affine quadrics of the form q(x1,...,xn)=m, where q is a non-degenerate integral quadratic form in n>3 variables and m is a non-zero integer. This gives asymptotic formulas for the density of integral points taking coprime polynomial values, which is a quantitative version of the arithmetic purity of strong approximation property off infinity for affine quadrics.}, author = {Cao, Yang and Huang, Zhizhong}, issn = {1090-2082}, journal = {Advances in Mathematics}, number = {3}, publisher = {Elsevier}, title = {{Arithmetic purity of the Hardy-Littlewood property and geometric sieve for affine quadrics}}, doi = {10.1016/j.aim.2022.108236}, volume = {398}, year = {2022}, } @article{10766, abstract = {Tension of the actomyosin cell cortex plays a key role in determining cell–cell contact growth and size. The level of cortical tension outside of the cell–cell contact, when pulling at the contact edge, scales with the total size to which a cell–cell contact can grow [J.-L. Maître et al., Science 338, 253–256 (2012)]. Here, we show in zebrafish primary germ-layer progenitor cells that this monotonic relationship only applies to a narrow range of cortical tension increase and that above a critical threshold, contact size inversely scales with cortical tension. This switch from cortical tension increasing to decreasing progenitor cell–cell contact size is caused by cortical tension promoting E-cadherin anchoring to the actomyosin cytoskeleton, thereby increasing clustering and stability of E-cadherin at the contact. After tension-mediated E-cadherin stabilization at the contact exceeds a critical threshold level, the rate by which the contact expands in response to pulling forces from the cortex sharply drops, leading to smaller contacts at physiologically relevant timescales of contact formation. Thus, the activity of cortical tension in expanding cell–cell contact size is limited by tension-stabilizing E-cadherin–actin complexes at the contact.}, author = {Slovakova, Jana and Sikora, Mateusz K and Arslan, Feyza N and Caballero Mancebo, Silvia and Krens, Gabriel and Kaufmann, Walter and Merrin, Jack and Heisenberg, Carl-Philipp J}, issn = {10916490}, journal = {Proceedings of the National Academy of Sciences of the United States of America}, number = {8}, publisher = {Proceedings of the National Academy of Sciences}, title = {{Tension-dependent stabilization of E-cadherin limits cell-cell contact expansion in zebrafish germ-layer progenitor cells}}, doi = {10.1073/pnas.2122030119}, volume = {119}, year = {2022}, } @article{10767, abstract = {The t-haplotype of mice is a classical model for autosomal transmission distortion. A largely non-recombining variant of the proximal region of chromosome 17, it is transmitted to more than 90% of the progeny of heterozygous males through the disabling of sperm carrying a standard chromosome. While extensive genetic and functional work has shed light on individual genes involved in drive, much less is known about the evolution and function of the rest of its hundreds of genes. Here, we characterize the sequence and expression of dozens of t-specific transcripts and of their chromosome 17 homologues. Many genes showed reduced expression of the t-allele, but an equal number of genes showed increased expression of their t-copy, consistent with increased activity or a newly evolved function. Genes on the t-haplotype had a significantly higher non-synonymous substitution rate than their homologues on the standard chromosome, with several genes harbouring dN/dS ratios above 1. Finally, the t-haplotype has acquired at least two genes from other chromosomes, which show high and tissue-specific expression. These results provide a first overview of the gene content of this selfish element, and support a more dynamic evolutionary scenario than expected of a large genomic region with almost no recombination.}, author = {Kelemen, Réka K and Elkrewi, Marwan N and Lindholm, Anna K. and Vicoso, Beatriz}, issn = {14712954}, journal = {Proceedings of the Royal Society B: Biological Sciences}, number = {1968}, pages = {20211985}, publisher = {The Royal Society}, title = {{Novel patterns of expression and recruitment of new genes on the t-haplotype, a mouse selfish chromosome}}, doi = {10.1098/rspb.2021.1985}, volume = {289}, year = {2022}, } @article{10768, abstract = {Among the most fascinated properties of the plant hormone auxin is its ability to promote formation of its own directional transport routes. These gradually narrowing auxin channels form from the auxin source toward the sink and involve coordinated, collective polarization of individual cells. Once established, the channels provide positional information, along which new vascular strands form, for example, during organogenesis, regeneration, or leave venation. The main prerequisite of this still mysterious auxin canalization mechanism is a feedback between auxin signaling and its directional transport. This is manifested by auxin-induced re-arrangements of polar, subcellular localization of PIN-FORMED (PIN) auxin exporters. Immanent open questions relate to how position of auxin source and sink as well as tissue context are sensed and translated into tissue polarization and how cells communicate to polarize coordinately. Recently, identification of the first molecular players opens new avenues into molecular studies of this intriguing example of self-organizing plant development.}, author = {Hajny, Jakub and Tan, Shutang and Friml, Jiří}, issn = {1369-5266}, journal = {Current Opinion in Plant Biology}, number = {2}, publisher = {Elsevier}, title = {{Auxin canalization: From speculative models toward molecular players}}, doi = {10.1016/j.pbi.2022.102174}, volume = {65}, year = {2022}, } @article{10769, abstract = {studiamos aspectos de Teoría Cuántica de Campos a densidad finita usando técnicas y conceptos de información cuántica. Nos enfocamos en fermiones de Dirac masivos con potencial químico en 1+1 dimensiones espacio-temporales. Usando la entropía de entrelazamiento en un intervalo, construimos la función c entrópica que es finita. Esta función c no es monótona, e incorpora el entrelazamiento de largo alcance proveniente de la superficie de Fermi. Motivados por trabajos previos de modelos en la red, calculamos numéricamente las entropías de Renyi y encontramos oscilaciones de Friedel. Seguidamente, analizamos la información mutua como una medida de correlación entre diferentes regiones. Usando una expansión de distancia grande desarrollada por Cardy, argumentamos que la información mutua detecta las correlaciones inducidas por la superficie de Fermi todavía al orden dominante en la expansión. Finalmente, analizamos la entropía relativa y sus generalizaciones de Renyi para distinguir estados con diferente carga. Encontramos que estados en diferentes sectores de superselección dan origen a un comportamiento super-extensivo en la entropía relativa.}, author = {Daguerre, L. and Torroba, G. and Medina Ramos, Raimel A and Solís, M.}, issn = {18501168}, journal = {Anales de la Asociacion Fisica Argentina}, number = {4}, pages = {93--98}, publisher = {Asociación Física Argentina}, title = {{Non relativistic quantum field theory: Dynamics and irreversibility}}, doi = {10.31527/analesafa.2021.32.4.93}, volume = {32}, year = {2022}, } @article{10771, abstract = {A critical overview of the theory of the chirality-induced spin selectivity (CISS) effect, that is, phenomena in which the chirality of molecular species imparts significant spin selectivity to various electron processes, is provided. Based on discussions in a recently held workshop, and further work published since, the status of CISS effects—in electron transmission, electron transport, and chemical reactions—is reviewed. For each, a detailed discussion of the state-of-the-art in theoretical understanding is provided and remaining challenges and research opportunities are identified.}, author = {Evers, Ferdinand and Aharony, Amnon and Bar-Gill, Nir and Entin-Wohlman, Ora and Hedegård, Per and Hod, Oded and Jelinek, Pavel and Kamieniarz, Grzegorz and Lemeshko, Mikhail and Michaeli, Karen and Mujica, Vladimiro and Naaman, Ron and Paltiel, Yossi and Refaely-Abramson, Sivan and Tal, Oren and Thijssen, Jos and Thoss, Michael and Van Ruitenbeek, Jan M. and Venkataraman, Latha and Waldeck, David H. and Yan, Binghai and Kronik, Leeor}, issn = {15214095}, journal = {Advanced Materials}, number = {13}, publisher = {Wiley}, title = {{Theory of chirality induced spin selectivity: Progress and challenges}}, doi = {10.1002/adma.202106629}, volume = {34}, year = {2022}, } @article{10772, abstract = {We introduce tropical corals, balanced trees in a half-space, and show that they correspond to holomorphic polygons capturing the product rule in Lagrangian Floer theory for the elliptic curve. We then prove a correspondence theorem equating counts of tropical corals to punctured log Gromov–Witten invariants of the Tate curve. This implies that the homogeneous coordinate ring of the mirror to the Tate curve is isomorphic to the degree-zero part of symplectic cohomology, confirming a prediction of homological mirror symmetry.}, author = {Arguez, Nuroemuer Huelya}, issn = {1469-7750}, journal = {Journal of the London Mathematical Society}, number = {1}, pages = {343--411}, publisher = {London Mathematical Society}, title = {{Mirror symmetry for the Tate curve via tropical and log corals}}, doi = {10.1112/jlms.12515}, volume = {105}, year = {2022}, } @article{10773, abstract = {The Voronoi tessellation in Rd is defined by locally minimizing the power distance to given weighted points. Symmetrically, the Delaunay mosaic can be defined by locally maximizing the negative power distance to other such points. We prove that the average of the two piecewise quadratic functions is piecewise linear, and that all three functions have the same critical points and values. Discretizing the two piecewise quadratic functions, we get the alpha shapes as sublevel sets of the discrete function on the Delaunay mosaic, and analogous shapes as superlevel sets of the discrete function on the Voronoi tessellation. For the same non-critical value, the corresponding shapes are disjoint, separated by a narrow channel that contains no critical points but the entire level set of the piecewise linear function.}, author = {Biswas, Ranita and Cultrera Di Montesano, Sebastiano and Edelsbrunner, Herbert and Saghafian, Morteza}, issn = {1432-0444}, journal = {Discrete and Computational Geometry}, pages = {811--842}, publisher = {Springer Nature}, title = {{Continuous and discrete radius functions on Voronoi tessellations and Delaunay mosaics}}, doi = {10.1007/s00454-022-00371-2}, volume = {67}, year = {2022}, } @inproceedings{10774, abstract = {We study the problem of specifying sequential information-flow properties of systems. Information-flow properties are hyperproperties, as they compare different traces of a system. Sequential information-flow properties can express changes, over time, in the information-flow constraints. For example, information-flow constraints during an initialization phase of a system may be different from information-flow constraints that are required during the operation phase. We formalize several variants of interpreting sequential information-flow constraints, which arise from different assumptions about what can be observed of the system. For this purpose, we introduce a first-order logic, called Hypertrace Logic, with both trace and time quantifiers for specifying linear-time hyperproperties. We prove that HyperLTL, which corresponds to a fragment of Hypertrace Logic with restricted quantifier prefixes, cannot specify the majority of the studied variants of sequential information flow, including all variants in which the transition between sequential phases (such as initialization and operation) happens asynchronously. Our results rely on new equivalences between sets of traces that cannot be distinguished by certain classes of formulas from Hypertrace Logic. This presents a new approach to proving inexpressiveness results for HyperLTL.}, author = {Bartocci, Ezio and Ferrere, Thomas and Henzinger, Thomas A and Nickovic, Dejan and Da Costa, Ana Oliveira}, booktitle = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}, isbn = {9783030945824}, issn = {16113349}, location = {Philadelphia, PA, United States}, pages = {1--19}, publisher = {Springer Nature}, title = {{Flavors of sequential information flow}}, doi = {10.1007/978-3-030-94583-1_1}, volume = {13182}, year = {2022}, } @article{10775, abstract = {List-decodability of Reed–Solomon codes has received a lot of attention, but the best-possible dependence between the parameters is still not well-understood. In this work, we focus on the case where the list-decoding radius is of the form r = 1-ε for ε tending to zero. Our main result states that there exist Reed–Solomon codes with rate Ω(ε) which are (1 - ε, O(1/ε))-list-decodable, meaning that any Hamming ball of radius 1-ε contains at most O(1/ε) codewords. This trade-off between rate and list-decoding radius is best-possible for any code with list size less than exponential in the block length. By achieving this trade-off between rate and list-decoding radius we improve a recent result of Guo, Li, Shangguan, Tamo, and Wootters, and resolve the main motivating question of their work. Moreover, while their result requires the field to be exponentially large in the block length, we only need the field size to be polynomially large (and in fact, almost-linear suffices). We deduce our main result from a more general theorem, in which we prove good list-decodability properties of random puncturings of any given code with very large distance.}, author = {Ferber, Asaf and Kwan, Matthew Alan and Sauermann, Lisa}, issn = {1557-9654}, journal = {IEEE Transactions on Information Theory}, number = {6}, pages = {3823--3828}, publisher = {IEEE}, title = {{List-decodability with large radius for Reed-Solomon codes}}, doi = {10.1109/TIT.2022.3148779}, volume = {68}, year = {2022}, } @article{10776, abstract = {Let K be a convex body in Rn (i.e., a compact convex set with nonempty interior). Given a point p in the interior of K, a hyperplane h passing through p is called barycentric if p is the barycenter of K∩h. In 1961, Grünbaum raised the question whether, for every K, there exists an interior point p through which there are at least n+1 distinct barycentric hyperplanes. Two years later, this was seemingly resolved affirmatively by showing that this is the case if p=p0 is the point of maximal depth in K. However, while working on a related question, we noticed that one of the auxiliary claims in the proof is incorrect. Here, we provide a counterexample; this re-opens Grünbaum’s question. It follows from known results that for n≥2, there are always at least three distinct barycentric cuts through the point p0∈K of maximal depth. Using tools related to Morse theory we are able to improve this bound: four distinct barycentric cuts through p0 are guaranteed if n≥3.}, author = {Patakova, Zuzana and Tancer, Martin and Wagner, Uli}, issn = {1432-0444}, journal = {Discrete and Computational Geometry}, pages = {1133--1154}, publisher = {Springer Nature}, title = {{Barycentric cuts through a convex body}}, doi = {10.1007/s00454-021-00364-7}, volume = {68}, year = {2022}, } @article{10787, abstract = {A species distributed across diverse environments may adapt to local conditions. We ask how quickly such a species changes its range in response to changed conditions. Szép et al. (Szép E, Sachdeva H, Barton NH. 2021 Polygenic local adaptation in metapopulations: a stochastic eco-evolutionary model. Evolution75, 1030–1045 (doi:10.1111/evo.14210)) used the infinite island model to find the stationary distribution of allele frequencies and deme sizes. We extend this to find how a metapopulation responds to changes in carrying capacity, selection strength, or migration rate when deme sizes are fixed. We further develop a ‘fixed-state’ approximation. Under this approximation, polymorphism is only possible for a narrow range of habitat proportions when selection is weak compared to drift, but for a much wider range otherwise. When rates of selection or migration relative to drift change in a single deme of the metapopulation, the population takes a time of order m−1 to reach the new equilibrium. However, even with many loci, there can be substantial fluctuations in net adaptation, because at each locus, alleles randomly get lost or fixed. Thus, in a finite metapopulation, variation may gradually be lost by chance, even if it would persist in an infinite metapopulation. When conditions change across the whole metapopulation, there can be rapid change, which is predicted well by the fixed-state approximation. This work helps towards an understanding of how metapopulations extend their range across diverse environments. This article is part of the theme issue ‘Species’ ranges in the face of changing environments (Part II)’.}, author = {Barton, Nicholas H and Olusanya, Oluwafunmilola O}, issn = {1471-2970}, journal = {Philosophical Transactions of the Royal Society B: Biological Sciences}, keywords = {General Agricultural and Biological Sciences, General Biochemistry, Genetics and Molecular Biology}, number = {1848}, publisher = {The Royal Society}, title = {{The response of a metapopulation to a changing environment}}, doi = {10.1098/rstb.2021.0009}, volume = {377}, year = {2022}, } @unpublished{10788, abstract = {We determine an asymptotic formula for the number of integral points of bounded height on a certain toric variety, which is incompatible with part of a preprint by Chambert-Loir and Tschinkel. We provide an alternative interpretation of the asymptotic formula we get. To do so, we construct an analogue of Peyre's constant $\alpha$ and describe its relation to a new obstruction to the Zariski density of integral points in certain regions of varieties.}, author = {Wilsch, Florian Alexander}, booktitle = {arXiv}, keywords = {Integral point, toric variety, Manin's conjecture}, title = {{Integral points of bounded height on a certain toric variety}}, doi = {10.48550/arXiv.2202.10909}, year = {2022}, } @article{10791, abstract = {The mammalian neocortex is composed of diverse neuronal and glial cell classes that broadly arrange in six distinct laminae. Cortical layers emerge during development and defects in the developmental programs that orchestrate cortical lamination are associated with neurodevelopmental diseases. The developmental principle of cortical layer formation depends on concerted radial projection neuron migration, from their birthplace to their final target position. Radial migration occurs in defined sequential steps, regulated by a large array of signaling pathways. However, based on genetic loss-of-function experiments, most studies have thus far focused on the role of cell-autonomous gene function. Yet, cortical neuron migration in situ is a complex process and migrating neurons traverse along diverse cellular compartments and environments. The role of tissue-wide properties and genetic state in radial neuron migration is however not clear. Here we utilized mosaic analysis with double markers (MADM) technology to either sparsely or globally delete gene function, followed by quantitative single-cell phenotyping. The MADM-based gene ablation paradigms in combination with computational modeling demonstrated that global tissue-wide effects predominate cell-autonomous gene function albeit in a gene-specific manner. Our results thus suggest that the genetic landscape in a tissue critically affects the overall migration phenotype of individual cortical projection neurons. In a broader context, our findings imply that global tissue-wide effects represent an essential component of the underlying etiology associated with focal malformations of cortical development in particular, and neurological diseases in general.}, author = {Hansen, Andi H and Pauler, Florian and Riedl, Michael and Streicher, Carmen and Heger, Anna-Magdalena and Laukoter, Susanne and Sommer, Christoph M and Nicolas, Armel and Hof, Björn and Tsai, Li Huei and Rülicke, Thomas and Hippenmeyer, Simon}, issn = {2753-149X}, journal = {Oxford Open Neuroscience}, number = {1}, publisher = {Oxford Academic}, title = {{Tissue-wide effects override cell-intrinsic gene function in radial neuron migration}}, doi = {10.1093/oons/kvac009}, volume = {1}, year = {2022}, } @unpublished{10792, abstract = {Background Proper cerebral cortical development depends on the tightly orchestrated migration of newly born neurons from the inner ventricular and subventricular zones to the outer cortical plate. Any disturbance in this process during prenatal stages may lead to neuronal migration disorders (NMDs), which can vary in extent from focal to global. Furthermore, NMDs show a substantial comorbidity with other neurodevelopmental disorders, notably autism spectrum disorders (ASDs). Our previous work demonstrated focal neuronal migration defects in mice carrying loss-of-function alleles of the recognized autism risk gene WDFY3. However, the cellular origins of these defects in Wdfy3 mutant mice remain elusive and uncovering it will provide critical insight into WDFY3-dependent disease pathology . Methods Here, in an effort to untangle the origins of NMDs in Wdfy3lacZ mice, we employed mosaic analysis with double markers (MADM). MADM technology enabled us to genetically distinctly track and phenotypically analyze mutant and wild type cells concomitantly in vivo using immunofluorescent techniques. Results We revealed a cell autonomous requirement of WDFY3 for accurate laminar positioning of cortical projection neurons and elimination of mispositioned cells during early postnatal life. In addition, we identified significant deviations in dendritic arborization, as well as synaptic density and morphology between wild type, heterozygous, and homozygous Wdfy3 mutant neurons in Wdfy3-MADM reporter mice at postnatal stages. Limitations While Wdfy3 mutant mice have provided valuable insight into prenatal aspects of ASD pathology that remain inaccessible to investigation in humans, like most animal models, they do not a perfectly replicate all aspects of human ASD biology. The lack of human data makes it indeterminate whether morphological deviations described here apply to ASD patients. Conclusions Our genetic approach revealed several cell autonomous requirements of Wdfy3 in neuronal development that could underly the pathogenic mechanisms of WDFY3-related ASD conditions. The results are also consistent with findings in other ASD animal models and patients and suggest an important role for Wdfy3 in regulating neuronal function and interconnectivity in postnatal life.}, author = {Schaaf, Zachary and Tat, Lyvin and Cannizzaro, Noemi and Green, Ralph and Rülicke, Thomas and Hippenmeyer, Simon and Zarbalis, K}, issn = {2693-5015}, pages = {30}, publisher = {Research Square}, title = {{WDFY3 cell autonomously controls neuronal migration}}, doi = {10.21203/rs.3.rs-1316167/v1}, year = {2022}, } @article{10797, abstract = {We consider symmetric partial exclusion and inclusion processes in a general graph in contact with reservoirs, where we allow both for edge disorder and well-chosen site disorder. We extend the classical dualities to this context and then we derive new orthogonal polynomial dualities. From the classical dualities, we derive the uniqueness of the non-equilibrium steady state and obtain correlation inequalities. Starting from the orthogonal polynomial dualities, we show universal properties of n-point correlation functions in the non-equilibrium steady state for systems with at most two different reservoir parameters, such as a chain with reservoirs at left and right ends.}, author = {Floreani, Simone and Redig, Frank and Sau, Federico}, issn = {0246-0203}, journal = {Annales de l'institut Henri Poincare (B) Probability and Statistics}, number = {1}, pages = {220--247}, publisher = {Institute of Mathematical Statistics}, title = {{Orthogonal polynomial duality of boundary driven particle systems and non-equilibrium correlations}}, doi = {10.1214/21-AIHP1163}, volume = {58}, year = {2022}, } @phdthesis{10799, abstract = {Because of the increasing popularity of machine learning methods, it is becoming important to understand the impact of learned components on automated decision-making systems and to guarantee that their consequences are beneficial to society. In other words, it is necessary to ensure that machine learning is sufficiently trustworthy to be used in real-world applications. This thesis studies two properties of machine learning models that are highly desirable for the sake of reliability: robustness and fairness. In the first part of the thesis we study the robustness of learning algorithms to training data corruption. Previous work has shown that machine learning models are vulnerable to a range of training set issues, varying from label noise through systematic biases to worst-case data manipulations. This is an especially relevant problem from a present perspective, since modern machine learning methods are particularly data hungry and therefore practitioners often have to rely on data collected from various external sources, e.g. from the Internet, from app users or via crowdsourcing. Naturally, such sources vary greatly in the quality and reliability of the data they provide. With these considerations in mind, we study the problem of designing machine learning algorithms that are robust to corruptions in data coming from multiple sources. We show that, in contrast to the case of a single dataset with outliers, successful learning within this model is possible both theoretically and practically, even under worst-case data corruptions. The second part of this thesis deals with fairness-aware machine learning. There are multiple areas where machine learning models have shown promising results, but where careful considerations are required, in order to avoid discrimanative decisions taken by such learned components. Ensuring fairness can be particularly challenging, because real-world training datasets are expected to contain various forms of historical bias that may affect the learning process. In this thesis we show that data corruption can indeed render the problem of achieving fairness impossible, by tightly characterizing the theoretical limits of fair learning under worst-case data manipulations. However, assuming access to clean data, we also show how fairness-aware learning can be made practical in contexts beyond binary classification, in particular in the challenging learning to rank setting.}, author = {Konstantinov, Nikola H}, isbn = {978-3-99078-015-2}, issn = {2663-337X}, keywords = {robustness, fairness, machine learning, PAC learning, adversarial learning}, pages = {176}, publisher = {Institute of Science and Technology Austria}, title = {{Robustness and fairness in machine learning}}, doi = {10.15479/at:ista:10799}, year = {2022}, } @article{10802, abstract = {Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. While many approaches have been developed for training fair models from data, little is known about the robustness of these methods to data corruption. In this work we consider fairness-aware learning under worst-case data manipulations. We show that an adversary can in some situations force any learner to return an overly biased classifier, regardless of the sample size and with or without degrading accuracy, and that the strength of the excess bias increases for learning problems with underrepresented protected groups in the data. We also prove that our hardness results are tight up to constant factors. To this end, we study two natural learning algorithms that optimize for both accuracy and fairness and show that these algorithms enjoy guarantees that are order-optimal in terms of the corruption ratio and the protected groups frequencies in the large data limit.}, author = {Konstantinov, Nikola H and Lampert, Christoph}, issn = {1533-7928}, journal = {Journal of Machine Learning Research}, keywords = {Fairness, robustness, data poisoning, trustworthy machine learning, PAC learning}, pages = {1--60}, publisher = {ML Research Press}, title = {{Fairness-aware PAC learning from corrupted data}}, volume = {23}, year = {2022}, } @article{10812, abstract = {Several promising strategies based on combining or cycling different antibiotics have been proposed to increase efficacy and counteract resistance evolution, but we still lack a deep understanding of the physiological responses and genetic mechanisms that underlie antibiotic interactions and the clinical applicability of these strategies. In antibiotic-exposed bacteria, the combined effects of physiological stress responses and emerging resistance mutations (occurring at different time scales) generate complex and often unpredictable dynamics. In this Review, we present our current understanding of bacterial cell physiology and genetics of responses to antibiotics. We emphasize recently discovered mechanisms of synergistic and antagonistic drug interactions, hysteresis in temporal interactions between antibiotics that arise from microbial physiology and interactions between antibiotics and resistance mutations that can cause collateral sensitivity or cross-resistance. We discuss possible connections between the different phenomena and indicate relevant research directions. A better and more unified understanding of drug and genetic interactions is likely to advance antibiotic therapy.}, author = {Römhild, Roderich and Bollenbach, Mark Tobias and Andersson, Dan I.}, issn = {1740-1534}, journal = {Nature Reviews Microbiology}, keywords = {General Immunology and Microbiology, Microbiology, Infectious Diseases}, pages = {478--490}, publisher = {Springer Nature}, title = {{The physiology and genetics of bacterial responses to antibiotic combinations}}, doi = {10.1038/s41579-022-00700-5}, volume = {20}, year = {2022}, }