{"year":"2023","acknowledgement":"Open access funding provided by Università degli Studi di Trieste within the CRUI-CARE Agreement. Funding was provided by Austrian Science Fund (Grant No. F65), Horizon 2020 (Grant No. 754411) and Österreichische Forschungsförderungsgesellschaft.","project":[{"grant_number":"F6504","name":"Taming Complexity in Partial Differential Systems","_id":"fc31cba2-9c52-11eb-aca3-ff467d239cd2"},{"call_identifier":"H2020","grant_number":"754411","_id":"260C2330-B435-11E9-9278-68D0E5697425","name":"ISTplus - Postdoctoral Fellowships"}],"doi":"10.1007/s00521-023-09033-7","publication":"Neural Computing and Applications","author":[{"first_name":"Federico","last_name":"Cornalba","id":"2CEB641C-A400-11E9-A717-D712E6697425","full_name":"Cornalba, Federico","orcid":"0000-0002-6269-5149"},{"full_name":"Disselkamp, Constantin","first_name":"Constantin","last_name":"Disselkamp"},{"full_name":"Scassola, Davide","first_name":"Davide","last_name":"Scassola"},{"first_name":"Christopher","last_name":"Helf","full_name":"Helf, Christopher"}],"external_id":{"arxiv":["2203.04579"]},"publication_identifier":{"eissn":["1433-3058"],"issn":["0941-0643"]},"ec_funded":1,"citation":{"short":"F. Cornalba, C. Disselkamp, D. Scassola, C. Helf, Neural Computing and Applications (2023).","ista":"Cornalba F, Disselkamp C, Scassola D, Helf C. 2023. Multi-objective reward generalization: improving performance of Deep Reinforcement Learning for applications in single-asset trading. Neural Computing and Applications.","apa":"Cornalba, F., Disselkamp, C., Scassola, D., &#38; Helf, C. (2023). Multi-objective reward generalization: improving performance of Deep Reinforcement Learning for applications in single-asset trading. <i>Neural Computing and Applications</i>. Springer Nature. <a href=\"https://doi.org/10.1007/s00521-023-09033-7\">https://doi.org/10.1007/s00521-023-09033-7</a>","chicago":"Cornalba, Federico, Constantin Disselkamp, Davide Scassola, and Christopher Helf. “Multi-Objective Reward Generalization: Improving Performance of Deep Reinforcement Learning for Applications in Single-Asset Trading.” <i>Neural Computing and Applications</i>. Springer Nature, 2023. <a href=\"https://doi.org/10.1007/s00521-023-09033-7\">https://doi.org/10.1007/s00521-023-09033-7</a>.","ama":"Cornalba F, Disselkamp C, Scassola D, Helf C. Multi-objective reward generalization: improving performance of Deep Reinforcement Learning for applications in single-asset trading. <i>Neural Computing and Applications</i>. 2023. doi:<a href=\"https://doi.org/10.1007/s00521-023-09033-7\">10.1007/s00521-023-09033-7</a>","ieee":"F. Cornalba, C. Disselkamp, D. Scassola, and C. Helf, “Multi-objective reward generalization: improving performance of Deep Reinforcement Learning for applications in single-asset trading,” <i>Neural Computing and Applications</i>. Springer Nature, 2023.","mla":"Cornalba, Federico, et al. “Multi-Objective Reward Generalization: Improving Performance of Deep Reinforcement Learning for Applications in Single-Asset Trading.” <i>Neural Computing and Applications</i>, Springer Nature, 2023, doi:<a href=\"https://doi.org/10.1007/s00521-023-09033-7\">10.1007/s00521-023-09033-7</a>."},"article_type":"original","language":[{"iso":"eng"}],"_id":"14451","oa":1,"oa_version":"Published Version","date_created":"2023-10-22T22:01:16Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publisher":"Springer Nature","scopus_import":"1","title":"Multi-objective reward generalization: improving performance of Deep Reinforcement Learning for applications in single-asset trading","quality_controlled":"1","status":"public","type":"journal_article","day":"05","month":"10","article_processing_charge":"Yes (via OA deal)","date_updated":"2023-10-31T10:58:28Z","date_published":"2023-10-05T00:00:00Z","publication_status":"epub_ahead","department":[{"_id":"JuFi"}],"abstract":[{"text":"We investigate the potential of Multi-Objective, Deep Reinforcement Learning for stock and cryptocurrency single-asset trading: in particular, we consider a Multi-Objective algorithm which generalizes the reward functions and discount factor (i.e., these components are not specified a priori, but incorporated in the learning process). Firstly, using several important assets (BTCUSD, ETHUSDT, XRPUSDT, AAPL, SPY, NIFTY50), we verify the reward generalization property of the proposed Multi-Objective algorithm, and provide preliminary statistical evidence showing increased predictive stability over the corresponding Single-Objective strategy. Secondly, we show that the Multi-Objective algorithm has a clear edge over the corresponding Single-Objective strategy when the reward mechanism is sparse (i.e., when non-null feedback is infrequent over time). Finally, we discuss the generalization properties with respect to the discount factor. The entirety of our code is provided in open-source format.","lang":"eng"}],"main_file_link":[{"url":"https://doi.org/10.1007/s00521-023-09033-7","open_access":"1"}]}