{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T21:29:31Z","timestamp":1773264571409,"version":"3.50.1"},"reference-count":25,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T00:00:00Z","timestamp":1773014400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"The Ministry of Education of Humanities and Social Science Project of China","award":["21YJCZH030"],"award-info":[{"award-number":["21YJCZH030"]}]},{"name":"Nature Science Foundation of Shaanxi Province","award":["2024JC-YBMS-601"],"award-info":[{"award-number":["2024JC-YBMS-601"]}]},{"name":"Key Research and Development Program of Shaanxi Province","award":["2023-YBSF-28"],"award-info":[{"award-number":["2023-YBSF-28"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems"],"abstract":"<jats:p>Portfolio selection is a fundamental task in quantitative finance that aims to allocate capital across assets to balance risk and return. While deep learning has shown great promise in this field, extracting reliable feature representations from non-stationary and noisy financial data remains a significant challenge. The existing models often fail to simultaneously capture the temporal dynamics of price series and complex inter-asset correlations, which limits their trading performance. To address these issues, we propose Denoising-Sequence-Correlation Reinforcement Learning (DSCRL), a novel portfolio selection framework based on deep reinforcement learning. DSCRL employs a dual-stream feature extraction network, where one stream aims to learn temporal market dynamics and the other aims to capture asset correlations, enabling more informative representations. A denoising module is further integrated to mitigate the impact of noise, ensuring stability and robustness in the learning process. Furthermore, a deterministic policy gradient (DPG)-based decision network is designed to directly optimize continuous portfolio weights and normalize them to satisfy budget constraints while preserving the importance. Extensive experiments conducted on multiple benchmark datasets demonstrate that DSCRL consistently outperforms both traditional financial heuristics and advanced deep reinforcement approaches. The results highlight its superior ability to achieve higher cumulative returns with lower volatility. Overall, DSCRL provides an effective and robust solution that strikes a better trade-off between pursuing profits and managing risks in dynamic financial markets.<\/jats:p>","DOI":"10.3390\/systems14030292","type":"journal-article","created":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T15:49:42Z","timestamp":1773071382000},"page":"292","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Novel Portfolio Selection Method via Deep Reinforcement Learning"],"prefix":"10.3390","volume":"14","author":[{"given":"Ni","family":"Gao","sequence":"first","affiliation":[{"name":"School of Economics and Finance, Xi\u2019an International Studies University, Xi\u2019an 710128, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yan","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Economics and Finance, Xi\u2019an International Studies University, Xi\u2019an 710128, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yiyue","family":"He","sequence":"additional","affiliation":[{"name":"School of Economics and Management, Northwest University, Xi\u2019an 710127, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Juan","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Journalism and Communication, Shaanxi Normal University, Xi\u2019an 710019, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lefang","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Economics and Finance, Xi\u2019an Jiaotong University, Xi\u2019an 710049, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,9]]},"reference":[{"key":"ref_1","first-page":"77","article-title":"Portfolio Selection","volume":"7","author":"Markowitz","year":"1952","journal-title":"J. Financ."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"654","DOI":"10.1016\/j.ejor.2017.11.054","article-title":"Deep Learning with Long Short-Term Memory Networks for Financial Market Predictions","volume":"270","author":"Fischer","year":"2018","journal-title":"Eur. J. Oper. Res."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"4468","DOI":"10.1609\/aaai.v36i4.20369","article-title":"FactorVAE: A Probabilistic Dynamic Factor Model Based on Variational Autoencoder for Predicting Cross-Sectional Stock Returns","volume":"36","author":"Duan","year":"2022","journal-title":"AAAI"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"643","DOI":"10.1609\/aaai.v35i1.16144","article-title":"DeepTrader: A Deep Reinforcement Learning Approach for Risk-Return Balanced Portfolio Management with Market Conditions Embedding","volume":"35","author":"Wang","year":"2021","journal-title":"AAAI"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"119707","DOI":"10.1016\/j.eswa.2023.119707","article-title":"Asset Correlation Based Deep Reinforcement Learning for the Portfolio Selection","volume":"221","author":"Zhao","year":"2023","journal-title":"Expert Syst. Appl."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"113973","DOI":"10.1016\/j.eswa.2020.113973","article-title":"Portfolio Optimization with Return Prediction Using Deep Learning and Machine Learning","volume":"165","author":"Ma","year":"2021","journal-title":"Expert Syst. Appl."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1357","DOI":"10.1111\/j.1467-6486.2006.00648.x","article-title":"Noise Trading and the Management of Operational Risk; Firms, Traders and Irrationality in Financial Markets","volume":"43","author":"Willman","year":"2006","journal-title":"J. Manag. Stud."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Dragomiretskiy, K., and Zosso, D. (2013). IEEE Transactions on Signal Processing Publication Information. IEEE Trans. Signal Process., 61.","DOI":"10.1109\/TSP.2013.2265116"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Xu, K., Zhang, Y., Ye, D., Zhao, P., and Tan, M. (2020, January 11). Relation-Aware Transformer for Portfolio Policy Learning. Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence; International Joint Conferences on Artificial Intelligence Organization, Yokohama, Japan.","DOI":"10.24963\/ijcai.2020\/641"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1080\/21642583.2019.1708830","article-title":"A novel swarm intelligence optimization approach: Sparrow search algorithm","volume":"8","author":"Xue","year":"2020","journal-title":"Syst. Sci. Control. Eng."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/S1001-0742(12)60029-1","article-title":"Forecasting of wheat price Based on multiscale analysis","volume":"25","author":"Wang","year":"2016","journal-title":"Chin. J. Manag. Sci."},{"key":"ref_12","first-page":"59","article-title":"Research on prediction modeling of stock market index based on CEEMDAN-LSTM","volume":"35","author":"He","year":"2020","journal-title":"Stat. Inf. Forum"},{"key":"ref_13","unstructured":"Bai, S., Kolter, J.Z., and Koltun, V. (2018). An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv."},{"key":"ref_14","first-page":"5998","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"875","DOI":"10.1109\/72.935097","article-title":"Learning to Trade via Direct Reinforcement","volume":"12","author":"Moody","year":"2001","journal-title":"IEEE Trans. Neural Netw."},{"key":"ref_16","first-page":"236","article-title":"Cost-Sensitive Portfolio Selection via Deep Reinforcement Learning","volume":"34","author":"Zhang","year":"2020","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_17","unstructured":"Li, B., and Hoi, S.C.H. (2012). On-Line Portfolio Selection with Moving Average Reversion. arXiv."},{"key":"ref_18","unstructured":"Jiang, Z., Xu, D., and Liang, J. (2017). A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem. arXiv."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1517","DOI":"10.1609\/aaai.v31i1.10728","article-title":"Portfolio Selection via Subset Resampling","volume":"31","author":"Shen","year":"2017","journal-title":"AAAI"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.1467-9965.1991.tb00002.x","article-title":"Universal Portfolios","volume":"1","author":"Cover","year":"1991","journal-title":"Math. Financ."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"579","DOI":"10.1613\/jair.1336","article-title":"Can We Learn to Beat the Best Stock","volume":"21","author":"Borodin","year":"2004","journal-title":"JAIR"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Yang, H., Liu, X.-Y., Zhong, S., and Walid, A. (2020, January 15\u201316). Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy. Proceedings of the First ACM International Conference on AI in Finance, New York, NY, USA.","DOI":"10.1145\/3383455.3422540"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Gao, Z., Gao, Y., Hu, Y., Jiang, Z., and Su, J. (2020, January 8\u201311). Application of Deep Q-Network in Portfolio Management. Proceedings of the 2020 5th IEEE International Conference on Big Data Analytics (ICBDA), Xiamen, China.","DOI":"10.1109\/ICBDA49040.2020.9101333"},{"key":"ref_24","unstructured":"Liang, Z., Chen, H., Zhu, J., Jiang, K., and Li, Y. (2018). Adversarial Deep Reinforcement Learning in Portfolio Management. arXiv."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Karzanov, D., Garz\u00f3n, R., Terekhov, M., Gulcehre, C., Raffinot, T., and Detyniecki, M. (2025). Regret-Optimized Portfolio Enhancement through Deep Reinforcement Learning and Future Looking Rewards. Proceedings of the 6th ACM International Conference on AI in Finance, Singapore, 15\u201318 November 2025, ACM.","DOI":"10.1145\/3768292.3770340"}],"container-title":["Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/3\/292\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T05:28:09Z","timestamp":1773206889000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/3\/292"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,9]]},"references-count":25,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["systems14030292"],"URL":"https:\/\/doi.org\/10.3390\/systems14030292","relation":{},"ISSN":["2079-8954"],"issn-type":[{"value":"2079-8954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,9]]}}}