{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/market-self-learning-of-signals-impact-and","title":"Market Self-Learning of Signals, Impact and Optimal Trading: Invisible Hand Inference with Free Energy","arxiv_id":"1805.06126","date":"2018-05-16","proceeding":null,"authors":["Igor Halperin","Ilya Feldshteyn"],"abstract":"We present a simple model of a non-equilibrium self-organizing market where\nasset prices are partially driven by investment decisions of a bounded-rational\nagent. The agent acts in a stochastic market environment driven by various\nexogenous \"alpha\" signals, agent's own actions (via market impact), and noise.\nUnlike traditional agent-based models, our agent aggregates all traders in the\nmarket, rather than being a representative agent. Therefore, it can be\nidentified with a bounded-rational component of the market itself, providing a\nparticular implementation of an Invisible Hand market mechanism. In such\nsetting, market dynamics are modeled as a fictitious self-play of such\nbounded-rational market-agent in its adversarial stochastic environment. As\nrewards obtained by such self-playing market agent are not observed from market\ndata, we formulate and solve a simple model of such market dynamics based on a\nneuroscience-inspired Bounded Rational Information Theoretic Inverse\nReinforcement Learning (BRIT-IRL). This results in effective asset price\ndynamics with a non-linear mean reversion - which in our model is generated\ndynamically, rather than being postulated. We argue that our model can be used\nin a similar way to the Black-Litterman model. In particular, it represents, in\na simple modeling framework, market views of common predictive signals, market\nimpacts and implied optimal dynamic portfolio allocations, and can be used to\nassess values of private signals. Moreover, it allows one to quantify a\n\"market-implied\" optimal investment strategy, along with a measure of market\nrationality. Our approach is numerically light, and can be implemented using\nstandard off-the-shelf software such as TensorFlow.","url_abs":"http://arxiv.org/abs/1805.06126v1","url_pdf":"http://arxiv.org/pdf/1805.06126v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"market-self-learning-of-signals-impact-and","repo_url":"https://github.com/harshit0511/Inv-RL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"self-learning","task_name":"Self-Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}