{"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/using-deep-q-learning-to-understand-the-tax","title":"Using deep Q-learning to understand the tax evasion behavior of risk-averse firms","arxiv_id":"1801.09466","date":"2018-01-29","proceeding":null,"authors":["Nikolaos D. Goumagias","Dimitrios Hristu-Varsakelis","Yannis M. Assael"],"abstract":"Designing tax policies that are effective in curbing tax evasion and maximize\nstate revenues requires a rigorous understanding of taxpayer behavior. This\nwork explores the problem of determining the strategy a self-interested,\nrisk-averse tax entity is expected to follow, as it \"navigates\" - in the\ncontext of a Markov Decision Process - a government-controlled tax environment\nthat includes random audits, penalties and occasional tax amnesties. Although\nsimplified versions of this problem have been previously explored, the mere\nassumption of risk-aversion (as opposed to risk-neutrality) raises the\ncomplexity of finding the optimal policy well beyond the reach of analytical\ntechniques. Here, we obtain approximate solutions via a combination of\nQ-learning and recent advances in Deep Reinforcement Learning. By doing so, we\ni) determine the tax evasion behavior expected of the taxpayer entity, ii)\ncalculate the degree of risk aversion of the \"average\" entity given empirical\nestimates of tax evasion, and iii) evaluate sample tax policies, in terms of\nexpected revenues. Our model can be useful as a testbed for \"in-vitro\" testing\nof tax policies, while our results lead to various policy recommendations.","url_abs":"http://arxiv.org/abs/1801.09466v1","url_pdf":"http://arxiv.org/pdf/1801.09466v1.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":"using-deep-q-learning-to-understand-the-tax","repo_url":"https://github.com/iassael/tax-evasion-dqn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"using-deep-q-learning-to-understand-the-tax","repo_url":"https://github.com/iassael/tax-evasion-torch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"q-learning","task_name":"Q-Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}