{"ID":2845927,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.14778","arxiv_id":"2511.14778","title":"Learning Interestingness in Automated Mathematical Theory Formation","abstract":"We take two key steps in automating the open-ended discovery of new mathematical theories, a grand challenge in artificial intelligence. First, we introduce $\\emph{FERMAT}$, a reinforcement learning (RL) environment that models concept discovery and theorem-proving using a set of symbolic actions, opening up a range of RL problems relevant to theory discovery. Second, we explore a specific problem through $\\emph{FERMAT}$: automatically scoring the $\\emph{interestingness}$ of mathematical objects. We investigate evolutionary algorithms for synthesizing nontrivial interestingness measures. In particular, we introduce an LLM-based evolutionary algorithm that features function abstraction, leading to notable improvements in discovering elementary number theory and finite fields over hard-coded baselines. We open-source the $\\emph{FERMAT}$ environment at this URL(https://github.com/trishullab/Fermat).","short_abstract":"We take two key steps in automating the open-ended discovery of new mathematical theories, a grand challenge in artificial intelligence. First, we introduce $\\emph{FERMAT}$, a reinforcement learning (RL) environment that models concept discovery and theorem-proving using a set of symbolic actions, opening up a range of...","url_abs":"https://arxiv.org/abs/2511.14778","url_pdf":"https://arxiv.org/pdf/2511.14778v1","authors":"[\"George Tsoukalas\",\"Rahul Saha\",\"Amitayush Thakur\",\"Sabrina Reguyal\",\"Swarat Chaudhuri\"]","published":"2025-11-05T18:59:17Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\"]","has_code":false,"code_links":[{"ID":607394,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2845927,"paper_url":"https://arxiv.org/abs/2511.14778","paper_title":"Learning Interestingness in Automated Mathematical Theory Formation","repo_url":"https://github.com/trishullab/Fermat","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}