Matěj Kripner

Main Research Interests

reasoning in AI (especially in math), deep learning, reinforcement learning, automated theorem proving, Lean

Projects

Ph.D. topic: Advancing Reasoning Capabilities of Statistical Systems (2024 - now)

• advisors: RNDr. Milan Straka, Ph.D., Mgr. Martin Schmid, Ph.D.

Curriculum Vitae

• 2024-now Ph.D. in Computational Linguistics, ÚFAL MFF UK
• 2022-2024 Mgr. in Artificial Intelligence, MFF UK
• 2019-2022 Bc. in Computer Science (specialization: Theoretical Computer Science), MFF UK

Selected Bibliography

• Google Scholar
• ORCID: 0009-0006-9530-3670
• Researcher ID: LMN-4478-2024

1. Matěj Kripner (2025): Symbolic World Models in Lean 4 for Reinforcement Learning. In: Proceedings of the RLC 2025 Workshop on Programmatic Reinforcement Learning, pp. 8-18, Reinforcement Learning Journal (RLJ), Edmonton, Canada (url, local PDF, bibtex)
2. Matěj Kripner, Michal Šustr, Milan Straka (2025): LeanTree: Accelerating White-Box Proof Search with Factorized States in Lean 4. In: Proceedings of the Second AI for Math Workshop, pp. 67-79, International Machine Learning Society (IMLS), Geneva, Switzerland (url, local PDF, bibtex)
3. Matěj Kripner (2024): Self-Supervised Summarization via Reinforcement Learning (masters thesis). In: (bibtex)
4. David Herel, Dominika Zogatová, Matěj Kripner, Tomáš Mikolov (2022): Emergence of Novelty in Evolutionary Algorithms. In: Proceedings of the ALIFE 2022: The 2022 Conference on Artificial Life, pp. 146-154, MIT Press, Cambridge, MA, USA (pdf, bibtex)