Miroslav Olšák
Miroslav Olšák has 4 papers in our library, in 2019–2025, most of them on Reasoning; often with Josef Urban, Bartosz Piotrowski, Cezary Kaliszyk.
OpenAlex · ORCID · h-index 8 · 37 works · 217 citations OpenAlex, 2026-10-14
📄 4 works in our library · 🔗 Cited 54 times in our library · ⭐ Paper Stars ranking: #7926
Affiliations
🎓University of CambridgeCambridge ·
United Kingdom2023–2025 · 3 papersEmployment 2023–2025🎓Charles UniversityPrague ·
Czechia2017–2023 · 11 papers🎓Universität InnsbruckInnsbruck ·
Austria2016–2023 · 9 papersEmployment 2016–2019Employment 2019–2021🔬Institut des Hautes Études ScientifiquesBures-sur-Yvette ·
France2021–2024 · 4 papersEmployment 2021–2023
Fields, topics and concepts
📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 4 worksArtificial Intelligence · 4 worksComputational Theory and Mathematics · 4 worksMathematics · 1 workGeometry and Topology · 1 work
🧭 Topics 1Research subjects and languages the work is about.Reasoning · 2 works
Works by year
2020s
2010s
Papers
- ENIGMA Anonymous: Symbol-Independent Inference Guiding Machine (system description) (2020 · 42 citations)
Learned guidance for saturation-style provers usually relies on symbol names. Such guidance cannot carry over to problems or libraries that use different names. This system description adds two name-agnostic clause classifiers to the ENIGMA framework of the E prover. Under either classifier, consistently renaming all… - Gold-medalist Performance in Solving Olympiad Geometry with AlphaGeometry2 (2025 · 31 citations)
AlphaGeometry2 (AG2) is Google DeepMind's successor to AlphaGeometry (AG1), a neuro-symbolic system for Euclidean geometry problems at International Mathematical Olympiad level. - Property Invariant Embedding for Automated Reasoning (2019 · 30 citations)
Machine-learned guidance for theorem provers depends on turning formulas into vectors. Most existing encoders, including the graph-based FormulaNet, learn a separate representation for each symbol name. They therefore cope poorly with the many fresh symbols that Skolemization and definitional clausification produce… - Graph2Tac: Online Representation Learning of Formal Math Concepts (2024 · 19 citations)
The paper tackles a practical gap in machine-learning provers for Coq: models trained offline cannot easily handle the new definitions, lemmas and proofs that appear when working in an unfamiliar formalization project.
Co-authors here
- Josef Urban ×2
- Bartosz Piotrowski
- Cezary Kaliszyk
- Jan Jakubův
- Jason Rute
- Junsu Kim
- Karel Chvalovský
- Martin Suda
- Quoc Viet Le
- Thang Luong
- Trieu H. Trinh
- Yuri Chervonyi
Venues
- Frontiers in artificial intelligence and applications: Frontiers in artificial intelligence and applications 2019
- Lecture notes in computer science: Lecture notes in computer science 2020
Cited by
Updated
This page in Markdown → https://md.research.elamur.ai/free/en/authors/miroslav-olsak/