Learning Collusion in Episodic, Inventory-Constrained Markets
Paul Friedrich, Barna Pasztor, Giorgia Ramponi. 2024. Learning Collusion in Episodic, Inventory-Constrained Markets. ArXiv.
CV
Education
- Ph.D in Machine Learning, ETH AI Center (ETH Zurich), 2025 (expected)
- Topic: Multi-agent Learning in General-sum Games
- Supervisors: Prof. Dr. Andreas Krause, Prof. Dr. Sven Seuken
- M.Sc. in Data Science, ETH Zurich, 2020
- B.Sc. in Mathematics with Economics, University College London, 2018
Work experience
- Data Scientist (2020 October - 2022 August)
- QuantumBlack, AI by McKinsey & Company (Zürich, Switzerland)
- Industries: Automotive, Telecommunication, Mining, Pharmaceutical, Retail
- Functions: Pricing, Marketing, Due Diligence, Forecasting
- Research Assistant (2019 February - 2020 February)
- Social Networks lab, ETH Zurich (Zürich, Switzerland)
- Supervisor: Prof. Dr. Ulrik Brandes
- Research topic: Graph Embedding & Visualization
Publications
Stochastic Bilevel Optimization with Lower-Level Contextual Markov Decision Processes
Vinzenz Thoma, Barna Pasztor, Andreas Krause, Giorgia Ramponi, Yifan Hu. 2024. Stochastic Bilevel Optimization with Lower-Level Contextual Markov Decision Processes. ArXiv.
Safe Model-Based Multi-Agent Mean-Field Reinforcement Learning
Matej Jusup, Barna Pásztor, Tadeusz Janik, Kenan Zhang, Francesco Corman, Andreas Krause, and Ilija Bogunovic. 2024. Safe Model-Based Multi-Agent Mean-Field Reinforcement Learning. In Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS '24). International Foundation for Autonomous Agents and Multiagent Systems, Richland, SC, 973–982.
Efficient Model-Based Multi-Agent Mean-Field Reinforcement Learning
Barna Pásztor, Andreas Krause, Ilija Bogunovic (2023). "Efficient Model-Based Multi-Agent Mean-Field Reinforcement Learning." Transactions on Machine Learning Research.
On the impact of publicly available news and information transfer to financial markets
Metod Jazbec, Barna Pásztor, Felix Faltings, Nino Antulov-Fantulin, Petter N Kolm (2021). "On the impact of publicly available news and information transfer to financial markets." Royal Society Open Science.
Stochastic Gradient Descent Works Really Well for Stress Minimization
Katharina Börsig, Ulrik Brandes, Barna Pásztor (2021). "Stochastic Gradient Descent Works Really Well for Stress Minimization." Graph Drawing and Network Visualization.
Academic Achievements
- Finalist, MeltingPot competition (NeurIPS 2023)
- Top 6 out of 117 teams
- Developed a Multi-Agent RL system to learn in challenging social dilemma environments
- Techniques and frameworks used: IMPALA, A2C, PopArt, reward shaping, curriculum learning, RLLib
- 2nd place, Computational Social Choice Competition (IJCAI 2023)
- Proposed a provably optimal and efficient voting rule for best-k candidate voting problem
Talks
Efficient Model-Based Multi-Agent Mean-Field Reinforcement Learning
Webinar at Machine Learning and Mean-Field Games Seminar, Online