Scholar
Seth Karten
Google Scholar ID: gzyxNfkAAAAJ
Princeton University
Citations & Impact
Citations
89
H-index
6
i10-index
5
Publications
13
Co-authors
16
list available
Publications
Resume (English only)
Academic Achievements
- - Recipient of the NSF Graduate Research Fellowship (2023)
- - Recipient of Princeton's Francis Robbins Upton Fellowship
- - Selected Publications:
- * Generative Simulacra: LLM Economist: Large Population Models and Mechanism Design in Multi-Agent Generative Simulacra
- * The PokeAgent Challenge: Competitive and Long Context Learning at Scale
- * PokéChamp: An Expert-level Minimax Language Agent
- * FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning
- * On the Role of Emergent Communication for Social Learning in Multi-Agent Reinforcement Learning
- * Towards True Lossless Sparse Communication in Multi-Agent Systems
- * Interpretable Learned Emergent Communication for Human-Agent Teams
Research Experience
- - Studied multi-agent pathfinding (MAPF) at Amazon
- - Developed multi-agent world models at Waymo
- - Studied emergent communication and decision-making in multi-agent teams
- - Studied learning hierarchical control primitives
Education
- - PhD Candidate in Computer Science, Princeton University, advised by Chi Jin
- - MS in Robotics, Carnegie Mellon University, advised by Katia Sycara
- - BS in Computer Science and Mathematics, Rutgers University, New Brunswick, received the Hagerty Award (2019), supervised by Kostas Bekris
Background
- Research interests: Developing foundation agents that can perform closed-loop decision-making in open-ended environments at scale; Research focus: Studying agents in complex environments to test fundamentals while enabling rich applications in embodied agents and economic systems.
Miscellany
- Personal interests and other information not detailed.
