Search papers, labs, and topics across Lattice.
University of Oxford
4
0
7
Other-Play's performance remains stable across varied implementations, challenging the notion that single-seed evaluations capture the full picture of zero-shot coordination robustness.
Meta-optimizing an AI data scientist can dramatically enhance the quality of synthetic datasets, outperforming traditional methods.
Hierarchical policies aren't just for horizon reduction; they can unlock powerful state-space abstraction for dramatic gains in offline goal-conditioned RL.
RSPG solves partially observable mean field games an order of magnitude faster than existing methods, unlocking the ability to model complex macroeconomic systems with heterogeneous agents and common noise.
