Vivek Myers

Scaling Goal-conditioned Reinforcement Learning with Multistep Quasimetric Distances

Training LLM Agents to Empower Humans

Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following

Accelerating Goal-Conditioned Reinforcement Learning Algorithms and Research

Offline Goal-conditioned Reinforcement Learning with Quasimetric Representations

Successor Representations Enable Emergent Compositional Instruction Following

Horizon Generalization in Reinforcement Learning