Sungtae Lee — accepted AI-conference papers

2024

Unsupervised Object Interaction Learning with Counterfactual Dynamics Models

Jongwook Choi, Sungtae Lee, Xinyu Wang, Sungryull Sohn, Honglak Lee

AAAI 2024technical

We present COIL (Counterfactual Object Interaction Learning), a novel way of learning skills of object interactions on entity-centric environments. The goal is to learn primitive behaviors that can induce interactions without external reward or any supervision. Existing skill discovery methods are l…

Cited by 6 Source ↗PDF ↗Scholar ↗

2021

Shortest-Path Constrained Reinforcement Learning for Sparse Reward Tasks

Sungryull Sohn, Sungtae Lee, Jongwook Choi, Harm H Van Seijen, Mehdi Fatemi, Honglak Lee

ICML 2021spotlight

We propose the k-Shortest-Path (k-SP) constraint: a novel constraint on the agent’s trajectory that improves the sample efficiency in sparse-reward MDPs. We show that any optimal policy necessarily satisfies the k-SP constraint. Notably, the k-SP constraint prevents the policy from exploring state-a…

Cited by 8 Source ↗PDF ↗code ↗Scholar ↗