Minimal offline RL scaffold featuring a DQN baseline on Peg Solitaire (7x7 and 4x4 variants), deterministic seeding, YAML config, JSONL metrics, and plotting utilities. Designed for reproducible ...
This repository contains the code for the paper: Conditional Abstraction Trees for Sample-efficient Reinforcement Learning. Mehdi Dadvar, Rashmeet Kaur Nayyar, and Siddharth Srivastava. 39th ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback