dataset
Snake-RL-Benchmarks
wILLIEWILLYWILLIe
or hover any field below to flag it
Overview
Name
Snake-RL-Benchmarks
Source
wILLIEWILLYWILLIe
Episodes
0
Robot count
0
Format
other
Description
A rigorous RL benchmark project on Snake. Features a comprehensive comparison of Online RL (Q-Learning, SARSA, DQN, PPO) vs. Offline Imitation Learning (Behavior Cloning) with detailed ablation studies on dataset sensitivity and reward shaping.
Robots used
null
Links
HuggingFace dataset
null