dataset
LongTailOfflineRL
AntonioAlgaida
or hover any field below to flag it
Overview
Name
LongTailOfflineRL
Source
AntonioAlgaida
Episodes
0
Robot count
0
Format
other
Description
Tackling the long-tail problem for AV motion planning with data-centric Offline RL. A comparative study on the Waymo dataset using criticality metrics (uncertainty, heuristics) to train robust, goal-conditioned, attention-based agents. [ArXiv: 2508.07029]
Robots used
null
Links
HuggingFace dataset
null