Image of a human arm shaking hands with a robotic arm representing Artificial intelligence research at Microsoft
Microsoft Research Lab – Redmond

Robot Learning Group

Downloads

January 2024

LLF Bench is a benchmark for evaluating learning agents that provides a diverse collection of interactive learning problems where the agent gets language feedback instead of rewards (as in RL) or action feedback (as in imitation learning).

Github for LLF-Bench

November 2022

The MoCapAct dataset contains training data and models for humanoid locomotion research. It consists of expert policies that are trained to track individual clip snippets and HDF5 files of noisy rollouts collected from each expert, including proprioceptive observations and actions.…

Download for Motion Capture with Actions (MoCapAct)