distracting-control
visual_distracting_control_suite
Visual Distracting Control Suite Benchmark
This dataset contains expert trajectories generated by a Proximal Policy Optimization (PPO) reinforcement learning agent trained on 4 environments of the Distracting Control Suite. For each environment we collect data with different levels of distraction, which we define below, and masks for the agent.
Levels of distraction:
None: Vanilla DeepMind Control Suite without visual distractions. The environment uses the default static background… See the full description on the dataset page: https://huggingface.co/datasets/hamza-adnan/visual_distracting_control_suite.visual_distracting_control_suite
Visual Distracting Control Suite with Agent Masks
Visual Distracting Control Suite with Agent Masks contains expert trajectories
for four continuous-control tasks from
DeepMind Control Suite. It adds
the visual variations of the
Distracting Control Suite
and provides two agent masks for every RGB observation: a ground-truth mask from
the simulator and a mask predicted with
SAM 2.1.
The dataset is designed for offline reinforcement learning, imitation learning,
latent-action… See the full description on the dataset page: https://huggingface.co/datasets/EpicPinkPenguin/visual_distracting_control_suite.