We study how visual representations pre-trained on diverse human video data can enable
data-efficient learning of downstream robotic manipulation tasks. Concretely, we pre-train a …

Large, high-capacity models trained on diverse datasets have shown remarkable successes
on efficiently tackling downstream applications. In domains from NLP to Computer Vision …

In reinforcement learning (RL), it is easier to solve a task if given a good representation.
While deep RL should automatically acquire such good representations, prior work often …

Recent unsupervised pre-training methods have shown to be effective on language and
vision domains by learning useful representations for multiple downstream tasks. In this …

Imitation learning from human demonstrations is a promising paradigm for teaching robots
manipulation skills in the real world. However, learning complex long-horizon tasks often …

We are motivated by the goal of generalist robots that can complete a wide range of tasks
across many environments. Critical to this is the robot's ability to acquire some metric of task …

Reinforcement learning is a powerful framework for robots to acquire skills from experience,
but often requires a substantial amount of online data collection. As a result, it is difficult to …

Unsupervised pre-training methods utilizing large and diverse datasets have achieved
tremendous success across a range of domains. Recent work has investigated such …

Natural agents can effectively learn from multiple data sources that differ in size, quality, and
types of measurements. We study this heterogeneity in the context of offline reinforcement …

While current vision algorithms excel at many challenging tasks, it is unclear how well they
understand the physical dynamics of real-world environments. Here we introduce Physion, a …