Robotic Offline RL from Internet Videos via Value-Function Pre-Training

TL;DR

This paper introduces V-PTR, a system that leverages large-scale human video datasets to pre-train value functions for robotic offline reinforcement learning, enabling better generalization and robustness in manipulation tasks.

Contribution

The paper presents a novel approach to use video data for pre-training value functions in robotic offline RL, bridging the gap between observation-only videos and RL requirements.

Findings

V-PTR improves policy performance on manipulation tasks.

Value functions learned from videos enhance downstream RL.

The system generalizes well across diverse tasks.

Abstract

Pre-training on Internet data has proven to be a key ingredient for broad generalization in many modern ML systems. What would it take to enable such capabilities in robotic reinforcement learning (RL)? Offline RL methods, which learn from datasets of robot experience, offer one way to leverage prior data into the robotic learning pipeline. However, these methods have a "type mismatch" with video data (such as Ego4D), the largest prior datasets available for robotics, since video offers observation-only experience without the action or reward annotations needed for RL methods. In this paper, we develop a system for leveraging large-scale human video datasets in robotic offline RL, based entirely on learning value functions via temporal-difference learning. We show that value learning on video datasets learns representations that are more conducive to downstream robotic offline RL than other approaches for learning from video data. Our system, called V-PTR, combines the benefits of pre-training on video data with robotic offline RL approaches that train on diverse robot data, resulting in value functions and policies for manipulation tasks that perform better, act robustly, and generalize broadly. On several manipulation tasks on a real WidowX robot, our framework produces policies that greatly improve over prior methods. Our video and additional details can be found at https://dibyaghosh.com/vptr/