REVIEW 8 cited by
Pre-Training for Robots: Offline RL Enables Learning New Tasks from a Handful of Trials
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Progress in deep learning highlights the tremendous potential of utilizing diverse robotic datasets for attaining effective generalization and makes it enticing to consider leveraging broad datasets for attaining robust generalization in robotic learning as well. However, in practice, we often want to learn a new skill in a new environment that is unlikely to be contained in the prior data. Therefore we ask: how can we leverage existing diverse offline datasets in combination with small amounts of task-specific data to solve new tasks, while still enjoying the generalization benefits of training on large amounts of data? In this paper, we demonstrate that end-to-end offline RL can be an effective approach for doing this, without the need for any representation learning or vision-based pre-training. We present pre-training for robots (PTR), a framework based on offline RL that attempts to effectively learn new tasks by combining pre-training on existing robotic datasets with rapid fine-tuning on a new task, with as few as 10 demonstrations. PTR utilizes an existing offline RL method, conservative Q-learning (CQL), but extends it to include several crucial design decisions that enable PTR to actually work and outperform a variety of prior methods. To our knowledge, PTR is the first RL method that succeeds at learning new tasks in a new domain on a real WidowX robot with as few as 10 task demonstrations, by effectively leveraging an existing dataset of diverse multi-task robot data collected in a variety of toy kitchens. We also demonstrate that PTR can enable effective autonomous fine-tuning and improvement in a handful of trials, without needing any demonstrations. An accompanying overview video can be found in the supplementary material and at thi URL: https://sites.google.com/view/ptr-final/
Forward citations
Cited by 8 Pith papers
-
Regularized Offline Policy Optimization with Posterior Hybrid Bayesian Belief
PhyB averages over the k worst dynamics models with entropy-weighted coefficients and uses Bregman-regularized policy iteration; it claims bounded pessimism, monotonic improvement, and top D4RL scores.
-
Regularized Offline Policy Optimization with Posterior Hybrid Bayesian Belief
PhyB approximates Bayesian expectations in offline RL as convex combinations over dynamics model subsets with bounded discrepancy, enabling regularized policy optimization with monotonic improvement guarantees.
-
Target-Aligned Bellman Backup for Cross-domain Offline Reinforcement Learning
Target-Aligned Bellman Backup (TABB) improves cross-domain offline RL by selecting source transitions according to their contribution to accurate target-domain Bellman target estimation.
-
RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots
RoboCasa supplies a large-scale kitchen simulator, generative assets, 100 tasks, and automated data pipelines that produce a clear scaling trend in imitation learning for generalist robots.
-
Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulation
A GPT-style model pre-trained on large video datasets achieves 94.9% success on CALVIN multi-task manipulation and 85.4% zero-shot generalization, outperforming prior baselines.
-
VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action Adaptation
VGAS uses best-of-N selection with a geometrically grounded critic and explicit regularization to improve success rates of few-shot VLA policies under limited data and distribution shifts.
-
MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations
MimicGen creates over 50K robot demonstrations from roughly 200 human ones, allowing imitation learning to achieve strong performance on complex long-horizon tasks like assembly and coffee preparation.
-
TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation
TinyVLA achieves faster inference and higher data efficiency than OpenVLA on robotic manipulation tasks by initializing from high-speed multimodal models and adding a diffusion policy decoder, without any pre-training phase.
Discussion (0). Sign in to comment.