Video Prediction Models as Rewards for Reinforcement Learning

• URL: http://arxiv.org/abs/2305.14343v2
• Date: Tue, 30 May 2023 17:38:44 GMT
• Title: Video Prediction Models as Rewards for Reinforcement Learning
• Authors: Alejandro Escontrela and Ademi Adeniji and Wilson Yan and Ajay Jain and Xue Bin Peng and Ken Goldberg and Youngwoon Lee and Danijar Hafner and Pieter Abbeel
• Abstract summary: VIPER is an algorithm that leverages pretrained video prediction models as action-free reward signals for reinforcement learning. We see our work as starting point for scalable reward specification from unlabeled videos.
• Score: 127.53893027811027
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Specifying reward signals that allow agents to learn complex behaviors is a long-standing challenge in reinforcement learning. A promising approach is to extract preferences for behaviors from unlabeled videos, which are widely available on the internet. We present Video Prediction Rewards (VIPER), an algorithm that leverages pretrained video prediction models as action-free reward signals for reinforcement learning. Specifically, we first train an autoregressive transformer on expert videos and then use the video prediction likelihoods as reward signals for a reinforcement learning agent. VIPER enables expert-level control without programmatic task rewards across a wide range of DMC, Atari, and RLBench tasks. Moreover, generalization of the video prediction model allows us to derive rewards for an out-of-distribution environment where no expert data is available, enabling cross-embodiment generalization for tabletop manipulation. We see our work as starting point for scalable reward specification from unlabeled videos that will benefit from the rapid advances in generative modeling. Source code and datasets are available on the project website: https://escontrela.me/viper

Related papers

• Pre-trained Visual Dynamics Representations for Efficient Policy Learning [33.62440075940917]
  We propose Pre-trained Visual Dynamics Representations (PVDR) to bridge the domain gap between videos and downstream tasks for efficient policy learning. The pre-trained visual dynamics representations capture the visual dynamics prior knowledge in the videos. This abstract prior knowledge can be readily adapted to downstream tasks and aligned with executable actions through online adaptation.
  arXiv  Detail & Related papers  (2024-11-05T15:18:02Z)
• Predicting Long-horizon Futures by Conditioning on Geometry and Time [49.86180975196375]
  We explore the task of generating future sensor observations conditioned on the past. We leverage the large-scale pretraining of image diffusion models which can handle multi-modality. We create a benchmark for video prediction on a diverse set of videos spanning indoor and outdoor scenes.
  arXiv  Detail & Related papers  (2024-04-17T16:56:31Z)
• Video Pre-trained Transformer: A Multimodal Mixture of Pre-trained Experts [2.457872341625575]
  We present Video Pre-trained Transformer. It uses four SOTA encoder models to convert a video into a sequence of compact embeddings. It learns using an autoregressive causal language modeling loss by predicting the words spoken in YouTube videos.
  arXiv  Detail & Related papers  (2023-03-24T17:18:40Z)
• Reinforcement Learning with Action-Free Pre-Training from Videos [95.25074614579646]
  We introduce a framework that learns representations useful for understanding the dynamics via generative pre-training on videos. Our framework significantly improves both final performances and sample-efficiency of vision-based reinforcement learning.
  arXiv  Detail & Related papers  (2022-03-25T19:44:09Z)
• Reward prediction for representation learning and reward shaping [0.8883733362171032]
  We propose learning a state representation in a self-supervised manner for reward prediction. We augment the training of out-of-the-box RL agents by shaping the reward using our reward predictor during policy learning.
  arXiv  Detail & Related papers  (2021-05-07T11:29:32Z)
• Generative Adversarial Reward Learning for Generalized Behavior Tendency Inference [71.11416263370823]
  We propose a generative inverse reinforcement learning for user behavioral preference modelling. Our model can automatically learn the rewards from user's actions based on discriminative actor-critic network and Wasserstein GAN.
  arXiv  Detail & Related papers  (2021-05-03T13:14:25Z)
• Learning from Weakly-labeled Web Videos via Exploring Sub-Concepts [89.06560404218028]
  We introduce a new method for pre-training video action recognition models using queried web videos. Instead of trying to filter out, we propose to convert the potential noises in these queried videos to useful supervision signals. We show that SPL outperforms several existing pre-training strategies using pseudo-labels.
  arXiv  Detail & Related papers  (2021-01-11T05:50:16Z)
• Semi-supervised reward learning for offline reinforcement learning [71.6909757718301]
  Training agents usually requires reward functions, but rewards are seldom available in practice and their engineering is challenging and laborious. We propose semi-supervised learning algorithms that learn from limited annotations and incorporate unlabelled data. In our experiments with a simulated robotic arm, we greatly improve upon behavioural cloning and closely approach the performance achieved with ground truth rewards.
  arXiv  Detail & Related papers  (2020-12-12T20:06:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.