Saved in:
Bibliographic Details
Main Authors: Hansen, Nicklas, Su, Hao, Wang, Xiaolong
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.16828
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914721856225280
author Hansen, Nicklas
Su, Hao
Wang, Xiaolong
author_facet Hansen, Nicklas
Su, Hao
Wang, Xiaolong
contents TD-MPC is a model-based reinforcement learning (RL) algorithm that performs local trajectory optimization in the latent space of a learned implicit (decoder-free) world model. In this work, we present TD-MPC2: a series of improvements upon the TD-MPC algorithm. We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perform 80 tasks across multiple task domains, embodiments, and action spaces. We conclude with an account of lessons, opportunities, and risks associated with large TD-MPC2 agents. Explore videos, models, data, code, and more at https://tdmpc2.com
format Preprint
id arxiv_https___arxiv_org_abs_2310_16828
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TD-MPC2: Scalable, Robust World Models for Continuous Control
Hansen, Nicklas
Su, Hao
Wang, Xiaolong
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Robotics
TD-MPC is a model-based reinforcement learning (RL) algorithm that performs local trajectory optimization in the latent space of a learned implicit (decoder-free) world model. In this work, we present TD-MPC2: a series of improvements upon the TD-MPC algorithm. We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perform 80 tasks across multiple task domains, embodiments, and action spaces. We conclude with an account of lessons, opportunities, and risks associated with large TD-MPC2 agents. Explore videos, models, data, code, and more at https://tdmpc2.com
title TD-MPC2: Scalable, Robust World Models for Continuous Control
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2310.16828