Transferable Reinforcement Learning via Generalized Occupancy Models
- URL: http://arxiv.org/abs/2403.06328v2
- Date: Tue, 28 May 2024 23:42:28 GMT
- Title: Transferable Reinforcement Learning via Generalized Occupancy Models
- Authors: Chuning Zhu, Xinqi Wang, Tyler Han, Simon S. Du, Abhishek Gupta,
- Abstract summary: Generalized occupancy models (GOMs) learn a distribution of successor features from a stationary dataset.
GOMs avoid compounding error while enabling rapid transfer across reward functions.
- Score: 36.53356539916603
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Intelligent agents must be generalists, capable of quickly adapting to various tasks. In reinforcement learning (RL), model-based RL learns a dynamics model of the world, in principle enabling transfer to arbitrary reward functions through planning. However, autoregressive model rollouts suffer from compounding error, making model-based RL ineffective for long-horizon problems. Successor features offer an alternative by modeling a policy's long-term state occupancy, reducing policy evaluation under new tasks to linear reward regression. Yet, policy improvement with successor features can be challenging. This work proposes a novel class of models, i.e., generalized occupancy models (GOMs), that learn a distribution of successor features from a stationary dataset, along with a policy that acts to realize different successor features. These models can quickly select the optimal action for arbitrary new tasks. By directly modeling long-term outcomes in the dataset, GOMs avoid compounding error while enabling rapid transfer across reward functions. We present a practical instantiation of GOMs using diffusion models and show their efficacy as a new class of transferable models, both theoretically and empirically across various simulated robotics problems. Videos and code at https://weirdlabuw.github.io/gom/.
Related papers
- AdaMerging: Adaptive Model Merging for Multi-Task Learning [68.75885518081357]
This paper introduces an innovative technique called Adaptive Model Merging (AdaMerging)
It aims to autonomously learn the coefficients for model merging, either in a task-wise or layer-wise manner, without relying on the original training data.
Compared to the current state-of-the-art task arithmetic merging scheme, AdaMerging showcases a remarkable 11% improvement in performance.
arXiv Detail & Related papers (2023-10-04T04:26:33Z) - Model-Based Reinforcement Learning with Multi-Task Offline Pretraining [59.82457030180094]
We present a model-based RL method that learns to transfer potentially useful dynamics and action demonstrations from offline data to a novel task.
The main idea is to use the world models not only as simulators for behavior learning but also as tools to measure the task relevance.
We demonstrate the advantages of our approach compared with the state-of-the-art methods in Meta-World and DeepMind Control Suite.
arXiv Detail & Related papers (2023-06-06T02:24:41Z) - Self-Supervised Reinforcement Learning that Transfers using Random
Features [41.00256493388967]
We propose a self-supervised reinforcement learning method that enables the transfer of behaviors across tasks with different rewards.
Our method is self-supervised in that it can be trained on offline datasets without reward labels, but can then be quickly deployed on new tasks.
arXiv Detail & Related papers (2023-05-26T20:37:06Z) - TOM: Learning Policy-Aware Models for Model-Based Reinforcement Learning
via Transition Occupancy Matching [28.743727234246126]
We propose a new "transition occupancy matching" (TOM) objective for model learning.
TOM is good to the extent that the current policy experiences the same distribution of transitions inside the model as in the real environment.
We show that TOM successfully focuses model learning on policy-relevant experience and drives policies faster to higher task rewards.
arXiv Detail & Related papers (2023-05-22T03:06:09Z) - Contrastive Value Learning: Implicit Models for Simple Offline RL [40.95632543012637]
We propose Contrastive Value Learning (CVL), which learns an implicit, multi-step model of the environment dynamics.
CVL can be learned without access to reward functions, but nonetheless can be used to directly estimate the value of each action.
Our experiments demonstrate that CVL outperforms prior offline RL methods on complex continuous control benchmarks.
arXiv Detail & Related papers (2022-11-03T19:10:05Z) - When to Update Your Model: Constrained Model-based Reinforcement
Learning [50.74369835934703]
We propose a novel and general theoretical scheme for a non-decreasing performance guarantee of model-based RL (MBRL)
Our follow-up derived bounds reveal the relationship between model shifts and performance improvement.
A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns.
arXiv Detail & Related papers (2022-10-15T17:57:43Z) - Fully Decentralized Model-based Policy Optimization for Networked
Systems [23.46407780093797]
This work aims to improve data efficiency of multi-agent control by model-based learning.
We consider networked systems where agents are cooperative and communicate only locally with their neighbors.
In our method, each agent learns a dynamic model to predict future states and broadcast their predictions by communication, and then the policies are trained under the model rollouts.
arXiv Detail & Related papers (2022-07-13T23:52:14Z) - Model Reprogramming: Resource-Efficient Cross-Domain Machine Learning [65.268245109828]
In data-rich domains such as vision, language, and speech, deep learning prevails to deliver high-performance task-specific models.
Deep learning in resource-limited domains still faces multiple challenges including (i) limited data, (ii) constrained model development cost, and (iii) lack of adequate pre-trained models for effective finetuning.
Model reprogramming enables resource-efficient cross-domain machine learning by repurposing a well-developed pre-trained model from a source domain to solve tasks in a target domain without model finetuning.
arXiv Detail & Related papers (2022-02-22T02:33:54Z) - Can Wikipedia Help Offline Reinforcement Learning? [12.12541097531412]
Fine-tuning reinforcement learning models has been challenging because of a lack of large scale off-the-shelf datasets.
Recent work has looked at tackling offline RL with improved results as result of the introduction of the Transformer architecture.
We investigate the transferability of pre-trained sequence models on other domains (vision, language) when finetuned on offline RL tasks.
arXiv Detail & Related papers (2022-01-28T13:55:35Z) - Evaluating model-based planning and planner amortization for continuous
control [79.49319308600228]
We take a hybrid approach, combining model predictive control (MPC) with a learned model and model-free policy learning.
We find that well-tuned model-free agents are strong baselines even for high DoF control problems.
We show that it is possible to distil a model-based planner into a policy that amortizes the planning without any loss of performance.
arXiv Detail & Related papers (2021-10-07T12:00:40Z) - Goal-Aware Prediction: Learning to Model What Matters [105.43098326577434]
One of the fundamental challenges in using a learned forward dynamics model is the mismatch between the objective of the learned model and that of the downstream planner or policy.
We propose to direct prediction towards task relevant information, enabling the model to be aware of the current task and encouraging it to only model relevant quantities of the state space.
We find that our method more effectively models the relevant parts of the scene conditioned on the goal, and as a result outperforms standard task-agnostic dynamics models and model-free reinforcement learning.
arXiv Detail & Related papers (2020-07-14T16:42:59Z)
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.