Active Fine-Tuning of Generalist Policies
- URL: http://arxiv.org/abs/2410.05026v1
- Date: Mon, 7 Oct 2024 13:26:36 GMT
- Title: Active Fine-Tuning of Generalist Policies
- Authors: Marco Bagatella, Jonas Hübotter, Georg Martius, Andreas Krause,
- Abstract summary: We propose AMF (Active Multi-task Fine-tuning) to maximize multi-task policy performance under a limited demonstration budget.
We derive performance guarantees for AMF under regularity assumptions and demonstrate its empirical effectiveness in complex and high-dimensional environments.
- Score: 54.65568433408307
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Pre-trained generalist policies are rapidly gaining relevance in robot learning due to their promise of fast adaptation to novel, in-domain tasks. This adaptation often relies on collecting new demonstrations for a specific task of interest and applying imitation learning algorithms, such as behavioral cloning. However, as soon as several tasks need to be learned, we must decide which tasks should be demonstrated and how often? We study this multi-task problem and explore an interactive framework in which the agent adaptively selects the tasks to be demonstrated. We propose AMF (Active Multi-task Fine-tuning), an algorithm to maximize multi-task policy performance under a limited demonstration budget by collecting demonstrations yielding the largest information gain on the expert policy. We derive performance guarantees for AMF under regularity assumptions and demonstrate its empirical effectiveness to efficiently fine-tune neural policies in complex and high-dimensional environments.
Related papers
- Policy Adaptation via Language Optimization: Decomposing Tasks for Few-Shot Imitation [49.43094200366251]
We propose a novel approach for few-shot adaptation to unseen tasks that exploits the semantic understanding of task decomposition.
Our method, Policy Adaptation via Language Optimization (PALO), combines a handful of demonstrations of a task with proposed language decompositions.
We find that PALO is able of consistently complete long-horizon, multi-tier tasks in the real world, outperforming state of the art pre-trained generalist policies.
arXiv Detail & Related papers (2024-08-29T03:03:35Z) - Meta-Reinforcement Learning Based on Self-Supervised Task Representation
Learning [23.45043290237396]
MoSS is a context-based Meta-reinforcement learning algorithm based on Self-Supervised task representation learning.
On MuJoCo and Meta-World benchmarks, MoSS outperforms prior in terms of performance, sample efficiency (3-50x faster), adaptation efficiency, and generalization.
arXiv Detail & Related papers (2023-04-29T15:46:19Z) - Learning Multi-Task Transferable Rewards via Variational Inverse
Reinforcement Learning [10.782043595405831]
We extend an empowerment-based regularization technique to situations with multiple tasks based on the framework of a generative adversarial network.
Under the multitask environments with unknown dynamics, we focus on learning a reward and policy from unlabeled expert examples.
Our proposed method derives the variational lower bound of the situational mutual information to optimize it.
arXiv Detail & Related papers (2022-06-19T22:32:41Z) - Leveraging convergence behavior to balance conflicting tasks in
multi-task learning [3.6212652499950138]
Multi-Task Learning uses correlated tasks to improve performance generalization.
Tasks often conflict with each other, which makes it challenging to define how the gradients of multiple tasks should be combined.
We propose a method that takes into account temporal behaviour of the gradients to create a dynamic bias that adjust the importance of each task during the backpropagation.
arXiv Detail & Related papers (2022-04-14T01:52:34Z) - Modular Adaptive Policy Selection for Multi-Task Imitation Learning
through Task Division [60.232542918414985]
Multi-task learning often suffers from negative transfer, sharing information that should be task-specific.
This is done by using proto-policies as modules to divide the tasks into simple sub-behaviours that can be shared.
We also demonstrate its ability to autonomously divide the tasks into both shared and task-specific sub-behaviours.
arXiv Detail & Related papers (2022-03-28T15:53:17Z) - Constructing a Good Behavior Basis for Transfer using Generalized Policy
Updates [63.58053355357644]
We study the problem of learning a good set of policies, so that when combined together, they can solve a wide variety of unseen reinforcement learning tasks.
We show theoretically that having access to a specific set of diverse policies, which we call a set of independent policies, can allow for instantaneously achieving high-level performance.
arXiv Detail & Related papers (2021-12-30T12:20:46Z) - Meta-Reinforcement Learning Robust to Distributional Shift via Model
Identification and Experience Relabeling [126.69933134648541]
We present a meta-reinforcement learning algorithm that is both efficient and extrapolates well when faced with out-of-distribution tasks at test time.
Our method is based on a simple insight: we recognize that dynamics models can be adapted efficiently and consistently with off-policy data.
arXiv Detail & Related papers (2020-06-12T13:34:46Z) - Learning Adaptive Exploration Strategies in Dynamic Environments Through
Informed Policy Regularization [100.72335252255989]
We study the problem of learning exploration-exploitation strategies that effectively adapt to dynamic environments.
We propose a novel algorithm that regularizes the training of an RNN-based policy using informed policies trained to maximize the reward in each task.
arXiv Detail & Related papers (2020-05-06T16:14:48Z)
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.