Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control

• URL: http://arxiv.org/abs/2405.16158v1
• Date: Sat, 25 May 2024 09:53:25 GMT
• Title: Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control
• Authors: Michal Nauman, Mateusz Ostaszewski, Krzysztof Jankowski, Piotr Miłoś, Marek Cygan,
• Abstract summary: BRO is a model-free algorithm to achieve near-optimal policies in the Dog and Humanoid tasks. BRO achieves state-of-the-art results, significantly outperforming the leading model-based and model-free algorithms. BRO is the first model-free algorithm to achieve near-optimal policies in the notoriously challenging Dog and Humanoid tasks.
• Score: 1.1404490220482764
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Sample efficiency in Reinforcement Learning (RL) has traditionally been driven by algorithmic enhancements. In this work, we demonstrate that scaling can also lead to substantial improvements. We conduct a thorough investigation into the interplay of scaling model capacity and domain-specific RL enhancements. These empirical findings inform the design choices underlying our proposed BRO (Bigger, Regularized, Optimistic) algorithm. The key innovation behind BRO is that strong regularization allows for effective scaling of the critic networks, which, paired with optimistic exploration, leads to superior performance. BRO achieves state-of-the-art results, significantly outperforming the leading model-based and model-free algorithms across 40 complex tasks from the DeepMind Control, MetaWorld, and MyoSuite benchmarks. BRO is the first model-free algorithm to achieve near-optimal policies in the notoriously challenging Dog and Humanoid tasks.

Related papers

• Monte Carlo Tree Search Boosts Reasoning via Iterative Preference Learning [55.96599486604344]
  We introduce an approach aimed at enhancing the reasoning capabilities of Large Language Models (LLMs) through an iterative preference learning process. We use Monte Carlo Tree Search (MCTS) to iteratively collect preference data, utilizing its look-ahead ability to break down instance-level rewards into more granular step-level signals. The proposed algorithm employs Direct Preference Optimization (DPO) to update the LLM policy using this newly generated step-level preference data.
  arXiv  Detail & Related papers  (2024-05-01T11:10:24Z)
• Reinforced In-Context Black-Box Optimization [64.25546325063272]
  RIBBO is a method to reinforce-learn a BBO algorithm from offline data in an end-to-end fashion. RIBBO employs expressive sequence models to learn the optimization histories produced by multiple behavior algorithms and tasks. Central to our method is to augment the optimization histories with textitregret-to-go tokens, which are designed to represent the performance of an algorithm based on cumulative regret over the future part of the histories.
  arXiv  Detail & Related papers  (2024-02-27T11:32:14Z)
• CARE: Confidence-rich Autonomous Robot Exploration using Bayesian Kernel Inference and Optimization [12.32946442160165]
  We consider improving the efficiency of information-based autonomous robot exploration in unknown and complex environments. We propose a novel lightweight information gain inference method based on Bayesian kernel inference and optimization (BKIO) We show the desired efficiency of our proposed methods without losing exploration performance in different unstructured, cluttered environments.
  arXiv  Detail & Related papers  (2023-09-11T02:30:06Z)
• 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)
• Improved Algorithms for Neural Active Learning [74.89097665112621]
  We improve the theoretical and empirical performance of neural-network(NN)-based active learning algorithms for the non-parametric streaming setting. We introduce two regret metrics by minimizing the population loss that are more suitable in active learning than the one used in state-of-the-art (SOTA) related work.
  arXiv  Detail & Related papers  (2022-10-02T05:03:38Z)
• Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels [112.63440666617494]
  Reinforcement learning algorithms can succeed but require large amounts of interactions between the agent and the environment. We propose a new method to solve it, using unsupervised model-based RL, for pre-training the agent. We show robust performance on the Real-Word RL benchmark, hinting at resiliency to environment perturbations during adaptation.
  arXiv  Detail & Related papers  (2022-09-24T14:22:29Z)
• PC-MLP: Model-based Reinforcement Learning with Policy Cover Guided Exploration [15.173628100049129]
  This work studies a model-based algorithm for both Kernelized Regulators (KNR) and linear Markov Decision Processes (MDPs) For both models, our algorithm guarantees sample complexity and only uses access to a planning oracle. Our method can also perform reward-free exploration efficiently.
  arXiv  Detail & Related papers  (2021-07-15T15:49:30Z)
• Stabilizing Deep Q-Learning with ConvNets and Vision Transformers under Data Augmentation [25.493902939111265]
  We investigate causes of instability when using data augmentation in off-policy Reinforcement Learning algorithms. We propose a simple yet effective technique for stabilizing this class of algorithms under augmentation. Our method greatly improves stability and sample efficiency of ConvNets under augmentation, and achieves generalization results competitive with state-of-the-art methods for image-based RL.
  arXiv  Detail & Related papers  (2021-07-01T17:58:05Z)
• Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planning [93.1435980666675]
  We show how optimistic exploration can be easily combined with state-of-the-art reinforcement learning algorithms. Our experiments demonstrate that optimistic exploration significantly speeds-up learning when there are penalties on actions.
  arXiv  Detail & Related papers  (2020-06-15T18:37:38Z)

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.