Learning in complex action spaces without policy gradients

• URL: http://arxiv.org/abs/2410.06317v1
• Date: Tue, 8 Oct 2024 19:49:34 GMT
• Title: Learning in complex action spaces without policy gradients
• Authors: Arash Tavakoli, Sina Ghiassian, Nemanja Rakićević,
• Abstract summary: We show that QMLE can be applied to complex action spaces with a controllable computational cost that is comparable to that of policy gradient methods. QMLE demonstrates strong performance on the DeepMind Control Suite, even when compared to the state-of-the-art methods.
• Score: 8.81420331399616
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Conventional wisdom suggests that policy gradient methods are better suited to complex action spaces than action-value methods. However, foundational studies have shown equivalences between these paradigms in small and finite action spaces (O'Donoghue et al., 2017; Schulman et al., 2017a). This raises the question of why their computational applicability and performance diverge as the complexity of the action space increases. We hypothesize that the apparent superiority of policy gradients in such settings stems not from intrinsic qualities of the paradigm, but from universal principles that can also be applied to action-value methods to serve similar functionality. We identify three such principles and provide a framework for incorporating them into action-value methods. To support our hypothesis, we instantiate this framework in what we term QMLE, for Q-learning with maximum likelihood estimation. Our results show that QMLE can be applied to complex action spaces with a controllable computational cost that is comparable to that of policy gradient methods, all without using policy gradients. Furthermore, QMLE demonstrates strong performance on the DeepMind Control Suite, even when compared to the state-of-the-art methods such as DMPO and D4PG.

Related papers

• Last-Iterate Global Convergence of Policy Gradients for Constrained Reinforcement Learning [62.81324245896717]
  We introduce an exploration-agnostic algorithm, called C-PG, which exhibits global last-ite convergence guarantees under (weak) gradient domination assumptions. We numerically validate our algorithms on constrained control problems, and compare them with state-of-the-art baselines.
  arXiv  Detail & Related papers  (2024-07-15T14:54:57Z)
• Improving Reward-Conditioned Policies for Multi-Armed Bandits using Normalized Weight Functions [8.90692770076582]
  Recently proposed reward-conditioned policies (RCPs) offer an appealing alternative in reinforcement learning. We show that RCPs are slower to converge and have inferior expected rewards at convergence, compared with classic methods. We refer to this technique as generalized marginalization, whose advantage is that negative weights for policies conditioned on low rewards can make the resulting policies more distinct from them.
  arXiv  Detail & Related papers  (2024-06-16T03:43:55Z)
• Learning Optimal Deterministic Policies with Stochastic Policy Gradients [62.81324245896716]
  Policy gradient (PG) methods are successful approaches to deal with continuous reinforcement learning (RL) problems. In common practice, convergence (hyper)policies are learned only to deploy their deterministic version. We show how to tune the exploration level used for learning to optimize the trade-off between the sample complexity and the performance of the deployed deterministic policy.
  arXiv  Detail & Related papers  (2024-05-03T16:45:15Z)
• Actor-Critic Reinforcement Learning with Phased Actor [10.577516871906816]
  We propose a novel phased actor in actor-critic (PAAC) method to improve policy gradient estimation. PAAC accounts for both $Q$ value and TD error in its actor update. Results show that PAAC leads to significant performance improvement measured by total cost, learning variance, robustness, learning speed and success rate.
  arXiv  Detail & Related papers  (2024-04-18T01:27:31Z)
• Policy Gradient for Rectangular Robust Markov Decision Processes [62.397882389472564]
  We introduce robust policy gradient (RPG), a policy-based method that efficiently solves rectangular robust Markov decision processes (MDPs) Our resulting RPG can be estimated from data with the same time complexity as its non-robust equivalent.
  arXiv  Detail & Related papers  (2023-01-31T12:40:50Z)
• Policy Gradient Method For Robust Reinforcement Learning [23.62008807533706]
  This paper develops the first policy gradient method with global optimality guarantee and complexity analysis for robust reinforcement learning under model mismatch. We show that the proposed robust policy gradient method converges to the global optimum gradient under direct policy parameterization. We then extend our methodology to the general model-free setting and design the robust actoriable parametric policy class and value function.
  arXiv  Detail & Related papers  (2022-05-15T17:35:17Z)
• Semi-On-Policy Training for Sample Efficient Multi-Agent Policy Gradients [51.749831824106046]
  We introduce semi-on-policy (SOP) training as an effective and computationally efficient way to address the sample inefficiency of on-policy policy gradient methods. We show that our methods perform as well or better than state-of-the-art value-based methods on a variety of SMAC tasks.
  arXiv  Detail & Related papers  (2021-04-27T19:37:01Z)
• Discrete Action On-Policy Learning with Action-Value Critic [72.20609919995086]
  Reinforcement learning (RL) in discrete action space is ubiquitous in real-world applications, but its complexity grows exponentially with the action-space dimension. We construct a critic to estimate action-value functions, apply it on correlated actions, and combine these critic estimated action values to control the variance of gradient estimation. These efforts result in a new discrete action on-policy RL algorithm that empirically outperforms related on-policy algorithms relying on variance control techniques.
  arXiv  Detail & Related papers  (2020-02-10T04:23:09Z)
• Reward-Conditioned Policies [100.64167842905069]
  imitation learning requires near-optimal expert data. Can we learn effective policies via supervised learning without demonstrations? We show how such an approach can be derived as a principled method for policy search.
  arXiv  Detail & Related papers  (2019-12-31T18:07:43Z)

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.