Rethinking the Suitability of Reinforcement Learning Algorithms Under Practical Transfer Constraints

• URL: http://arxiv.org/abs/2607.17326v1
• Date: Sun, 19 Jul 2026 16:29:13 GMT
• Title: Rethinking the Suitability of Reinforcement Learning Algorithms Under Practical Transfer Constraints
• Abstract summary: We provide two insights to reinforcement-learning practitioners.<n>First, comparing the sample efficiency of different algorithms is often an insufficient criterion in transfer-oriented settings.<n>Second, domain randomization can help different kinds of algorithms learn robust policies.
• Score: 9.42374503870091
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Transfer-oriented reinforcement learning requires evaluating algorithms along dimensions that go beyond standard sample efficiency. We focus on two dimensions: practical efficiency, which asks whether conclusions about algorithm suitability change under wall-clock rather than interaction-based budgets, and robustness under dynamics mismatch, which asks how different learning paradigms respond to variability in the training distribution induced by domain randomization. We provide two insights to reinforcement-learning practitioners. First, comparing the sample efficiency of different algorithms is often an insufficient criterion in transfer-oriented settings. The wall-clock time required to train a decent policy is an important consideration for practitioners, and we find that the sample-inefficient PPO algorithm can produce a performant policy faster than relatively more sample-efficient algorithms such as SAC and TD-MPC2, validating the common understanding of massively parallel training paradigms. Second, domain randomization can help different kinds of algorithms learn robust policies. In particular, although PPO, SAC, and TD-MPC2 represent different RL paradigms - on-policy, off-policy, and model-based learning and planning, respectively - we find that domain randomization affects all three algorithms in a similar way. To the best of our knowledge, this is the first controlled comparison of the effect of domain-randomization coverage on PPO, SAC, and TD-MPC2 under the same transfer protocol. Taken together, these two insights highlight the importance of evaluating RL algorithms not only by sample efficiency, but also by practical considerations such as training time and the algorithms' ability to produce usable policies.

Related papers

• Breaking the Computational Barrier: Provably Efficient Actor-Critic for Low-Rank MDPs [53.412166189410904]
  We use supervised learning as a computational proxy to establish a hierarchy of commonly adopted RL oracles under low-rank Markov Decision Processes (MDPs)<n>Motivated by this observation, we propose a novel optimistic actor-critic algorithm that relies solely on the policy evaluation oracle.<n>We prove that our algorithm outperforms the existing sample complexity guarantees for low-rank MDPs while avoiding computationally expensive planning or optimization oracles commonly assumed in prior works.
  arXiv  Detail & Related papers  (2026-05-02T04:46:54Z)
• Max-Entropy Reinforcement Learning with Flow Matching and A Case Study on LQR [22.463270191334434]
  Soft actor-critic (SAC) is a popular algorithm for max-entropy reinforcement learning.<n>We propose a variant of the SAC algorithm that parameterizes the policy with flow-based models.<n>Online variant of flow matching enables policy update with only samples from a user-specified sampling distribution.
  arXiv  Detail & Related papers  (2025-12-29T21:23:52Z)
• What Matters for Batch Online Reinforcement Learning in Robotics? [65.06558240091758]
  The ability to learn from large batches of autonomously collected data for policy improvement holds the promise of enabling truly scalable robot learning.<n>Previous works have applied imitation learning and filtered imitation learning methods to the batch online RL problem.<n>We analyze how these axes affect performance and scaling with the amount of autonomous data.
  arXiv  Detail & Related papers  (2025-05-12T21:24:22Z)
• An advantage based policy transfer algorithm for reinforcement learning with measures of transferability [5.926203312586109]
  Reinforcement learning (RL) enables sequential decision-making in complex and high-dimensional environments.<n>This paper proposes an off-policy Advantage-based Policy Transfer algorithm, APT-RL, for fixed domain environments.
  arXiv  Detail & Related papers  (2023-11-12T04:25:53Z)
• Statistically Efficient Variance Reduction with Double Policy Estimation for Off-Policy Evaluation in Sequence-Modeled Reinforcement Learning [53.97273491846883]
  We propose DPE: an RL algorithm that blends offline sequence modeling and offline reinforcement learning with Double Policy Estimation. We validate our method in multiple tasks of OpenAI Gym with D4RL benchmarks.
  arXiv  Detail & Related papers  (2023-08-28T20:46:07Z)
• Distillation Policy Optimization [5.439020425819001]
  We introduce an actor-critic learning framework that harmonizes two data sources for both evaluation and control. This framework incorporates variance reduction mechanisms, including a unified advantage estimator (UAE) and a residual baseline. Our results showcase substantial enhancements in sample efficiency for on-policy algorithms, effectively bridging the gap to the off-policy approaches.
  arXiv  Detail & Related papers  (2023-02-01T15:59:57Z)
• Generalized Policy Improvement Algorithms with Theoretically Supported Sample Reuse [15.134707391442236]
  We develop a new class of model-free deep reinforcement learning algorithms for data-driven, learning-based control. Our Generalized Policy Improvement algorithms combine the policy improvement guarantees of on-policy methods with the efficiency of sample reuse.
  arXiv  Detail & Related papers  (2022-06-28T02:56:12Z)
• Deterministic and Discriminative Imitation (D2-Imitation): Revisiting Adversarial Imitation for Sample Efficiency [61.03922379081648]
  We propose an off-policy sample efficient approach that requires no adversarial training or min-max optimization. Our empirical results show that D2-Imitation is effective in achieving good sample efficiency, outperforming several off-policy extension approaches of adversarial imitation.
  arXiv  Detail & Related papers  (2021-12-11T19:36:19Z)
• A Subgame Perfect Equilibrium Reinforcement Learning Approach to Time-inconsistent Problems [4.314956204483074]
  We establish a subgame perfect equilibrium reinforcement learning framework for time-inconsistent (TIC) problems. We propose a new class of algorithms, called backward policy iteration (BPI), that solves SPERL and addresses both challenges. To demonstrate the practical usage of BPI as a training framework, we adapt standard RL simulation methods and derive two BPI-based training algorithms.
  arXiv  Detail & Related papers  (2021-10-27T09:21:35Z)
• Learning Sampling Policy for Faster Derivative Free Optimization [100.27518340593284]
  We propose a new reinforcement learning based ZO algorithm (ZO-RL) with learning the sampling policy for generating the perturbations in ZO optimization instead of using random sampling. Our results show that our ZO-RL algorithm can effectively reduce the variances of ZO gradient by learning a sampling policy, and converge faster than existing ZO algorithms in different scenarios.
  arXiv  Detail & Related papers  (2021-04-09T14:50:59Z)
• Variance-Reduced Off-Policy Memory-Efficient Policy Search [61.23789485979057]
  Off-policy policy optimization is a challenging problem in reinforcement learning. Off-policy algorithms are memory-efficient and capable of learning from off-policy samples.
  arXiv  Detail & Related papers  (2020-09-14T16:22:46Z)

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.