mlsanityData to 7 Oct 2026

Intelligent Switching for Reset-Free RL

arXiv 2405.01684 ↗First posted 2 May 2024cs.LG cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

In the real world, the strong episode resetting mechanisms that are needed to train agents in simulation are unavailable. The textitresetting assumption limits the potential of reinforcement learning in the real world, as providing resets to an agent usually requires the creation of additional handcrafted mechanisms or human interventions. Recent work aims to train agents (textitforward) with learned resets by constructing a second (textitbackward) agent that returns the forward agent to the initial state. We find that the termination and timing of the transitions between these two agents are crucial for algorithm success. With this in mind, we create a new algorithm, Reset Free RL with Intelligently Switching Controller (RISC) which intelligently switches between the two agents based on the agent's confidence in achieving its current goal. Our new method achieves state-of-the-art performance on several challenging environments for reset-free RL.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 23 of its 30 references that are in the map. First posted on arXiv 2 May 2024.

Builds on

23 of 30 references.

References · 23

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  2. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  3. 2021Deep Reinforcement Learning at the Edge of the Statistical Precipicecited by 781
  4. 2020Towards Continual Reinforcement Learning: A Review and Perspectivescited by 268
  5. 2017Reverse Curriculum Generation for Reinforcement Learningcited by 185
  6. 2020Automatic Curriculum Learning For Deep RL: A Short Surveycited by 141
  7. 2017Intrinsic Motivation and Automatic Curricula via Asymmetric Self-Playcited by 139
  8. 2023Human-Timescale Adaptation in an Open-Ended Task Spacecited by 136
  9. 2020The Ingredients of Real-World Robotic Reinforcement Learningcited by 111
  10. 2021Reset-Free Reinforcement Learning via Multi-Task Learning: Learning Dexterous Manipulation Behaviors without Human Interventioncited by 90
  11. 2017Leave no Trace: Learning to Reset for Safe and Autonomous Reinforcement Learningcited by 86
  12. 2017Time Limits in Reinforcement Learningcited by 72
11 more
  1. 2020Reset-Free Lifelong Learning with Skill-Space Planningcited by 38
  2. 2021Autonomous Reinforcement Learning: Formalism and Benchmarkingcited by 33
  3. 2021Autonomous Reinforcement Learning via Subgoal Curriculacited by 32
  4. 2020Continual Learning of Control Primitives: Skill Discovery via Reset-Gamescited by 29
  5. 2023Self-Improving Robots: End-to-End Autonomous Visuomotor Reinforcement Learningcited by 29
  6. 2022A State-Distribution Matching Approach to Non-Episodic Reinforcement Learningcited by 21
  7. 2022Demonstration-Bootstrapped Autonomous Practicing via Multi-Task Reinforcement Learningcited by 15
  8. 2020Ecological Reinforcement Learningcited by 13
  9. 2022Automating Reinforcement Learning with Example-based Resetscited by 11
  10. 2019How Should an Agent Practice?cited by 8
  11. 2023Demonstration-free Autonomous Reinforcement Learning via Implicit and Bidirectional Curriculumcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 3.

4 papers.

  1. 2025Learning Without Time-Based Embodiment Resets in Soft-Actor Criticcited by 1
  2. 2026Reset-Free Reinforcement Learning for Real-World Agile Driving: An Empirical Studycited by 0
  3. 2026REVERSAL-BENCH: A Reversibility Axis and Reset Oracle for Measuring the Reset-Free RL Cliffcited by 0
  4. 2026Latent evolving World Action Modelcited by 0