mlsanityData to 14 Oct 2026

Continual Deep Reinforcement Learning with Task-Agnostic Policy Distillation

arXiv 2411.16532 ↗First posted 25 Nov 2024cs.LGcited by 3

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations

Abstract (arXiv)

Central to the development of universal learning systems is the ability to solve multiple tasks without retraining from scratch when new data arrives. This is crucial because each task requires significant training time. Addressing the problem of continual learning necessitates various methods due to the complexity of the problem space. This problem space includes: (1) addressing catastrophic forgetting to retain previously learned tasks, (2) demonstrating positive forward transfer for faster learning, (3) ensuring scalability across numerous tasks, and (4) facilitating learning without requiring task labels, even in the absence of clear task boundaries. In this paper, the Task-Agnostic Policy Distillation (TAPD) framework is introduced. This framework alleviates problems (1)-(4) by incorporating a task-agnostic phase, where an agent explores its environment without any external goal and maximizes only its intrinsic motivation. The knowledge gained during this phase is later distilled for further exploration. Therefore, the agent acts in a self-supervised manner by systematically seeking novel states. By utilizing task-agnostic distilled knowledge, the agent can solve downstream tasks more efficiently, leading to improved sample efficiency. Our code is available at the repository: https://github.com/wabbajack1/TAPD.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 4 of its 9 references that are in the map. First posted on arXiv 25 Nov 2024.

Builds on

4 of 9 references.

References · 4

  1. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  2. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  3. 2017Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agentscited by 269
  4. 2023To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Reviewcited by 72

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2025MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learningcited by 11
  2. 2025Distillation-Guided Structural Transfer for Continual Learning Beyond Sparse Distributed Memorycited by 2
  3. 2026CIDER: Continual Interactive Distillation for Embodied Reinforcement Learningcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026PCSD: Persistent Consistency for Self-Distillation in Agentic Reinforcement Learning
  2. 2026Self-Distillation Enables Continual Learning
  3. 2024Reset & Distill: A Recipe for Overcoming Negative Transfer in Continual Reinforcement Learning
  4. 2025A Survey of Continual Reinforcement Learning
  5. 2025Self-Composing Policies for Scalable Continual Reinforcement Learning
  6. 2025An Empirical Study of Deep Reinforcement Learning in Continuing Tasks
  7. 2026Scalable In-Context Reinforcement Learning with Recurrent Algorithm Distillation
  8. 2026CADENT: Gated Hybrid Distillation for Sample-Efficient Transfer in Reinforcement Learning