mlsanityData to 7 Oct 2026

On Neural Consolidation for Transfer in Reinforcement Learning

arXiv 2210.02240 ↗First posted 5 Oct 2022cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Although transfer learning is considered to be a milestone in deep reinforcement learning, the mechanisms behind it are still poorly understood. In particular, predicting if knowledge can be transferred between two given tasks is still an unresolved problem. In this work, we explore the use of network distillation as a feature extraction method to better understand the context in which transfer can occur. Notably, we show that distillation does not prevent knowledge transfer, including when transferring from multiple tasks to a new one, and we compare these results with transfer without prior distillation. We focus our work on the Atari benchmark due to the variability between different games, but also to their similarities in terms of visual features.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 26 of its 33 references that are in the map. First posted on arXiv 5 Oct 2022.

Builds on

26 of 33 references.

References · 26

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  3. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  4. 2016Learning without Forgettingcited by 2,465
  5. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,313
  6. 2017Gradient Episodic Memory for Continual Learningcited by 1,871
  7. 2018World Modelscited by 1,444
  8. 2016Progressive Neural Networkscited by 1,332
  9. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,313
  10. 2020Gradient Surgery for Multi-Task Learningcited by 1,210
  11. 2017Continual Learning with Deep Generative Replaycited by 1,152
  12. 2015Prioritized Experience Replaycited by 1,070
14 more
  1. 2017Rainbow: Combining Improvements in Deep Reinforcement Learningcited by 774
  2. 2017Lifelong Learning with Dynamically Expandable Networkscited by 689
  3. 2018Progress & Compress: A scalable framework for continual learningcited by 516
  4. 2019Model-Based Reinforcement Learning for Ataricited by 495
  5. 2015Policy Distillationcited by 414
  6. 2017PathNet: Evolution Channels Gradient Descent in Super Neural Networkscited by 362
  7. 2017Distral: Robust Multitask Reinforcement Learningcited by 231
  8. 2015Actor-Mimic: Deep Multitask and Transfer Reinforcement Learningcited by 176
  9. 2018Dopamine: A Research Framework for Deep Reinforcement Learningcited by 121
  10. 2016Less-forgetting Learning in Deep Neural Networkscited by 97
  11. 2020Learning to Continually Learncited by 82
  12. 2019Task Agnostic Continual Learning via Meta Learningcited by 53
  13. 2018Progressive Reinforcement Learning with Distillation for Multi-Skilled Motion Controlcited by 12
  14. 2019A Complementary Learning Systems Approach to Temporal Difference Learningcited by 11

Built on by

Not cited yet.