Continual Evolution Strategies in Control Tasks
Not on the map. Why?
Abstract (arXiv)
We study Evolution Strategies (ES) for continual control, where agents must adapt to changing tasks without forgetting previous ones. On sequential MuJoCo locomotion tasks, naive ES suffers from severe catastrophic forgetting. Replay substantially improves retention and can induce positive transfer, while larger replay budgets reduce plasticity. Overall, these results show that ES can support continual adaptation in control and that replay is an effective mechanism for mitigating forgetting.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 10 of its 20 references that are in the map. First posted on arXiv 3 Aug 2026.
Builds on
10 of 20 references.
References · 10
- 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
- 2017Gradient Episodic Memory for Continual Learningcited by 1,898
- 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
- 2018Experience Replay for Continual Learningcited by 1,095
- 2017Evolution Strategies as a Scalable Alternative to Reinforcement Learningcited by 574
- 2020Class-incremental learning: survey and performance evaluation on image classificationcited by 449
- 2020Towards Continual Reinforcement Learning: A Review and Perspectivescited by 274
- 2019Continual Learning for Robotics: Definition, Framework, Learning Strategies, Opportunities and Challengescited by 93
- 2021Replay in Deep Learning: Current Approaches and Missing Biological Elementscited by 72
- 2025Achieving Deep Continual Learning via Evolutioncited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.
- 2026Continual Reinforcement Learning with Neuroevolutioncited by 0
- 2026Continual Reinforcement Learning with Neuroevolution1 Oct 2026