mlsanityData to 14 Oct 2026

Provably Sub-Linear Two-Timescale NeuroEvolution with Online Plasticity

arXiv 2606.20817 ↗First posted 18 Jun 2026cs.NE cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

NeuroEvolution of Augmenting Topologies (NEAT) is a widely used neuroevolution algorithm for learning neural network architectures and weights for control tasks. However, standard offline optimisation searches for connection strengths directly, which can scale poorly in high-dimensional weight spaces and more difficult continuous control problems. Hybrid methods that combine neuroevolution with online learning can address this challenge, but their theoretical properties remain underexplored. This paper gives the first regret analysis for a general NeuroEvolutionary Online Learning (NEOL) framework, which decouples learning into two timescales: an outer loop for architecture search and an inner loop for online weight adaptation via rewardmodulated plasticity. Under mild conditions, we prove that NEOL achieves sublinear regret. Empirically, under fixed interaction budgets on four standard control benchmarks, a NEAT-based NEOL implementation achieves higher final fitness and lower variance than pure NEAT, and is competitive with strong reinforcement learning (RL) baselines on several tasks. The results are supported byWilcoxon rank-sum tests and ablation studies. Overall, the findings show that online plasticity can improve the sample efficiency and robustness of two-timescale neuroevolution. Code is available at https://github.com/boobaa2001/NeuroEvolution Online Learning NEOL.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 14 of its 55 references that are in the map. First posted on arXiv 18 Jun 2026.

Builds on

14 of 55 references.

References · 14

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  3. 2019Online Learning: A Modern Introduction Using Convex Optimizationcited by 417
  4. 2017Evolving Deep Neural Networkscited by 95
  5. 2018Differentiable plasticity: training plastic neural networks with backpropagationcited by 84
  6. 2018Evolution-Guided Policy Gradient in Reinforcement Learningcited by 83
  7. 2018Eligibility Traces and Plasticity on Behavioral Time Scales: Experimental Support of neoHebbian Three-Factor Learning Rulescited by 67
  8. 2021Evolving Reinforcement Learning Algorithmscited by 49
  9. 2020Meta-Learning through Hebbian Plasticity in Random Networkscited by 47
  10. 2017Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networkscited by 29
  11. 2022Neuroevolution is a Competitive Alternative to Reinforcement Learning for Skill Discoverycited by 23
  12. 2023First Steps Towards a Runtime Analysis of Neuroevolutioncited by 1
2 more
  1. 2023Discovering Effective Policies for Land-Use Planning with Neuroevolutioncited by 1
  2. 2025Runtime Analysis of Evolutionary NAS for Multiclass Classificationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Continual Reinforcement Learning with Neuroevolution
  2. 2025When Does Neuroevolution Outcompete Reinforcement Learning in Transfer Learning Tasks?
  3. 2025Gradient-Based Neuroplastic Adaptation for Concurrent Optimization of Neuro-Fuzzy Networks
  4. 2024NeoRL: Efficient Exploration for Nonepisodic RL
  5. 2025Learning Upper Lower Value Envelopes to Shape Online RL: A Principled Approach
  6. 2024A Natural Extension To Online Algorithms For Hybrid RL With Limited Coverage
  7. 2026Policy Plasticity Matters in Offline-to-Online Reinforcement Learning: Refitting Offline Policies for Online Adaptation
  8. 2022On Practical Reinforcement Learning: Provable Robustness, Scalability, and Statistical Efficiency