analiza aiqbench
This paper is clearly AI/ML-centric: it proposes joint neural architecture search and hyperparameter optimization for reinforcement learning policies used in autonomous driving. The main contribution is a method for co-evolving network structure and training settings, with reported gains in reward, parameter count, and stability over NAS alone, measured in a single simulator. Readers should weigh these results as simulator-based and from a prior-work lineage (EMNAS), with generalization to real-world driving still unproven.
Hyperparameter and Neural Architecture Search via Evolutionary Optimization for Reinforcement Learning: HPNAS-RL
arXiv:2610.36882v1 Announce Type: new Abstract: In this study, we propose a dual-branch evolutionary framework that simultaneously performs Neural Architecture Search (NAS) and Hyperparameter Optimization (HPO) for Reinforcement Learning (RL) in Autonomous Driving (AD) tasks. While NAS has shown strong potential in automating architecture design for Convolutional Neural Networks (CNNs), its application to RL remains limited, especially in scenarios that require joint optimization of architectural and training dynamics. Building on our prior work on Evolutionary Multi-Objective NAS (EMNAS), the proposed framework evolves discrete network architectures and continuous hyperparameters in parallel under a unified multi-objective search strategy. In this setup, NAS targets the convolutional layers of the RL policy network - responsible for visual feature extraction, while HPO focuses on optimizing critical RL hyperparameters. To accelerate convergence and improve generalization, we incorporate parallel evaluation and a teacher-student policy transfer mechanism. Experiments in a high-fidelity driving simulator show that the joint NAS-HPO framework consistently outperforms standalone NAS, yielding policies with higher rewards, fewer parameters, and improved stability. These results highlight the benefits of jointly optimizing model structure and training dynamics for scalable and effective RL agent design.
Abstract:In this study, we propose a dual-branch evolutionary framework that simultaneously performs Neural Architecture Search (NAS) and Hyperparameter Optimization (HPO) for Reinforcement Learning (RL) in Autonomous Driving (AD) tasks. While NAS has shown strong potential in automating architecture design for Convolutional Neural Networks (CNNs), its application to RL remains limited, especially in scenarios that require joint optimization of architectural and training dynamics. Building on our prior work on Evolutionary Multi-Objective NAS (EMNAS), the proposed framework evolves discrete network architectures and continuous hyperparameters in parallel under a unified multi-objective search strategy. In this setup, NAS targets the convolutional layers of the RL policy network - responsible for visual feature extraction, while HPO focuses on optimizing critical RL hyperparameters. To accelerate convergence and improve generalization, we incorporate parallel evaluation and a teacher-student policy transfer mechanism. Experiments in a high-fidelity driving simulator show that the joint NAS-HPO framework consistently outperforms standalone NAS, yielding policies with higher rewards, fewer parameters, and improved stability. These results highlight the benefits of jointly optimizing model structure and training dynamics for scalable and effective RL agent design.
Machine Learning (cs.LG)
arXiv:2610.36882 [cs.LG]
(or arXiv:2610.36882v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.36882 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 273 cuvinte
Subiecte conexe
Efficient Hyperparameter Optimization for LLM Reinforcement Learning
arXiv · 3 iun. 2026
Continuous Delayed-Memory Stochastic Gradient Descent and Continuous-Time Reinforcement Learning from History of Astrophysical Time Series Studies
arXiv · 21 sept. 2026
Non-Stationary Functional Bilevel Optimization
arXiv · 24 iul. 2026
Reinforcement learning to choose optimizers
arXiv · 3 sept. 2026