LLM-Driven Composite Neural Architecture Search for Multi-Source RL State Encoding
By: Yu Yu , Qian Xie , Nairen Cao and more
Potential Business Impact:
Helps robots learn faster from many senses.
Designing state encoders for reinforcement learning (RL) with multiple information sources -- such as sensor measurements, time-series signals, image observations, and textual instructions -- remains underexplored and often requires manual design. We formalize this challenge as a problem of composite neural architecture search (NAS), where multiple source-specific modules and a fusion module are jointly optimized. Existing NAS methods overlook useful side information from the intermediate outputs of these modules -- such as their representation quality -- limiting sample efficiency in multi-source RL settings. To address this, we propose an LLM-driven NAS pipeline that leverages language-model priors and intermediate-output signals to guide sample-efficient search for high-performing composite state encoders. On a mixed-autonomy traffic control task, our approach discovers higher-performing architectures with fewer candidate evaluations than traditional NAS baselines and the LLM-based GENIUS framework.
Similar Papers
Tutorial on Large Language Model-Enhanced Reinforcement Learning for Wireless Networks
Networking and Internet Architecture
AI helps wireless networks learn and adapt better.
Enhancing Autonomous Driving Systems with On-Board Deployed Large Language Models
Artificial Intelligence
Helps self-driving cars handle unexpected situations better.
A Continuous Encoding-Based Representation for Efficient Multi-Fidelity Multi-Objective Neural Architecture Search
Machine Learning (CS)
Finds best computer designs faster for complex jobs.