Researchers have developed a novel reinforcement learning framework that utilizes compact parametrized quantum circuits (PQCs) for device modeling, particularly in scenarios with limited data. This framework employs a graph neural network policy optimized via proximal policy optimization to search for optimal circuit architectures. The approach demonstrated superior performance compared to classical baselines, achieving significantly lower mean absolute error and tighter fold variability for both Gallium Nitride High Electron Mobility Transistors (GaN HEMTs) and Nanowire FETs (NWFETs). These findings highlight the efficacy of classically simulated PQCs selected by RL as effective surrogates for device datasets, even without explicit physical constraints. AI
IMPACT This research demonstrates a novel application of quantum machine learning and reinforcement learning for improving accuracy and variability in device modeling, potentially impacting semiconductor design and manufacturing.
RANK_REASON The cluster contains a research paper published on arXiv detailing a new framework for device modeling using quantum machine learning. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- compact parametrized quantum circuits (PQCs)
- GaN HEMTs
- Graph Neural Network (GNN)
- Khemchik
- leave-one-group-out cross-validation (LOGOCV)
- Nanowire FETs
- NWFETs
- Proximal Policy Optimization (PPO)
- Reinforcement Learning (RL)-Based Energy Efficient Resource Allocation for Energy Harvesting-Powered Wireless Body Area Network
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
