{"id":"arxiv/2605.27420","url":"https://arxiv.org/abs/2605.27420","tags":["quant-ph","physics.app-ph"],"score":0,"title":"Hybrid Classical-Quantum Neural Networks for Multi-Characteristic Co-Optimization of Recessed-Gate AlGaN/GaN MIS-HEMTs","domain":"quant-ph","source":"arxiv","authors":["Rushat Rai","Pei-Jie Chang","Doan Viet Nguyen","Yuan-Chieh Chiu","Niall Tumilty","Yun-Yuan Wang","Simon See","Wen-Jay Lee","Tai-Yue Li","Nan-Yow Chen","Tian-Li Wu"],"reviews":0,"subject":"quant-ph","summary":"Optimizing recessed-gate AlGaN/GaN MIS-HEMTs requires accurate multi-characteristic models, but experimental semiconductor datasets remain costly and encode process-induced variability that simulations cannot faithfully reproduce. This work proposes a hybrid classical-quantum neural network (HQNN) for joint optimization of six electrical targets from a 24-dimensional fabrication/process vector. We systematically screen quantum-circuit templates to extract circuit-design guidance, then select a final HQNN and compare it directly with classical baselines. On 468 experimental fabricated devices…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-05-19","reactions":{}}