Optimization of Projected Phase Change Memory for Analog In‐Memory Computing Inference (Adv. Electron. Mater. 6/2023)

Large Neural Networks In article number 2201190, Ning Li and co-workers systematically tune phase change memory electrical properties using a projection liner, which also reduces resistance drift and read noise. The inference accuracy of large deep neural networks is improved for both the short term and long term after programming. The device optimization space is identified to achieve the best accuracy for large neural networks with real applications.