Preventing Vanishing Gradient Problem of Hardware Neuromorphic System by Implementing Imidazole‐Based Memristive ReLU Activation Neuron (Adv. Mater. 24/2023)

Memristive Neuron Devices It is essential to introduce an appropriate activation function for the training of the deep neural networks. In article number 2300023, Byung Chul Jang, Sung-Yool Choi, and co-workers demonstrate memristive rectifying linear unit (ReLU) activation neurons for hardware neuromorphic systems by introducing imidazole-based memristors. The proposed memristive ReLU neurons will enable highly integrated hardware neural networks by connecting adjacent layers, with prevented performance degradation due to the vanishing gradient problem.