Fig. 4: The device flexibility test and implementation of CNN using the weight change behavior of the proposed device. | Microsystems & Nanoengineering

Fig. 4: The device flexibility test and implementation of CNN using the weight change behavior of the proposed device.

From: Ionic liquid multistate resistive switching characteristics in two terminal soft and flexible discrete channels for neuromorphic computing

Fig. 4

The synaptic plasticity of the a unbending, b bending, and c stretching devices under a continuous pulse train with a pulse of 1 ms. d Diagram of neural networks for the recognition of CIFAR-10 data. e Simulated accuracy of ideal and real devices. f Device endurance performance. The experimental STDP results show g antisymmetric Hebbian and h antisymmetric anti-Hebbian learning rules

Back to article page