Bioinspired neural networks for deep learning
Intelligent Computing and Systems
The Intelligent Computing and Systems (ICS) Lab is driving theoretical and practical innovation in circuit design, dynamic analysis, and applications of memristor-based neural networks, neuromorphic computing, and brain-inspired computing. The Lab studies the characteristics of neuromorphic computing, explores semi-supervised learning data set processing technology, designs and improves deep learning algorithms. ICS researchers use neuromorphic computing circuits to implement the improved deep learning algorithms, to compress the model to reduce the calculation cost and file size of the model to maintain accuracy, and choose the appropriate neural network framework with acceleration technology to support large-scale deep learning network applications.