PhD Student

Research Direction: His research interests lie in sparse training, epitopological learning, and quantum computing. Most of his recent research focuses on how to create novel and more efficient artificial intelligence algorithms with the inspiration of brain. To achieve this target, he designed epitopological sparse deep learning in artificial neural networks, which realized the connectivity sparseness in brain, following the intuition of epitopological learning. The next stage of his research will contain 4 steps:

1) Achieve the neuron-wise sparse training in ANNs and combine it with the connectivity-wise sparse training.

2) Explore the initialization of weights of sparse training.

3) Accelerate the process of sparse operation on hardware.

4) Try to find the connections between sparse structure of ANNs and quantum field.