Master of Engineering (Vehicle Engineering), Wuhan University of Technology, China, 2004;

Doctor of Engineering (Automation and Computer Aided Engineering), Chinese University of Hong Kong, China, 2007. Artificial neural network.

National Natural Science Foundation of Computational Neuroscience (Youth): Optimized recurrent neural network cluster design based on KKT conditions (2009-2011);

National Natural Science Foundation of China (General): The research work of Deep Learning Neural Network (20 13-20 16) based on sparse coding model focuses on the cross direction of computer science and cognitive neuroscience, including artificial neural network and computational neuroscience. On the one hand, I am interested in uncovering the mysteries of the brain, especially the mechanism by which the brain processes sensory information and decision-making information. The main tools used are hierarchical computing model and Bayesian theory. We are also trying to use functional magnetic resonance imaging (fMRI) combined with machine learning to explore the working mechanism of the brain. On the other hand, I am interested in brain-inspired calculation methods. The research focus is to design recursive neural network to solve optimization related problems. We are trying to combine more knowledge of cognitive neuroscience to improve the accuracy and efficiency of deep learning model in object recognition and detection.

Some work has been done on the information processing mechanism of the ventral visual pathway of the brain, and a series of hierarchical models have been established to explain the response characteristics of neurons at all levels of the pathway (including V 1, V2, IT and other regions). Two typical tasks are to improve HMAX model, increase sparsity and feedback connection, which can better explain a series of neuroscience data. The related results were published in PLoS ONE (20 14) and Neural computing(20 10) respectively.

With regard to brain-inspired computing methods, in recent ten years, most of the work has focused on the theory and method of solving optimization problems by recursive neural networks, deeply excavating the characteristics of existing models, designing a series of new models, and publishing related achievements in several IEEE journals. Some work has also been done in deep learning. In IJCCNN 2013 German traffic sign detection competition, convolution neural network was used to get the second and fourth place of two kinds of signs. In addition to target recognition and detection, image salient region detection is also a concerned application. Referring to a theory in psychology, the inverse hierarchy theory, a hierarchical model is constructed, which can well predict the gaze point of human eyes in the image. This achievement was accepted by CVPR 14, an important conference on computer vision. First Prize of Natural Science of Ministry of Education (ranked third): Neurodynamic Optimization Model and Its Application (20 12)

Excellent postdoctoral fellow in Tsinghua University (2009)

ICONIP 20 12: Best Paper Award (20 12) [1] P. Qi, X. Hu, "Learning nonlinear statistical laws in natural images by modeling the outer product of image intensity", neural computing (accepted).

[2] X. Hu, J. Zhang, P. Qi, B. Zhang, "Simulating the response characteristics of neurons with hierarchical K-means model", Neurocomputing, vol. 134, pp. 198-205, 20 14.

[3] X. Hu, J. Zhang, J. Li, B. Zhang, "Sparse-regularized HMAX for Visual Recognition", Volume 1,No. 1, Page12, 20/kloc-0.

[4] X. Hu, J. Wang, "Solving assignment problems with continuous-time and discrete-time improved dual networks", IEEE Transactions on Neural Networks and Learning Systems, Vol.23, No.5, pp.821-827, 20 12.

[5] X. Hu, B. Zhang, "Gaussian Attractor Network for Memory and Recognition Based on Empirical Dependence Learning", Neurocomputing, Vol.22, No.5, pp.1333-1357,2010.

[6] X. Hu, C. Sun, B. Zhang, "Design of Recursive Neural Networks for Solving Constrained Least Squares Deviation Problems", Journal of IEEE Neural Networks, No.21Volume 7, No.1 1073- 1086, 20/kloc.

[7] X. Hu, B. Zhang, "Alternative recurrent neural networks for solving variational inequalities and related optimization problems", IEEE Transactions on Systems, People and Cybernetics-Part B, Volume 39, No.6, pp. 1640- 1645, 2009/kloc-0.

[8] X. Hu and B. Zhang, "A New Recurrent Neural Network for Solving Convex Quadratic Programming Problems and Its Application in the k- winner Take All Problem", Journal of IEEE Neural Networks, Vol.20, No.4, pp.654-664, April 2009.