Zekun Hao, Yu Liu, Hongwei Qin, Junjie Yan, Xiu Li, Xiaolin Hu, “Scaleaware face detection,” Proc. of the 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, July 21 – 26. 

Tiancheng Sun, Yulong Wang, Jian Yang, Xiaolin Hu, “Convolutional neural networks with two pathways for image style recognition,” IEEE Transactions on Image Processing, 2017.The gram matrix technique proposed by Gatys et al. is used to classify image styles. Three benchmark datasets are experimented, WikiPaintings, Flickr Style and AVA Style. 

J. Wu, L. Ma, X. Hu, “Delving deeper into convolutional neural networks for camera relocalization,” Proc. of IEEE International Conference on Robotics and Automation (ICRA), Singapore, May 29 June 3, 2017.We present three techniqus for enhancing the performance of convolutional neural networks for camera relocalizationare. 

F. Liao, X. Hu, S. Song, “Emergence of V1 recurrent connectivity pattern in artificial neural network,”Computational and Systems Neuroscience (Cosyne), Salt Lake City, Feb. 23  26, 2017.


Y. Zhao, X. Jin, X. Hu, “Recurrent convolutional neural network for speech processing,” Proc. of the 42nd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, USA, March 59, 2017.Applications of recurrent CNN to speech processing. 
Z. Cheng, Z. Deng, X. Hu, B. Zhang, T. Yang*, “Efficient reinforcement learning of a reservoir network model of parametric working memory achieved with a cluster population winnertakeall readout mechanism,” Journal of Neurophysiology, vol.114, no. 6, 32963305, 2015.Learning of a reservoir network for working memory of monkey brain. 

X. Li, S. Qian, F. Peng, J. Yang*, X. Hu, and R. Xia, "Deep convolutional neural network and multiview stacking ensemble in Ali mobile recommendation algorithm competition," The First International Workshop on Mobile Data Mining & Human Mobility Computing (ICDM 2015).The team won the Ali competition. Rank 1st over 7186 teams. . 

M. Liang, X. Hu*, B. Zhang, “Convolutional neural networks with intralayer recurrent connections for scene labeling,” Advances in Neural Information Processing (NIPS), Montréal, Canada, Dec. 712, 2015.An application of the recurrent CNN. It achieves excellent performance on the Stanford Background and SIFT Flow datasets. 

Y. Zhou, X. Hu*, B. Zhang, “Interlinked convolutional neural networks for face parsing,” International Symposium on Neural Networks (ISNN), Jeju, Korea, Oct. 1518, 2015, pp. 222231.A twostage pipeline is proposed for face parsing and both stages use iCNN, which is a set of CNNs with interlinkage in the convolutional layers.. 

M. Liang, X. Hu*, “Recurrent convolutional neural network for object recognition,” Proc. of the 28th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, USA, June 712, 2015, pp. 33673375.cudaconvnet2 configs (used in the paper) Typical deep learning models for object recognition have feedforward architectures including HMAX and CNN.This is a crude approximation of the visual pathway in the brain since there are abundant recurrent connections in the visual cortex. We show that adding recurrent connections to CNN improves its performance in object recognition. 

X. Zhang, Q. Zhang, X. Hu*, B. Zhang, “Neural representation of threedimensional acoustic space in the human temporal lobe,” Frontiers in Human Neuroscience, vol. 9, article 203, 2015. doi: 10.3389/fnhum.2015.00203Humans are able to localize the sounds in the environment. How the locations are encoded in the cortex remains elusive. Using fMRI and machine learning techniques, we investigated how the temporal cortex of humans encodes the 3D acoustic space. 

M. Liang, X. Hu*, “Predicting eye fixations with higherlevel visual features,” IEEE Transactions on Image Processing, vol. 24, no. 3, pp. 11781189, 2015.There is a debate about whether lowlevel features or highlevel features are more important for prediction eye fixations. Through experiments, we show that midlevel features and objectlevel features are indeed more effective for this task. We obtained stateoftheart results on several benchmark datasets including Toronto, MIT, Kootstra and ASCMN at the time of submission. 

M. Liang, X. Hu*, “Feature selection in supervised saliency prediction,” IEEE Transactions on Cybernetics, vol. 45, no. 5, pp. 900912, 2015.(Download the computed saliency maps here) There is a trend for incorporating more and more features for supervised learning of visual saliency on natural images. We find much redundancy among these features by showing that a small subset of features leads to excellent performance on several benchmark datasets. In addition, these features are robust across different datasets. 

Q. Zhang, X. Hu*, B. Zhang, “Comparison of L1Norm SVR and Sparse Coding Algorithms for Linear Regression,” IEEE Transactions on Neural Networks and Learning Systems, vol. 26, no. 8, pp. 18281833, 2015.The close connection between the L1norm support vector regression (SVR) and sparse coding (SC) is revealed and some typical algorithms are compared for linear regression. The results show that the SC algorithms outperform the L1SVR algorithms in efficiency. The SC algorithms are then used to design RBF networks, which are more efficient than the wellknown orthogonal least squares algorithm. 
T. Shi, M. Liang, X. Hu*, “A reverse hierarchy model for predicting eye fixations,” Proc. of the 27th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, USA, June 2427, 2014, pp. 28222829.We present a novel approach for saliency detection in natural images. The idea is from a theory in cognitive neuroscience, called reverse hierarchy theory, which proposes that attention propagates from the top level of the visual hierarchy to the bottom level. 

X. Hu*, J. Zhang, J. Li, B. Zhang, “Sparsityregularized HMAX for visual recognition,” PLOS ONE, vol. 9, no. 1, e81813, 2014.We show that a deep learning model with alternating sparse coding/ICA and local max pooling can learn higherlevel features on images without labels. After training on a dataset with 1500 images, in which there were 150 unaligned faces, 6 units on the top layer became face detectors. This took a few hours on a laptop computer with 2 cores, in contrast to Google's 16,000 cores in a similar project. 

X. Hu*, J. Zhang, P. Qi, B. Zhang, “Modeling response properties of V2 neurons using a hierarchical Kmeans model,” Neurocomputing, vol. 134, pp. 198205, 2014.We show that the simple data clustering algorithm, Kmeans can be used to model some properties of V2 neurons if we stack them into a hierarchical structure. It is more biologically feasible than the sparse DBN for doing the same thing because it can be realized by competitive hebbian learning. This is an extended version of our ICONIP'12 paper. 

P. Qi, X. Hu*, “Learning nonlinear statistical regularities in natural images by modeling the outer product of image intensities,” Neural Computation, vol. 26, no. 4, pp. 693–711, 2014.This is a hierarchical model aimed at modeling the properties of complex cells in the primary visual cortex (V1). It can be regarded as a simplified version of Karklin and Lewicki's model published in 2009. 
20112010200920082007X. Hu and J. Wang, “Solving the kwinnerstakeall problem and the oligopoly CournotNash equilibrium problem using the general projection neural networks.” Proc. of 14th International Conference on Neural Information Processing (ICONIP), Kitakyushu, Japan, Nov. 1316, 2007, pp. 703712. S. Liu, X. Hu and J. Wang, “Obstacle Avoidance for Kinematically Redundant Manipulators Based on an Improved Problem Formulation and the Simplified Dual Neural Network”, Proc. of IEEE ThreeRivers Workshop on Soft Computing in Industrial Applications, Passau, Bavaria, Germany, August 13, 2007, pp. 6772. X. Hu and J. Wang, “Convergence of a recurrent neural network for nonconvex optimization based on an augmented Lagrangian function,” Proc. of 4th International Symposium on Neural Networks, Part III, Nanjing, China, June 37, 2007. 2006X. Hu and J. Wang, “Solving extended linear programming problems using a class of recurrent neural networks,” Proc. of 13th International Conference on Neural Information Processing, Part II, Hong Kong, Oct. 36, 2006. 