Yuchen Zhang



  I am a post-doc researcher at Stanford University, hosted by Percy Liang and Moses Charikar. The goal of my research is to develop generic tools for building the next generation of artificial intelligence. I am interested in convex and non-convex optimization, deep learning, distributed algorithms, statistical learning theory, crowdsourcing, etc. Prior to Stanford, I got Ph.D. in computer science from UC Berkeley, where I was very fortunate to be advised by Michael Jordan and Martin Wainwright. I got a Master degree in statistics from Berkeley, and a Bachelor degree in computer science from Tsinghua University.

I enjoyed a number of wonderful internships over the years. I have been working at Microsoft Research Asia on modeling user clicks in web search and online advertising. I also worked with Vanja Josifovski at Google on recommender systems, with Lin Xiao at Microsoft Research on optimization algorithms, and with the web search team of Baidu on burst detection.

Address: Gates 254    Email: zhangyuc (at) cs.stanford.edu    [Google Scholar]

 

Manuscripts

 

Convexified Convolutional Neural Networks [arXiv]

Y. Zhang, P. Liang, M. Wainwright

 

Learning Halfspaces and Neural Networks with Random Initialization [arXiv]

Y. Zhang, JD. Lee, M. Wainwright, MI. Jordan

 

Splash: User-friendly Programming Interface for Parallelizing Stochastic Algorithms [arXiv] [slides] [project website]

Y. Zhang, MI. Jordan

 

Optimal prediction for sparse linear models? Lower bounds for coordinate-separable M-estimators [arXiv]

Y. Zhang, M. Wainwright, MI. Jordan

 

On Bayes Risk Lower Bounds [arXiv]

X. Chen, A. Guntuboyina, Y. Zhang

 

Optimality Guarantees for Distributed Statistical Estimation [arXiv]

J. Duchi, MI. Jordan, M. Wainwright, Y. Zhang

 

Journal Publications

 

Spectral Methods meet EM: A Provably Optimal Algorithm for Crowdsourcing [pdf]

Y. Zhang , X. Chen, D. Zhou, MI. Jordan

Journal of Machine Learning Research

 

Divide and Conquer Kernel Ridge Regression: A Distributed Algorithm with Minimax Optimal Rates [pdf]

Y. Zhang , J. Duchi, M. Wainwright

Journal of Machine Learning Research

 

Communication-Efficient Algorithms for Statistical Optimization [pdf]

Y. Zhang , J. Duchi, M. Wainwright

Journal of Machine Learning Research

 

The Antimagicness of the Cartesian Product of Graphs [pdf]
Y. Zhang and X. Sun
Theoretical Computer Science

 

Conference Proceedings

 

Local Maxima in the Likelihood of Gaussian Mixture Models: Structural Results and Algorithmic Consequences [arXiv]

C. Jin, Y. Zhang, S. Balakrishnan, M. Wainwright, MI. Jordan

Neural Information Processing System (NIPS’16)

 

L1-regularized Neural Networks are Improperly Learnable in Polynomial Time [pdf] [arXiv] [slides]

Y. Zhang, JD. Lee, MI. Jordan

International Conference on Machine Learning (ICML'16)

 

Distributed Estimation of Generalized Matrix Rank: Efficient Algorithms and Lower Bounds [pdf] [arXiv] [slides]

Y. Zhang, M. Wainwright, MI. Jordan

International Conference on Machine Learning (ICML'15)

 

DiSCO: Communication-Efficient Distributed Optimization of Self-Concordant Loss [pdf] [arXiv] [slides]

Y. Zhang, L. Xiao

International Conference on Machine Learning (ICML'15)

 

Stochastic Primal-Dual Coordinate Method for Regularized Empirical Risk Minimization [pdf] [arXiv] [slides]

Y. Zhang, L. Xiao

International Conference on Machine Learning (ICML'15)

 

Spectral Methods meet EM: A Provably Optimal Algorithm for Crowdsourcing [pdf] [slides]

Y. Zhang, X. Chen, D. Zhou, MI. Jordan

Neural Information Processing System (NIPS’14)

 

Lower Bounds on the Performance of Polynomial-time Algorithms for Sparse Linear Regression [pdf] [arXiv] [slides]

Y. Zhang , M. Wainwright, MI. Jordan

Annual Conference on Learning Theory (COLT'14)

 

Taxonomy Discovery for Personalized Recommendation [pdf] [slides]

Y. Zhang A. Ahmed, V. Josifovski, A. Smola

ACM International Conference on Web Search and Data Mining (WSDM'14)

 

Information-theoretic Lower Bounds for Distributed Statistical Estimation with Communication Constraints [pdf] [slides]

Y. Zhang, J. Duchi, MI. Jordan, M. Wainwright

Neural Information Processing System (NIPS'13)

 

Divide and Conquer Kernel Ridge Regression [pdf] [slides]

Y. Zhang , J. Duchi, M. Wainwright

Annual Conference on Learning Theory (COLT'13)

 
Communication-Efficient Algorithms for Statistical Optimization [pdf]

Y. Zhang , J. Duchi, M. Wainwright

Neural Information Processing System (NIPS'12)

 

Understanding Click Noise: A Noise-aware Click Model for Web Search [pdf]

W. Chen, D. Wang, Y. Zhang, Q. Yang.

ACM International Conference on Web Search and Data Mining (WSDM'12)

 

User-click Modeling for Understanding and Predicting Search-behavior [pdf]

Y. Zhang, W. Chen, D. Wang, Q. Yang.

ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD'11)

 

Characterize Search Intent Diversity into Click Models [pdf]

B. Hu, Y. Zhang, W. Chen, G. Wang, Q. Yang.  

International World Wide Web Conference (WWW'11)

 

Learning Click Model via Probit Bayesian Inference [pdf]

Y. Zhang, D. Wang, G. Wang, W. Chen, Z. Zhang, B. Hu, L. Zhang.

ACM Conference on Information and Knowledge Management (CIKM'10)


Extracting independent rules: a new perspective of boosting [pdf]

Y. Zhang, L. Zhang.

International Symposium on Artificial Intelligence and Mathematics (ISAIM'10).

 

Explore click models for search ranking [pdf]

D. Wang, W. Chen, G. Wang, Y. Zhang, B. Hu.

ACM Conference on Information and Knowledge Management (CIKM'10)

 

Incorporating Post-Click Behaviors Into a Click Model [pdf]
F. Zhong, D. Wang, G. Wang, W. Chen, Y. Zhang, Z. Chen, H. Wang

ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR'10)