Majid Janzamin UCI Wordmark


Publications


Note: authorship order of specified publications by asterisk symbol * follows the convention in theoretical computer science of alphabetical author order.

 

Preprints:

 

Journal Publications:

 

Conference Publications:

 

Workshop/Conference Presentations:

  • Generalization Bounds for Neural Networks through Tensor Factorization. ISMP (International Symposium on Optimization), Pittsburgh, PA, July 2015.

  • Score Function Features for Discriminative Learning. ICLR (International Conference on Learning Representations), San Diego, CA, May 2015.

  • Provable Learning of Overcomplete Latent Variable Models: Semi-supervised and Unsupervised Settings. NIPS: Workshop on Optimization for Machine Learning, Montreal, Canada, Dec. 2014.

  • Provable Learning of Overcomplete Latent Variable Models: Semi-supervised and Unsupervised Settings. New England Machine Learning Day, Microsoft Research, Cambridge, MA, May 2014.

  • Guaranteed Non-Orthogonal Tensor Decomposition through Alternating Rank-1 Updates. Workshop on Electrical Flows, Graph Laplacians, and Algorithms: Spectral Graph Theory and Beyond, ICERM, Brown University, Providence, RI, April 2014.

  • When are Overcomplete Topic Models Identifiable? Uniqueness of Tensor Tucker Decompositions with Structured Sparsity. GlobalSIP: Symposium on Optimization in Machine Learning and Signal Processing, Austin, TX, Dec. 2013.

  • High-Dimensional Covariance Decomposition into Sparse Markov and Independence Domains. 50th Annual Allerton Conference on Communication, Control, and Computing, Monticello, IL, Oct. 2012