Tag Archives: randomized optimization

[Best student paper award] Welcome to my “Divide-and-Conquer Anchoring (DCA)” talk at ICDM Dallas Dec 8

Is it possible to finish a 60000×10000 matrix decomposition (NMF, PCA, etc)  or completion in 6 seconds on your laptop’s matlab? Can we make it even faster by a simple distributable scheme? How to summarize a huge-scale dataset (ratings, movie, … Continue reading

Posted in Tianyi's work | Tagged , , , , , , | Leave a comment

Semi-Soft GoDec: >4 times faster, auto-determined k

Here is a good news of GoDec (pertaining to our ICML 2011 paper): Semi-Soft GoDec is released. Different from the ordinary GoDec which imposes hard threshholding to both the singular values of the low-rank part L and the entries of the … Continue reading

Posted in Tianyi's work | Tagged , , , , , , , | 2 Comments

News about GoDec code and ICML 2011 paper

We recently published a google site for Go Decomposition (GoDec), presented on ICML 2011. Now you can find all the available information and upcoming news about GoDec on http://sites.google.com/site/godecomposition  On the new site, there are 3 resources about GoDec we … Continue reading

Posted in Tianyi's work | Tagged , , , , , , , | 5 Comments

GoDec: Randomized Low-rank & Sparse Matrix Decomposition in Noisy Case

This paper is accepted by ICML 2011 for presentation. Now the final version is ready and can be downloaded from here GO. Abstract: Low-rank and sparse structures have been profoundly studied in matrix completion and compressed sensing. In this paper, we develop … Continue reading

Posted in Tianyi's work | Tagged , , , , , , , , | 2 Comments