Tag Archives: structured learning

Multi-task Copula – A semiparametric joint prediction model for multiple outputs with sparse graph structure

Our paper “Multi-task Copula by Sparse Graph Regression“ has been accepted by KDD 2014 this year. So we can talk at the conference which is at NYC, between August 24-27. Before that, let me introduce this new method. In summary, we tackle … Continue reading

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

Compressed Labeling: An important extension of Hamming Compressed Sensing; at NIPS now

We are just informed that our submission “Compressed Labeling (CL) on Distilled Labelsets (DL) for Multi-label Learning” is accepted by Machine Learning Journal (Springer). Online first PDF can be downloaded here. CL is an important application and extension of Hamming … 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

Multi-label Learning via Structured Decomposition and Group Sparsity

This paper is now available on arxiv: http://arxiv.org/abs/1103.0102 In multi-label learning, each sample is associated with several labels. Existing works indicate that exploring correlations between labels improve the prediction performance. However, embedding the label correlations into the training process significantly … Continue reading

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