## Research

Solving machine learning problems using neural networks.

Learn more.

Developing a new learned representation for complex data.

Learn more.

Data representations using graphs and their applications.

Learn more.

Learning models to deal with consecutive/temporal data.

Learn more.

The estimation of values for incomplete data in a matrix.

Learn more.

The decomposition of a matrix into a product of multiple matrixes.

Learn more.

Learning a distance function over data.

Learn more.

The fusion of various modalities of data into a single representation for effective use.

Learn more.

Supervised learning where data has several labels instead of a single one.

Learn more.

Determining optimal points in non-smooth non-convex objective functions.

Learn more.

We replace squared l_2 (Frobenius) norm with others such as l_1, l_infinity, p-order l_2, and others

Learn more.

Machine learning models that induce sparsity to help us determine critical features in data.

Learn more.

Transfering knowledge between machine learning models.

Learn more.