I am generally interested in studying, developing, and applying mathematical tools and machine learning algorithms to various problems.

Fairness and Bias in Machine Learning

Machine learning, particularly automated decision-making, is used in many sensitive aspects of our society; however, very little is done that addresses the serious and critical issues (e.g. gender, racial, social bias) that could arise in the data and algorithms. My latest endeavors and interests lie in studying the fairness of widely used, simple machine learning algorithms and in developing fairness-promoting algorithms. 

My research is supported by UCLA Racial and Social Justice Seed Grant (co-PI), 2023-2024.

Matrix and Tensor Factorizations for Learning Tasks

Topic modeling is an unsupervised machine learning technique used to reveal hidden themes from large datasets. A popular technique for topic modeling that provides a low-rank approximation of a matrix is nonnegative matrix factorization (NMF). My interests lie in studying variations of NMF and their applications to different settings. 

Multi-modal tensor data arises in many settings and tensor decompositions have many applications in machine learning. I have a great interest in applications of tensor decompositions in machine learning and extending ideas from matrix numerical linear algebra to tensors.

Data Completion Techniques

Matrix completion is the task of imputing the missing entries of a partially observed matrix from a subset of known entries. Incomplete data is highly prevalent for a multitude of reasons. In today's data-driven world, data completion is essential, whether it is the main goal as in recommender systems or a pre-processing step for learning tasks. I have a general interest in matrix completion techniques and downstream tasks with incomplete data. My prior work presents an iterative method for low-rank matrix completion that takes into account sparsity-based structure in the missing entries. 

Dimensionality Reduction Techniques

Dimension reduction is the task of reducing the number of features in a dataset while best preserving certain aspects of the data. I have an interest in studying the fairness and unique properties of dimensionality reduction techniques.  My prior work seeks to understand the classical Multidimensional Scaling (MDS) algorithm, a popular technique for pattern recognition problems and a visualization technique.