Machine Learning Scientist currently working in Visa Inc. USA. I completed my PhD from Rutgers University, New Brunswick in 2022. My research interests lie in the field of statistical machine learning, graph machine learning, distributed optimization and signal processing.
Working on feature learning and dimension reduction techniques for big data in distributed settings. Developing a distrbuted training system for autoencoder for feature learning. Focus is to develop efficient algorithms as well as analyze their mathematical convergence guarantees.
Assisted in teaching the course and conducting labs for the courses Digital Signal Processing and Principle of Electrical Engineering.
Used dictionary learning and deep learning techniques on the single channel audio source separation problem.
Assisted in teaching the course and conducting labs for the courses Principle of Communication Systems and Modeling Complex Systems.
Department of Electrical and Computer Engineering.
Research Area: Machine Learning and Distributed Optimization.
Courses: Linear Algebra, Machine Learning, Information Theory and Coding, Stochastic Signals and Systems, Real Analysis, Optimum Signal Processing, Quantum Information Science.
GPA: 3.92/4
Department of Electronics and Communication Engineering.
Major: Communication and Signal Processing.
Courses: Convex Optimization, Probability and Random Processes, Statistical Signal Processing, Compressive Sensing, Stochastic Estimation and Control.
GPA: 9.44/10
Department of Electronics and Communication Engineering.
GPA: 9.16/10
Cleaned a given heterogenous dataset consisting of categorical, ordinal, binary, textual features.
Feature Engineering to make the data usual in machine learning algorithms.
Built a whole system consisting of various modules like classfication, regression etc. to predict the missing features in the data.
Part of an autonomous vehicle project.
Used LIDAR and image data to detect lanes for road navigation.
Used dictionary learning and orthogonal matching pursuit for recovering EEG signals from partially sampled data.
Developed a text-based speaker verification model that can be used for systems that need user authentication .