I am an undergraduate student majoring in Math with a minor in Physics. I am interested in Scientific Machine learning. I am currently working on NeuralODEs and their application to sequence modeling(like LSTMs and Time-series forecasting). I work with black-box differential equation solvers for calculating the gradients without using backpropagation thereby eliminating the memory costs of large models. I am also interested in Information Theory and its applications to learning systems(like Deep neural nets). I wanted to work in the intersection of Stochastic Differential equations and Deep neural nets in the future. I am also interested to work on other statistical concepts like Optimal Transport etc. I am a self-taught programmer with a passion for programming systems that scale-up. The programming languages I currently work with are Python, Julia, C++, and some MATLAB. I am more into the mathematics of machine learning rather than production-based ML models. I wish to understand the concept of interpretability and causal inference for machine learning models(which means I am more interested in what the model is learning from data rather than how good it is learning from the given data).

One of my other interests is Deep learning for medical research (or biological sciences in general). I am currently working towards using Deep learning techniques from computer vision and segmentation models(like U-Nets) for analyzing the Brain tumor MRIs of Glioblastoma patients.