National University of Singapore
Member since 2 years
Specialises In (based on submitted proposals)
I hold a Bachelor and Master of Statistics degree from Indian Statistical Institute and I am currently pursuing PhD in the field of deep learning at National university of Singapore. Prior to that, I spent 4 years as a data scientist working for Flipkart.com, where I developed neural network model behind Flipkart’s current demand forecasting engine (LSTM based mixture density network) which forecasts sale of millions of listed products of Flipkart with a very high degree of accuracy everyday. I specialize in machine learning, time series forecasting, semi-supervised learning, deep learning, computer vision etc. Recently I was involved in developing a deep learning model which will be able to predict glaucoma (a type of eye disease) from 3D OCT image of optical nerve head.
In my PhD I am exploring more fundamental aspects of AI, like how we can learn very good disentangled representation (lower dimensional) of high dimensional data (eg. word2vec), few keywords that describe my research is semi-supervised learning, disentanglement, multi task learning, generative models (Generative adversarial network, variational inference) etc. Also, I am mostly working with images (medical images) where labels are expensive to obtain.