Disha Shrivastava

Hello everyone! I am Disha Shrivastava, a fifth-year PhD student in Machine Learning at Mila, working with Hugo Larochelle and Danny Tarlow. I worked at the Learning for Code team at Google Brain, Montreal as a Student Researcher from August 2019 to June 2022. I also worked as a Research Scientist Intern at DeepMind London with Yujia Li and David Choi. I am interested in developing deep learning models for understanding and generating programs by making use of information such as local and global structure, execution-guided program states and human feedback. My research interests include designing assistive tools for code-autocompletion, program synthesis, incorporating domain-specific context in large language models and reasoning for complex question-answering tasks. I am also interested in meta-learning and out-of-distribution generalization. Prior to starting my PhD, I worked at IBM Research, India as a Research Software Engineer. My work focussed on unsupervised construction of knowledge graphs, metrics for computational creativity and topical coherence, and reasoning for maths question-answering. I graduated from IIT Delhi with a Masters in Computer Technology where I developed a data and model-parallel framework for training of deep networks in Apache Spark. I hold a Bachelors degree in Electronics and Communication Engineering from BIT Mesra. I co-organized the Deep Learning for Code Workshops at ICLR 2022-23 and AIPLANS Workshop at NeurIPS 2021. I started the ML4Code Reading Group at Mila. My hobbies are singing, reading books and occasional blogging!