Generating Grammar Rules for Syntax-Guided Synthesis using Deep Neural Network:

Conditional Linear Integer Arithmetic (CLIA) Track synthesizes function in Linear Integer Arithmetic (LIA) combined with the if-then-else construct (ite). The Specification is stated in the form of logical constraints for the SyGuS problem. Even though CVC4 solves most of the benchmark problems but the synthesis time is slow for some of the simple problems and even timeouts in some cases. In this project, we predicted the grammar rules for a given set of logical constraints to restrict the search space consequently improving the synthesis time.

Synthesizing Program Sketches from Logical Specifications using Machine Learning techniques:

This is a Research Project done during the course Program Synthesis meets Machine Learning (Jan'20, Jun'20). I worked in this project along with Stanly Samuel under the guidance of Prof. Deepak D'Souza, Prof. Chiranjib Bhattacharyya, and Dr. Sriram Rajamani (Microsoft).

Learning in Sparse Reward Environment:

This is a Survey Project done during the course Machine Learning (Jan'20, Jun'20). I worked in this project along with Mariamma Antony(PhD), Vivek Khandelwal(Masters), and Jagriti Singh(PhD) under the guidance of Shubham Gupta.