Graduate Student || The Pennsylvania State University || nud83 [at] psu.edu
I am currently an Informatics Ph.D. student at College of Information Sciences and Technology, The Pennsylvania State University
I am advised by Dr. C. Lee Giles. My advisory committee includes Dr. Dan Kifer and Dr. Ankur Mali
Keywords: Deep Learning, Machine Learning, Math Language Processing, Math Understanding in Neural Networks, Neuro-Symbolic AI, Explainability and Rules Extraction
Current Neural Network architectures are glorified state machines. They can solve problems equivalent to regular grammars. Any problem in context-free and counter languages can be solved using state machines when restricted in their input size. In my research, I study the learnability and stability of RNN and Transformer architectures on formal grammars like Tomita, Dyck, and $a^nb^nc^nd^n$, etc. My work models fundamental math problems like addition, multiplication, counting, arithmetic, and precision as strings of symbols and studies the symbol manipulation ability of neural networks.
Stability Analysis of Various Symbolic Rule Extraction Methods from Recurrent Neural Network
Neisarg Dave, Daniel Kifer, C. Lee Giles, Ankur Mali
ICLR 2024 Workshop on Bridging the Gap Between Practice and Theory in Deep Learning
Math Question Solving and MCQ Distractor Generation with attentional GRU Networks
Neisarg Dave, Riley Owen Bakes, Bart Pursel, C. Lee Giles
In Processings of Educational Data Mining, 2021.
BBookX: Creating Semi-Automated Textbooks to Support Student Learning and Decrease Student Costs
Pursel, Bart, Crystal M. Ramsay, Neisarg Dave, Chen Liang, and C. Lee Giles.
In iTextbooks@ AIED, pp. 81-86. 2019.
Distractor generation for multiple choice questions using learning to rank
Liang, Chen, Xiao Yang, Neisarg Dave, Drew Wham, Bart Pursel, and C. Lee Giles.
In Proceedings of the thirteenth workshop on innovative use of NLP for building educational applications, pp. 284-290. 2018.