Chaitanya K. Joshi
Research Engineer at A*STAR, Singapore
Publications
I am broadly interested in Deep Learning: my recent work focuses on Graph Neural Networks and their applications to Combinatorial Optimization. Previously, I worked on Attention & Memory Networks for Natural Langauge Processing.
Visit my Google Scholar profile for up-to-date papers and citations.
2020
Learning TSP Requires Rethinking Generalization
Chaitanya K. Joshi, Quentin Cappart, Louis-Martin Rousseau, Thomas Laurent, and Xavier Bresson
ArXiv preprint, 2020 (Under review)
tl;dr What architectures, inductive biases and learning paradigms enable zero-shot generalization to large-scale instances in learning-driven TSP solvers?
Benchmarking Graph Neural Networks
Vijay Prakash Dwivedi*, Chaitanya K. Joshi*, Thomas Laurent, Yoshua Bengio, and Xavier Bresson
ArXiv preprint, 2020 (Under review), *Equal contribution
tl;dr Open-source benchmarking framework to identify scalable and powerful GNN architectures, and track the progress of graph representation learning.
2019
Multi-Graph Transformer for Free-Hand Sketch Recognition
Peng Xu, Chaitanya K. Joshi, and Xavier Bresson
ArXiv preprint, 2019 (Under review)
tl;dr Representation learning for free-hand drawings using GNNs and Transformers.
On Learning Paradigms for the Travelling Salesman Problem
Chaitanya K. Joshi, Thomas Laurent, and Xavier Bresson
NeurIPS 2019, Graph Representation Learning Workshop
tl;dr How do learning paradigms impact zero-shot generalization to large-scale instances in learning-driven TSP solvers?
An Efficient Graph ConvNet for the Travelling Salesman Problem
Chaitanya K. Joshi, Thomas Laurent, and Xavier Bresson
INFORMS Annual Meeting, 2019 (Invited Talk); ArXiv preprint, 2019
tl;dr Deep Graph ConvNets paired with parallelized graph search can learn TSP up to few hundred cities, but fall short of classical solvers.
Working Women and Caste in India: A Study of Social Disadvantage using Feature Attribution
Kuhu Joshi and Chaitanya K. Joshi
ICLR 2019, AI for Social Good Workshop
tl;dr Feature attribution to interpret decision tree models and study the generational imact of caste on women’s work force participation in India.
Graph ConvNets for the Travelling Salesman Problem
Chaitanya K. Joshi
Bachelor’s Thesis, Nanyang Technological University, Singapore
(
Best Final Year Thesis Gold Medal, 2018-19)
2018
Question-answering and Chatbots using Memory Networks
Chaitanya K. Joshi
Book chapter, Hands-On NLP with Python (R. Arumugam and R. Shanmugamani, eds.), pp. 175-199, Packt Publishers, 2018
2017
Personalization in Goal-oriented Dialog
Chaitanya K. Joshi, Fei Mi, and Boi Faltings
NeurIPS 2017, Conversational AI Workshop
tl;dr New datasets and Memory Network architectures for studying personalized dialog systems based on user profiles.