Aditya Kapoor
Hello! I am Aditya and I am currently a Predoctoral Research Scholar at the Robotics and Automation group at TCS Research & Innovation. I received my B.E. in Computer Science from BITS Pilani, Goa. I have interned at Robocomp as a part of Google Summer Of Code 2021 with Dr.Luis Manso and did my undergraduate thesis at the Biorobotics Lab, CMU with Dr. Howie Choset.

I am the founder of StraysCue which is a software product of HyphenBridge Foundation, non-profit organization that works towards the welfare of urban stray animals. StraysCue's vision is to mitigate the internal management efficiency and increase public transparency of the organisations that facilitate the well-being of stray animals in metropolitan cities via its software product.

I am a firm beleiver of democratizing AI education and research. In the same spirit, I spend quite some time on side research ideas with enthusiastic individuals. I am also a core member of CORD AI serving the same mission to democratize AI research among young professionals across different colleges in India.
This information was last updated in June 2023.

CV   |   Google Scholar   |   GitHub   |   LinkedIn   |   Twitter

Research Motivation


TL;DR

I am broadly interested to do research in the applications of cognitive science in AI by studying intelligence in both humans and artificial agents. I believe that studying and emulating human intelligence is critical to build truly general-purpose intelligent systems that can autonomously acquire complex behaviors through learning via interacting with its environments and other agents in it.


I believe to work towards AGI, we need to focus on reproducing the fundamental mechanisms and properties og the human brain. It is these fundamental attributes and functions that give rise to so many emergent behaviors like creativity, social dynamics etc that make humans so unique and different from other not so intelligent life forms. I believe to get somewhere near to building AGI like beings, we as AI researchers should understand the the underlying concepts of the human brain via cognitive sciences, psychology and neuroscience. Reproducing or even abstracting these concepts via black box mechanisms can enable us to develop semi-autonomous or totally-autonomous agents/ products. It may also enable cognitive scientists to better understand the different plausible interactions between these abstractions that lead to even more complex behaviors or verify existing/ postulated theories. The interplay between the two fields can help us to understand previously unseen concepts and open up new areas of research. We never know, there might be a new and upcoming field "Artificial cognitive sciences" pretty soon.



I am also interested in addressing social challenges via my research and technological innovations.

Feel free to get in touch if you are interested in working with me via email at f20170963g [at] alumni [dot] bits-pilani [dot] ac [dot] in.

Publications


SocNavGym: A Reinforcement Learning Gym for Social Navigation
Aditya Kapoor*, Sushant Swamy*, Pilar Bachiller, Luis Mansor (* indicates equal contribution)
IEEE RO-MAN, 2023
PDF | Bibtex

Concept-based Anomaly Detection in Retail Stores for Automatic Correction using Mobile Robots
Aditya Kapoor, Vartika Sengar, Nijil George, Vighnesh Vatsal, Jayvardhana Gubbi, Balamuralidhar P, Arpan Pal
IEEE Systems, Man, and Cybernetics Society (SMCS) 2023
PDF | Bibtex

Learning Cooperative Multi-Agent Policies with Partial Reward Decoupling
Benjamin Freed*, Aditya Kapoor*, Ian Abraham, Jeff Schneider, Howie Choset (* indicates equal contribution)
IEEE RA-L, presented at ICRA 2022
PDF | Bibtex

Challenges in Applying Robotics to Retail Store Management
Vartika Sengar*, Aditya Kapoor*, Nijil George*, Vighnesh Vatsal*, Jayvardhana Gubbi, Balamuralidhar P, Arpan Pal (* indicates equal contribution)
ICRA 2022 Workshop - Challenges in Applying Academic Research to Real-World Robotics
PDF | BibTex

Auto-TransRL: Autonomous Composition of Vision Pipelines for Robotic Perception
Aditya Kapoor, Vartika Sengar, Nijil George, Vighnesh Vatsal, Jayvardhana Gubbi
ICRA 2022 Workshop - Robotic Perception and Mapping: Emerging Techniques
PDF | BibTex

Supervised category learning: When do participants use a partially diagnostic feature?
Sujith Thomas, Aditya Kapoor, Narayanan Srinivasan
CogSci 2021
PDF | BibTex

A toolkit to generate social navigation datasets
Rishabh Baghel, Aditya Kapoor, Pilar Bachiller, Ronit R Jorvekar, Daniel Rodriguez-Criado, Luis J Manso
Workshop of Physical Agents 2020
PDF | BibTex

Effect of a colour-based descriptor and stimuli presentation mode in unsupervised categorization
Sujith Thomas, Aditya Kapoor, Narayanan Srinivasan
CogSci 2020
PDF

Website template courtesy