I am a PhD Student in the Computer Science Department at the University of North Carolina at Chapel Hill. I am interested in Large Language Models for Decision Making, Reinforcement Learning, Active Learning, and AI for Health. I am focusing on a) developing architecture and methods to build a tabular foundational model. b) enhancing the performance of black-box decision-makers (LLMs) using ideas from Active Learning and Reinforcement Learning.
- Generative Models: Foundational models for tabular data, images, languages, sets, etc.
- Large Foundation Models: train and evaluate large language models and multi-modal models.
- Model-based Reinforcement Learning: leverage world-models and RL for planning problems.
- Active Feature Acquisition: acquire the most informative features for cost-sensitive problems.
Previously, I was a Research Intern at QTIM, Harvard University/MIT and MGH, where I developed deep learning models to predict gene expression of oncologic drivers of brain metastases from multi-sequence MRI, working with Prof. Jayashree Kalpathy-Cramer, Prof. Bruce Rosen, Prof. Albert Kim and Prof. Christopher Bridge.
As an undergraduate student at BITS Pilani Goa, I was affiliated with APPCAIR AI Labs where I worked with Prof. Ashwin Srinivasan on generating drug-like molecules for specific targets using Large Language Models and worked on reliable model compression. I also had the opportunity to work with Prof. Narendra Ahuja at the University of Illinois Urbana-Champaign on plant phenotype prediction.
News
Aug 2024 | Started PhD in Computer Science at UNC Chapel HillJuly 2024 | Presented my work on DL for brain metastases at ISBI 2024 in Athens, Greece
June 2024 | Completed B.E. Electronics and MSc. Biology with Minor in Data Science from BITS Pilani Goa
Feb 2024 | Drug Discovery paper accepted at AAAI 2024 main track
Publications / Preprints
Dynamic Information Sub-Selection for Decision Support
Hung-Tien Huang, Maxwell Lennon, , Sean Sylvia, Junier B Oliva
Under Review
preprint
Multimodal Deep Learning-Based Prediction of Immune Checkpoint Inhibitor Efficacy in Brain Metastases
Tobias R. Bodenmann, Nelson Gil, Felix J. Dorfner, Mason C. Cleveland, Jay B. Patel, , Melisa S. Guelen, Dagoberto Pulido-Arias, Jayashree Kalpathy-Cramer, Jean-Philippe Thiran, Bruce R. Rosen, Elizabeth Gerstner, Albert E. Kim, Christopher P. Bridge
CaPTion Workshop, MICCAI 2024
paper
Generating Novel Leads for Drug Discovery Using LLMs for Logical Feedback
, Ashwin Srinivasan, Tirtharaj Dash, Lovekesh Vig, Arijit Roy, Sowmya Krishnan, Raviprasad Aduri
Accepted at AAAI 2024 | AAAI Conference on Artificial Intelligence
paper
Deep Learning-based Non-Invasive Molecular Profiling of Brain Metastases from MR Imaging
, Tiago Goncalves*, Tobias R. Bodenmann, Syed Rakin Ahmed, Jay B. Patel, Praveer Singh, Katharina V. Hoebel, Mason C. Cleveland, Felix Dorfner, Dagoberto Pulido-Arias, Bruce R. Rosen, Jaime S. Cardoso, Jayashree Kalpathy-Cramer, Elizabeth Gerstner, Albert E. Kim, Christopher P. Bridge
(* = Equal Contribution)
ISBI 2024 | IEEE International Symposium on Biomedical Imaging
abstract
IKD+: Reliable Low Complexity Deep Models for Retinopathy Classification
, Rohit Rajesh, Tirtharaj Dash, Lovekesh Vig, Tanmay Tulsidas Verlekar, Md Mahmudul Hasan, Tariq Khan, Erik Meijering, Ashwin Srinivasan
IEEE ICIP 2023 | IEEE International Conference on Image Processing
paper
A Deep Learning Framework Enables Non-Invasive detection of Tumor Mutational Burden in Brain Metastases
Syed Rakin Ahmed*, , Christopher Bridge, Jay Patel, Ken Chang, Mishka Gidwani, Praveer Singh, Elizabeth Gerstner, Albert Kim, Priscilla Brastianos, Jayashree Kalpathy-Cramer
(* = Equal Contribution)
RSNA 2023 | Radiological Society of North America
abstract
Efficient Integration of Molecular Representation and Message-Passing Neural Networks for Predicting Small Molecule Drug-like Properties
, Mrunmay Mohan Shelar, Revanth Harinarthini, Hemanth Bandaru, Nahush Harihar Kumta, Ojas Wadhwani, Raviprasad Aduri
ICDD 2022 | International Conference on Drug Discovery
poster