I am a final year BS-MS student at the Indian Institute of Science Education and Research (IISER), Pune, India. Currently, I am a Research Fellow at the International Institute of Information Technology (IIIT), Hyderabad advised by Professor CV Jawahar and Dr. U. Deva Priyakumar. I am a part of Healthcare with AI (HAI) team at IIIT.

I am interested in building systems that can understand and generate multimodal data like image+text, video+audio, etc. For integration of such systems in the real world, it is also important that the systems are interpretable and robust. This motivates me to work on interpretability and robustness of multimodal AI systems. My current research focuses on buliding interpretable multimodal systems for application in healthcare and drug generation & discovery.

Updates

• [Jan 2021] LigGPT: Molecular Generation using Transformer-Decoder Model shorter version accepted as oral presentation at AAAI-SDA 2021 workshop!
• [Jan 2021] LigGPT: Molecular Generation using Transformer-Decoder Model preprint ready!
• [Jan 2021] MMBERT got accepted to ISBI 2021!
• [Oct 2020] Submitted MMBERT: Multimodal BERT for Improved VQA to the International Symposium of Biomedical Imaging (ISBI) 2021.
• [May 2020] I will be joining Healthcare with AI (HAI) team at IIIT-Hyderabad as a Research Fellow!

Email | GitHub | Twitter | Resume | NLP Resume | CV Resume | LinkedIn

Achievements

• All India Rank 2302 in IIT-JEE Advanced 2016 among ~ 1.2 million candidates
• All Indian Rank 69 in KVPY 2016 conducted by the Indian Institute of Science, Bangalore. I am funded throughout my degree by KVPY organization.
• National top 1% in National Physics Graduate Examination (NGPE) 2019.
• 2 x Kaggle Expert. Only 8% of Kaggle users are Experts.
• 16th position (top 2%) in PANDA Competition on Kaggle among 1010 participants.
• Selected in Madhava Mathematics Competition 2018, conducted by TIFR, Mumbai.

Publications

	LigGPT: Molecular Generation using a Transfomer-Decoder Model Viraj Bagal, Rishal Agarwal and U. Deva Priyakumar Under review Paper | AAAI_Paper_Version | Code | Slides | Video Presentation
	MMBERT: Multimodal BERT for Improved Medical VQA Yash Khare*, Viraj Bagal*, Minesh Mathew, Adithi Devi, U. Deva Priyakumar and CV Jawahar International Symposium on Biomedical Imaging (ISBI), 2021 Paper | Code | Slides | Video Presentation

Personal Projects

	Keyword Spotting from Speech Data Code | Model Comparisons
	Text Classification using Graph Neural Networks Notebook | Model Comparisons
	Semi-supervised Ecommerce Product Matching Code
	Wheat Head Object Detection Code
	CycleGAN: From Photos to Monet paintings and viceversa Code
	Re-implementation of Ethan Harris et al. FMix: Enhancing Mixed Sample Data Augmentation Article | Code
	Re-implementation of Ekagra et al. Jointly Learning Convolutional Representations to Compress Radiological Images and Classify Thoracic Diseases in the Compressed Domain. ICVGIP 2018 Report | Code
	Prostate cANcer graDe Assessment (PANDA) Challenge, Kaggle Leaderboard: Stood 16th. Top 2%.