Vinayak Gupta

ML Researcher
Lawrence Livermore National Lab

CV

Research Interests

  • LLMs for Sequential Data: Multi-Modal Sequences, Time Series Reasoning, Point Processes, and Recommendation Systems.

Education

Ph.D. in Machine Learning

Indian Institute of Technology, Delhi
2017 - 2022

B.Tech in Computer Science

Indian Institute of Information Technology, Jabalpur
2013 - 2017

Skills

  • Proficient: Python: Pytorch.
  • Basics: Tensorflow and MATLAB.

Teaching Experience

  • Information Retrieval, Machine Learning, Data Mining, Data Structures, and Computer Networks.

Service

  • PC Member: AAAI, IJCAI, WWW, WSDM, SIGIR, IEEE TPAMI, ACM TOIS, ICML Workshops, etc.

About Me

I am a Researcher in the AI research group (AIRG) at the Lawrence Livermore National Laboratory. I work on designing ML models for temporal sequences. Recently, I have been exploring large language models (LLMs) for modeling time-sensitive healthcare data. In the past, I have also delved into deep generative models and graph-based models for sequential data mining and recommendation systems.

Previously, I was a Postdoctoral Scholar in the Paul G. Allen School of Computer Science at the University of Washington where I worked with Tim Althoff. Even before, I was a Research Scientist at IBM Research, and completed my Ph.D. from the Department of Computer Science at the Indian Institute of Technology Delhi where I worked with Srikanta Bedathur and Abir De. Even before, I received my Bachelor's in CS from the Indian Institute of Information Technology (IIIT) Jabalpur. I have also worked as an Applied Scientist-Intern at Amazon, and as an R&D-Intern at Siemens.

Some News

  • [Aug '24] Joined LLNL as an ML Researcher!
  • [Jul '23] IIT Delhi's Nominee for SIGKDD and ACM India PhD Dissertation Awards.
  • [Apr '23] Joined University of Washington as a Postdoctoral Scholar!
  • [Oct '22] Featured as AI Expert in IndiaAI! By NASSCOM and Ministry of IT, Govt. of India.
  • [Jun '22] Adjudged as Challenger (Top 3) in NASSCOM AI Gamechangers of India 2022!
  • [Oct '21] Received the Outstanding Doctoral Paper award at AI-ML Systems 2021!

Few Publications (View All )

Are Language Models Actually Useful for Time Series Forecasting?
Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, and Tom Hartvigsen.
Neural Information Processing Systems (NeurIPS), 2024. (Spotlight Paper)
[Code]
Language Models Still Struggle to Zero-shot Reason about Time Series.
Mike A. Merrill, Mingtian Tan, Vinayak Gupta, Tom Hartvigsen, and Tim Althoff.
Findings of Emperical Methods in Natural Langauge Processing (EMNLP), 2024.
[Code]
Learning Temporal Point Processes for Efficient Retrieval of Continuous Time Event Sequences.
Vinayak Gupta, Srikanta Bedathur, and Abir De.
AAAI Conference on Artificial Intelligence (AAAI), 2022.
[Code]
ProActive: Self-Attentive Temporal Point Process Flows for Activity Sequences.
Vinayak Gupta and Srikanta Bedathur.
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2022.
[Code]
Learning Temporal Point Processes with Intermittent Observations.
Vinayak Gupta, Srikanta Bedathur, Sourangshu Bhattacharya, and Abir De.
International Conference on Artificial Intelligence and Statistics (AISTATS), 2021.
[Code]

About Me (The real one!)

Wow! You're still reading? Since you've come all the way here, let me tell you some interesting stuff about me. I grew up entirely in Delhi (the city of Smoke 😅) but was born in Allahabad (the city of Prime Ministers 😎). Before I went on with my seemingly never-ending degree-grabbing spree, I was a dorky student at DPS Indirapuram (a Dipsite!), an even dorkier student at Cambridge School Noida, and even dorkier at the school before (got the point?).

During my amazing four years as an undergrad, I was a member of the ABU Robocon team that was placed 9th all over India (our bots played badminton!), a rhythm guitarist in the institute rock band, a member of the counseling committee (never did anything here), and organizer of many unsung fests and clubs. I have, on occasion, also been known to do something useful there such as designing sensor network protocols and as an assistant for maintaining the networking and internet infra of the entire campus.