I am a research scientist at Vanderbilt University, USA. My research interests include multi-agent systems, robust machine learning, and decision-making under uncertainty. I am honored to be one of the recipients of the Google AI Impact Scholar Award (2021) for Social Good.

Prior to this, I was a Post-Doctoral Research Fellow at the Stanford Intelligent Systems Lab at Stanford University, USA, working under Prof. Mykel Kochenderfer. I was awarded the 2019 CARS post-doctoral fellowship by the Center of Automotive Research at Stanford (CARS). Before joining Stanford, I was a PhD student at Vanderbilt University’s Computational Economics Research Lab under Prof. Eugene Vorobeychik. My thesis was nominated for the Victor Lesser Distinguished Dissertation Award 2020.

I am particularly intersted in applying AI to real-world problems, especially ones that have societal impacts. Much of my disseration focussed on using AI to aid proactive emergency response. My current work focusses on designing decentralized decision-making systems for emergency response, understanding how crowdsourced reports can be used to infer emergency scenarios, optimizing para-transit routing, trying to understand how to tackle wildfires and heatwaves, and designing deep learning methods for aiding conservation. At Stanford, I was one of the mentors for Stanford’s CS+Social Good Impact Lab, helping undergraduate students use AI for social good. My research is supported by Cisco, National Science Foundation, Google AI for Social Good, and Microsoft AI for Earth.

If you are interested in any of these research areas or want to explore AI for social good, feel free to reach out. I am happy to chat, learn, help, and collaborate!


Updates

  • We are releasing WildfireDB, an open-source dataset with over 17 million data points that link wildfire occurrences with weather, topography, and other relevant features. See our Neurips 2021 (Track on Datasets and Benchmarks) paper here.
  • Our paper "Practitioner-Centric Approach for Early Incident Detection Using Crowdsourced Data for Emergency Services", in collaboration with George Mason University, has been accepted at ICDM 2021. See the paper here.
  • We gave a tutorial on multi-agent emergency response systems at IEEE SmartComp 2021. The videos of the talk are available on youtube.
  • Our paper "The Raptor Join Operator for Processing Big Raster + Vector Data", in collaboration with Stanford and UC Riverside, has been accepted at ACM SIGSPATIAL 2021. See the paper here.
  • Happy to receive a Microsoft AI for Earth grant to design deep learning methods for the conservation of monarch butterflies. See a pre-print of our paper here.
  • Our work on learning incident prediction models from sparse data for emergency response has been accepted at IEEE SmartComp 2021. See the paper here.
  • I am excited to be named one of Google AI's Impact Scholars for the social good program! Excited to work with Helpmum, based out of Nigeria, to create data-driven methods for vaccine delivery. Read more on the Google AI blog here.
  • Our work on Energy and Emission Prediction for Mixed-Vehicle Transit Fleets Using Multi-Taskand Inductive Transfer Learning has been accepted at ECML 2021.
  • I gave a talk at the Los Alamos National Lab's Seminar Series on "Multi-Agent Systems for Emergency Response." See the slides here.
  • I am co-organizing the Workshop on Data-Driven and Intelligent Cyber-Physical Systems (DI-CPS) at ACM CPS Week.
  • I am one of the PC members for the Autonomous Agents for Social Good Workshop at AAMAS-21! Consider attending if you are interested in how agent-based systems can help solve societal problems.
  • Our work on using hierarchical planning to solve large scale resource allocation problems formulated as multi-agent semi-Markovian decision processes has been accepted at ICCPS 2021. See the paper here
  • Our work on processing large scale vector and raster datasets has been accepted at ICDE 2021 (in collaboration with Stanford and UC Riverside). See the paper here here
  • We have created WildfireDB, the first open-source and comprehensive database with over 2 million data points, that links wildfire occurrences with relevant covariates extracted from satellite imagery. We presented details about the database at the AI for Earth Sciences workshop at NeurIPS 2020. See the paper here, access the database here, and see the spotlight talk here.
  • Our work on creating uncertainty aware wildfire management strategies and designing principled emergency response pipelines have been acepted to the AAAI Fall Symposium Series Workshop in AI for Social Good.
  • I gave a tutorial on Smart Emergency Response at the NSF Computational Sustainability Doctorial Symposium 2020 on October 18. See the talk here
  • I have been invited to be a PC member for AAMAS-21 and AAAI-21.
  • I gave a talk at the Utah Center for Data Science's Summer Seminar Series on the 19th of June about robust machine learning models and smart emergency response. See the poster here, and listen to the talk here
  • I gave a talk about how robust incident prediction can combat poaching at the Cambridge Environmental Data Science Group's AI4ER Seminar Series on the 30th of June. If you are interested in the intersection of data science and environment, listen to the talk here.
  • I am one of the PC members of the amazing Harvard CRCS workshop on AI for Social Good. Consider submitting a paper or attend to hear about how AI can transform the society. Find details here.
  • A pre-print of our survey paper on how AI can be used to aid Emergency Response in smart cities is out and feedback would be appreciated.
  • Our work on robust spatial-temporal incident prediction was accepted at the 36th Conference on Uncertainty in Artificial Intelligence (UAI 2020).
  • We are creating Stat Resp, an integrated open-source toolchain of statistical methods to aid emergency response.