Welcome!
I am currently working as postdoctoral research associate at the University of Manchester. where I am working on probabilistic modeling for human-in-the-loop under the supervision of Prof. Samuel Kaski.
I have completed my Ph.D. from the Department of Computer Science at Indian Institute of Technology, Hyderabad, under the supervision of Dr P.K Srijith and Dr. Maunendra Sankar Desakar. You can find my CV here for more details.
My research focuses on developing machine learning models and algorithms for probabilistic analysis of social network dynamics. I am exploring across various domains like event modeling, spatio-temporal modeling, graph neural network, social networks and point process.
News
Our work "Bayesian Neural Hawkes Process" is accepted at JDSA, 2023.
"Time-to-Event Modeling with Hypernetwork based Hawkes Process" accepted at KDD 2023.
Attended ELLIS Summer School on ML for Healthcare at Manchester.
Attended London Symposium on Information Theory at UCL.
Joined as postdoctoral associate at the University of Manchester.
Successfully defended thesis titled "Improving Temporal and Spatial Event Modeling using Hawkes Process and Neural Networks".
Presented a talk on 'Continual Learning for Time-to-Event Modeling' in Workshop on Continual Lifelong Learning in ACML 2022.
Our work on Hawkes Process for discriminative classification is accepted at IJCNN, 2022. Arxiv is available here.
Honored to serve as Web Chair at ACML, 2022
Awarded Intel India Research Fellowship, 2021
Multi-view Hypergraph Convolution Network for Semantic Annotation in LBSNs accepted at ASONAM 2021
Presenting poster at WiML, NeurIPS, 2020
Work on semantic annotations in LBSNs using spatio-temporal Hawkes process accepted at SIGSPATIAL, 2020. The Arxiv version of paper is here.
Attended Microsoft Academic Research Summit 2020 and Amazon Research Days, 2019
Grateful to attend ACM India Grad Cohort, 2019 at IIT Delhi
Won second prize in Honeywell AI and ML Hackathon, 2019.
ML for social cause: During Kerala floods, we used our expertise to extract information regarding need and availability of various resources. For more details, please visit here.
Work on predicting best answerers in community question answering sites accepted at RecSys, 2018.