I am a PhD candidate in the Intelligent Sensory Information Systems group at the University of Amsterdam, advised by Prof. Marcel Worring. My current research focus is representation learning on graph-structured data, in particular multimodal learning using hypergraphs.
I have worked on developing graph based deep learning models that are scalable to datasets with many modalities, with applications in recommedner systems, link predictions and neuroimaging. I am also part of the ASGARD project, whose aim is to develop AI powered solutions for forensic investigations that can directly contribute towards the Law Enforcement Agencies (LEAs) technological autonomy across EU.
I graduated from Indian Institute of Technology, Kanpur in 2015 with a B.Tech-M.Tech dual degree in Electrical Engineering. My formal background is in speech signal processing. I worked on developing Voice Activity Detection (VAD) algorithms during my master’s thesis and during my internship at NTU Singapore.
PhD in Artifical Intelligence, 2017-Ongoing
Unioversity of Amsterdam
B.Tech-M.Tech, Electrical Engineering, 2010-2015
Indian Institute of Technology, Kanpur
[June 2020] Our paper titled Fusing Structural and Functional MRIs using Graph Convolutional Networks for Autism Classification has been accepted at Medical Imaging with Deep Learning (MIDL).
[April 2020] Presented our work on Learning Represetnations in Hypergraphs at ELLIS Workshop on Geometric and Relational Deep Learning .
[August 2019] Our paper titled HyperLearn A Distributed Approach for Representation Learning in Datasets With Many Modalities has been accepted as Brave New Idea at ACM Multimedia .