Electrical and Computer Engineering
menu MENU

Communications and Signal Processing Seminar

Seminar by Paramveer Dhillon

Paramveer DhillonAssistant ProfessorUniversity of Michigan, School of Information
WHERE:
1005 EECS BuildingMap
SHARE:

Abstract coming soon.

Biography

I am an Assistant Professor in the School of Information (SI) and Computer Science & Engineering (courtesy) at the University of Michigan, where I research and teach various topics in Artificial Intelligence (AI), broadly defined.

I got my AM in Statistics and PhD in Computer Science from the University of Pennsylvania where I was advised by Profs. Lyle UngarDean Foster (now at Amazon), and James Gee. During my time at Penn, I also worked closely with Dr. Brian Avants on topics related to Machine Learning in Brain Imaging. My PhD thesis on Advances in Spectral Learning with Applications to Text Analysis and Brain Imaging proposed novel statistical methods for problems in Text Modeling/NLP and Brain Imaging and was awarded the Morris and Dorothy Rubinoff Best Disseration Award. Specifically, it proposed statistically and computationally efficient methods for the problem of learning word embeddings in NLP and for the problem of data-driven parcellation/segmentation of human brain images. Our methods not only gave predictive accuracies that were better or comparable to the state-of-the-art statistical methods (circa 2015) but also had strong theoretical guarantees. Please look at our JMLR 2015 and NeuroImage 2014 papers for more details. I also did other research in my PhD on establishing connections between PCA and ridge regression (cf. JMLR 2013) and on provably faster row and column subsampling algorithms for least squares regression (cf. NeurIPS 2013a,b).

Towards the end of my PhD, I got interested in computational social science and causal inference. After finishing my PhD, I proceeded to complete a Postdoc with Prof. Sinan Aral at MIT. At MIT, I worked on several social science problems (e.g. finding influential individuals in a social network with realistic real-world assumptions (cf. Nature Human Behaviour 2018), devising revenue maximizing price discrimination strategies for newspapers, and designing sequential interventions for news websites to help them maintain sustained user engagement). At MIT, I was also involved with the Initiative on the Digital Economy on studying the economic and societal impacts of AI.

Sponsored by

ECEKLA

Organizer

Judi Jones(734) 763-8557

Faculty Host

Clay Scott