Priyank Jaini

| CV | Google Scholar |
| LinkedIn | Github | Twitter |

I am a Research Scientist in the Brain team at Google Research in Toronto.

Previously, I was a post-doctoral researcher hosted by Max Welling at the University of Amsterdam, where I was part of AMLAB and the UvA-Bosch Delta Lab. Before moving to Amsterdam, I completed my PhD at the University of Waterloo and Vector Institute under the supervision of Pascal Poupart and Yaoliang Yu. In (seemingly) another life, I completed my undergraduate at IIT-Kanpur (Indian Institute of Technology, Kanpur) where I studied Mathematics and Statistics.

I host weekly office hours on Monday through the ML Collective initiative. If you'd like to talk about career directions, industry job market, graduate school, or research on probabilistic modeling, please reserve a spot here.

I am interested in machine learning and optimization with a particular focus on building tractable probabilistic models for reasoning under uncertainty. Most of my research has focussed on this through the lens of a variety of models including classical methods like probabilistic graphical models, mixture models, and Bayesian inference. Recently, my research has mainly focussed on developing flexible, expressive, and efficient generative models through works on Normalizing Flows, Energy Based Models, Diffusion Models, and Stein Variational Gradient Descent. I am particularly interested in incorporating inductive biases in the form of symmetries through equivariances in probabilistic modelling and applying to downstream tasks like molecular generation and modelling many-body particle systems.


Learning Equivariant Energy Based Models
with Equivariant Stein Variational Gradient Descent

Priyank Jaini*, Lars Holdijk*, Max Welling
NeurIPS 2021

Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
Emiel Hoogeboom*, Didrik Nielsen*, Priyank Jaini, Patrick Forré, Max Welling
NeurIPS 2021
Paper | Code

Self-Normalizing Flows
Thomas Andy Keller, Jorn Peters, Priyank Jaini, Emiel Hoogeboom, Patrick Forré, Max Welling
ICML 2021  (Spotlight)
Paper | Code

Sampling in Combinatorial Spaces with SurVAE Flow Augmented MCMC
Priyank Jaini, Didrik Nielsen, Max Welling
Paper | Code

SurVAE Flows: Surjections to Bridge the Gap between VAEs and Flows
Didrik Nielsen, Priyank Jaini, Emiel Hoogeboom, Ole Winther, Max Welling
NeurIPS 2020  (Long Oral Presentation)
Paper | Code | Video

Tails of Lipschitz Triangular Flows
Priyank Jaini, Ivan Kobyzev, Marcus Brubaker, Yaoliang Yu
ICML 2020

Learning Directed Acyclic Graph SPNs in Sub-Quadratic Time
Amur Ghose, Priyank Jaini, Pascal Poupart,
International Journal of Approximate Reasoning (IJAR) 2020

A Positivstellensatz for Conditional SAGE Signomials
Allen Wang, Priyank Jaini, Pascal Poupart, Yaoliang Yu

Sum-of-Squares Polynomial Flows
Priyank Jaini, Kira A. Selby, Yaoliang Yu
ICML 2019  (Long Oral Presentation)

Deep Homogeneous Mixture Models: Representation, Separation, and Approximation
Priyank Jaini, Pascal Poupart, Yaoliang Yu
NeurIPS 2018
Paper | Video

Prometheus: Directly Learning Acyclic Directed Graph Structures for Sum-Product Networks
Priyank Jaini, Amur Ghose, Pascal Poupart
Probabilistic Graphical Models (PGM) 2018

Online Bayesian Transfer Learning for Sequential Data Modelling
Priyank Jaini, Zhitang Chen, Pabla Carbajal*, Edith Law*, Laura Middleton*,
Kayla Regan*, Mike Schaekermann*, James Tung*, Pascal Poupart
* helped with data collection
ICLR 2017

Accuracy Maximization Analysis for Natural Tasks and Principles of Multiplicative Noise and Filter Correlation in Neural Coding
Johannes Burge, Priyank Jaini
PLoS Computational Biology 2017
Paper | Code

Linking Normative Models of Natural Tasks to Descriptive Models of Neural Response
Priyank Jaini, Johannes Burge
Journal of Vision (JoV) 2017
Paper | Code

Online and Distributed Learning of Gaussian Mixture Models by Bayesian Moment Matching
Priyank Jaini, Pascal Poupart
Approximate Bayesian Inference Workshop (AABI) 2017

Online Algorithms for Sum-Product Networks with Continuous Variables
Priyank Jaini, Abdullah Rashwan, Han Zhao, Yue Liu, Ershad Banijamali Liu, Zhitang Chen, Pascal Poupart
Probabilistic Graphical Model (PGM) 2016

  Selected Awards
  • Doctoral Dissertation Award, University of Waterloo (2020)
  • Borealis AI Graduate Fellowship (2018-2019)
  • Huawei Graduate Scholarship in Artificial Intelliegnce (2018-2019)
  • MITACS Accelerate Graduate Research Grant (2019)
  • Cheriton Graduate Scholarship, University of Waterloo (2017-2019)
  • Huawei Noah's Ark Lab Distinguished Collaborator Award (2016)
  • Graduate Excellence Award, University of Waterloo (2016)

Template taken shamelessly from here