My Mission

To advance state-of-the-art methods in AI, and make it easy for people to develop AI tools and conduct research in the field.

About Me

Research Scientist & Adjunct Professor

Issam Laradji is a research scientist at ServiceNow Research and an Adjunct professor at University of British Columbia, specializing in building AI solutions that require minimal human effort. With a postdoc from McGill’s Graphics lab and a PhD from the University of British Columbia, his research interests include natural language processing, 2D/3D computer vision, and optimization. He is passionate about bringing together a community of AI enthusiasts and making AI accessible to everyone. In pursuit of this vision, he has created many libraries including Haven-AI in order to empower individuals to build AI solutions for a wide range of real-life applications and research problems.

My Research

WACV2020

Consistency-based Learning

A Weakly Supervised Consistency-based Learning Method for COVID-19 Segmentation in CT Images

Paper Code
ECCV2020

Embedding Propagation

Smoother Manifold for Few-Shot Classification

Paper Code
ECCV2018

Where are the blobs

Counting by Localization with Point Supervision

Paper Code
ICIP2020

WISE-Net

Instance Segmentation with Point Supervision

Paper Code
ICML2018 Workshop

M-ADDA

Unsupervised Domain Adaptation with Deep Metric Learning

Paper Code
IJCNN2019

GP-DRF

Efficient Deep Gaussian Process Models for Variable-Sized Input

Paper Code
ArXiv2017

Let's Make Block Coordinate Descent Go Fast

Faster Greedy Rules, Message-Passing, Active-Set Complexity, and Superlinear Convergence

Paper Code
BMVC2019

Where are the masks

Instance Segmentation with Image-level Supervision

Paper Code
NeurIPS2019

SLS

Painless Stochastic Gradient: Interpolation, Line-Search, and Convergence Rates

Paper Code
ECML2018

MASAGA

A Linearly-Convergent Stochastic First-Order Method for Optimization on Manifolds

Paper Code
UAI2016

Kaczmarz

Convergence rates for greedy Kaczmarz algorithms, and faster randomized Kaczmarz rules using the orthogonality graph

Paper Code
ICML2015

Coordinate Descent

Coordinate Descent Converges Faster with the Gauss-Southwell Rule Than Random Selection

Paper Code

Testimonies

"Works very fast and maintains world-class standards." - John

"Very collaborative." - Johnny

"Very positive." - Jackson