Chris Maddison

Faculty Member

Assistant Professor, Department of Computer Science, University of Toronto

Assistant Professor, Department of Statistical Sciences, University of Toronto

Canada CIFAR Artificial Intelligence Chair

Chris’s goal is to understand and improve the algorithms that agents can use to learn from data and reason about their experience. This goal is typically framed through the language of statistics, and solved using algorithms for probabilistic inference or optimization. Chris has worked on gradient estimation, variational inference, optimization, and Monte Carlo methods. He is currently interested in learning useful and robust representations, with a particular interest in techniques that learn representations from pretext tasks.  Chris is an Open Philanthropy AI Fellow. Chris received a NIPS Best Paper Award in 2014, and was one of the founding members of the AlphaGo project.

Research Interests

  • Representation Learning
  • Discrete Search
  • Optimization & Inference

Highlights

  • Open Philanthropy AI Fellow
  • NIPS Best Paper Award in 2014
  • Founding members of the AlphaGo project

Publications

Conditional neural processes

Marta Garnelo and Dan Rosenbaum and Chris J Maddison and Tiago Ramalho and David Saxton and Murray Shanahan and Yee Whye Teh and Danilo J Rezende and SM Eslami

2018

On empirical comparisons of optimizers for deep learning

Dami Choi and Christopher J Shallue and Zachary Nado and Jaehoon Lee and Chris J Maddison and George E Dahl

2019

Tighter variational bounds are not necessarily better

Tom Rainforth and Adam R Kosiorek and Tuan Anh Le and Chris J Maddison and Maximilian Igl and Frank Wood and Yee Whye Teh

2017

Doubly reparameterized gradient estimators for Monte Carlo objectives

George Tucker and Dieterich Lawson and Shixiang Gu and Chris J Maddison

2018

Gradient Estimation with Stochastic Softmax Tricks

Max B Paulus and Dami Choi and Daniel Tarlow and Andreas Krause and Chris J Maddison

202

Hamiltonian descent for composite objectives

Brendan O'Donoghue and Chris J Maddison

2019

Hamiltonian descent methods

Chris J Maddison and Daniel Paulin and Yee Whye Teh and Brendan O'Donoghue and Arnaud Doucet

2018

Oops I Took A Gradient: Scalable Sampling for Discrete Distributions

Will Grathwohl and Kevin Swersky and Milad Hashemi and David Duvenaud and Chris J Maddison

2018

Lossy Compression for Lossless Prediction

Yann Dubois and Benjamin Bloem-Reddy and Karen Ullrich and Chris J Maddison

2021

Optimal Representations for Covariate Shift

Yangjun Ruan and Yann Dubois and Chris J Maddison

2022

Improving Lossless Compression Rates via Monte Carlo Bits-Back Coding

Yangjun Ruan and Karen Ullrich and Daniel Severo and James Townsend and Ashish Khisti and Arnaud Doucet and Alireza Makhzani and Chris J Maddison

2021

Selecting actions to be performed by a reinforcement learning agent using tree search

Thore Kurt Hartwig Graepel and Shih-Chieh Huang and David Silver and Arthur Clement Guez and Laurent Sifre and Ilya Sutskever and Christopher Maddison

2021

Twisted variational sequential monte carlo

Dieterich Lawson and George Tucker and Christian A Naesseth and Chris Maddison and Ryan P Adams and Yee Whye Teh

2018

Dual space preconditioning for gradient descent

Chris J Maddison and Daniel Paulin and Yee Whye Teh and Arnaud Doucet

2021

Learning to cut by looking ahead: Cutting plane selection via imitation learning

Max B Paulus and Giulia Zarpellon and Andreas Krause and Laurent Charlin and Chris Maddison

2022

Learning Generalized Gumbel-max Causal Mechanisms

Guy Lorberbom and Daniel Johnson and Chris J Maddison and Daniel Tarlow and Tamir Hazan

2021

Augment with Care: Contrastive Learning for the Boolean Satisfiability Problem

Haonan Duan and Pashootan Vaezipoor and Max B Paulus and Yangjun Ruan and Chris J Maddison

2022

Learning Branching Heuristics for Propositional Model Counting

Pashootan Vaezipoor and Gil Lederman and Yuhuai Wu and Chris J Maddison and Roger Grosse and Edward Lee and Sanjit A Seshia and Fahiem Bacchus

2021

Direct policy gradients: Direct optimization of policies in discrete action spaces

Guy Lorberbom and Chris J Maddison and Nicolas Heess and Tamir Hazan and Daniel Tarlow

2020

Neural network systems implementing conditional neural processes for efficient learning

Tiago Miguel Sargento Pires Ramalho and Dan Rosenbaum and Marta Garnelo and Christopher Maddison and Seyed Mohammadali Eslami and Yee Whye Teh and Danilo Jimenez Rezende

2021

The machine learning for combinatorial optimization competition (ml4co): Results and insights

Maxime Gasse and Simon Bowly and Quentin Cappart and Jonas Charfreitag and Laurent Charlin and Didier Chételat and Antonia Chmiela and Justin Dumouchelle and Ambros Gleixner and Aleksandr M Kazachkov and Elias Khalil and Pawel Lichocki and Andrea Lodi and Miles Lubin and Chris J Maddison and Morris Christopher and Dimitri J Papageorgiou and Augustin Parjadis and Sebastian Pokutta and Antoine Prouvost and Lara Scavuzzo and Giulia Zarpellon and Linxin Yang and Sha Lai and Akang Wang and Xiaodong Luo and Xiang Zhou and Haohan Huang and Shengcheng Shao and Yuanming Zhu and Dong Zhang and Tao Quan and Zixuan Cao and Yang Xu and Zhewei Huang and Shuchang Zhou and Chen Binbin and He Minggui and Hao Hao and Zhang Zhiyu and An Zhiwu and Mao Kun

2022

Unbiased gradient estimation with balanced assignments for mixtures of experts

Wouter Kool and Chris J Maddison and Andriy Mnih

2021

Contrastive Learning Can Find An Optimal Basis For Approximately View-Invariant Functions

Daniel D Johnson and Ayoub El Hanchi and Chris J Maddison

2023

Bayesian nonparametrics for offline skill discovery

Valentin Villecroze and Harry Braviner and Panteha Naderian and Chris Maddison and Gabriel Loaiza-Ganem

2022

Stochastic Reweighted Gradient Descent

Ayoub El Hanchi and David Stephens and Chris Maddison

2022

Learning to Extend Program Graphs to Work-in-Progress Code

Xuechen Li and Chris J Maddison and Daniel Tarlow

2021

Training a policy neural network and a value neural network

Thore Kurt Hartwig Graepel and Shih-Chieh Huang and David Silver and Arthur Clement Guez and Laurent Sifre and Ilya Sutskever and Christopher Maddison

2018

Rao-Blackwellizing the Straight-Through Gumbel-Softmax Gradient Estimator

Max B Paulus and Chris J Maddison and Andreas Krause

2021

Probabilistic invariant learning with randomized linear classifiers

Leonardo Cotta and Gal Yehuda and Assaf Schuster and Chris J Maddison

2023

Identifying the risks of lm agents with an lm-emulated sandbox

Yangjun Ruan and Honghua Dong and Andrew Wang and Silviu Pitis and Yongchao Zhou and Jimmy Ba and Yann Dubois and Chris J Maddison and Tatsunori Hashimoto

2024