UCL Centre for Artificial Intelligence


NeurIPS 2020

We are delighted to announce that there have been 29 papers accepted to NeurIPS 2020 this year by UCL researchers, many of whom are part of the AI Centre. Congratulations to all successful authors!

A Class of Algorithms for General Instrumental Variable Models
Niki Kilbertus (Helmholtz AI) · Matt Kusner (University College London) · Ricardo Silva (University College London)

Online Matrix Completion with Side Information
Mark Herbster (University College London) · Stephen Pasteris (University College London) · Fai Yu Lisa Tse (University College London)

Matern Gaussian processes on Riemannian manifolds
Viacheslav Borovitskiy (St. Petersburg Department of Steklov Mathematical Institute of Russian Academy of Sciences (PDMI RAS)) · Alexander Terenin (Petuum, Inc. and Imperial College London) · Peter Mostowsky (St. Petersburg State University) · Marc Deisenroth (University College London)

BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images
Thu Nguyen-Phuoc (University of Bath) · Christian Richardt (University of Bath) · Long Mai (Adobe Research) · Yongliang Yang (University of Bath) · Niloy Mitra (University College London)

Probabilistic Active Meta-Learning
Jean Kaddour (Imperial College London) · Steindor Saemundsson (Imperial College London) · Marc Deisenroth (University College London)

A Loss Function for Generative Neural Networks Based on Watson’s Perceptual Model
Steffen Czolbe (University of Copenhagen) · Oswin Krause (University of Copenhagen) · Ingemar Cox (University College London) · Christian Igel (University of Copenhagen)

A Non-Asymptotic Analysis for Stein Variational Gradient Descent
Anna Korba (Gatsby Unit - UCL) · Adil SALIM (KAUST) · Michael Arbel (UCL) · Giulia Luise (University College London) · Arthur Gretton (Gatsby Unit, UCL)

PAC-Bayesian Bound for the Conditional Value at Risk
Zakaria Mhammedi (The Australian National University and Data61) · Benjamin Guedj (Inria & University College London) · Robert Williamson (ANU)

Online Multitask Learning with Long-Term Memory
Mark Herbster (University College London) · Stephen Pasteris (University College London) · Fai Yu Lisa Tse (University College London)

The NetHack Learning Environment
Heinrich Küttler (Facebook AI Research) · Nantas Nardelli (University of Oxford) · Alexander Miller (Facebook AI Research) · Roberta Raileanu (NYU) · Marco Selvatici (Imperial College London) · Edward Grefenstette (DeepMind) · Tim Rocktäschel (University College London Facebook AI Research)

Barking up the right tree: an approach to search over molecule synthesis DAGs
John Bradshaw (University of Cambridge/MPI IS Tübingen) · Brooks Paige (University College London) · Matt Kusner (University College London) · Marwin Segler (BenevolentAI) · José Miguel Hernández-Lobato (University of Cambridge)

Learning Fair and Transferable Representations
Luca Oneto (University of Genoa) · Michele Donini (Amazon) · Giulia Luise (University College London) · Carlo Ciliberto (Imperial College London) · Massimiliano Pontil (IIT) · Andreas Maurer ()

Deep Reinforcement Learning with Stacked Hierarchical Attention for Text-based Games
Yunqiu Xu (University of Technology Sydney) · Meng Fang (Tencent) · Ling Chen (" University of Technology, Sydney, Australia") · Yali Du (University College London) · Joey Tianyi Zhou (IHPC, A*STAR) · Chengqi Zhang (University of Technology Sydney)

Replica-Exchange Nos\'e-Hoover Dynamics for Bayesian Learning on Large Datasets
Rui Luo (University College London) · Qiang Zhang (University College London) · Yaodong Yang (University College London) · Jun Wang (JD AI Research & UCL)

The Wasserstein Proximal Gradient Algorithm
Adil SALIM (KAUST) · Anna Korba (Gatsby Unit - UCL) · Giulia Luise (University College London)

Disentangling Human Error from Ground Truth in Segmentation of Medical Images
Le Zhang (University College London) · Ryutaro Tanno (Microsoft Research / UCL) · Moucheng Xu (University College London) · Chen Jin (University College London) · Joseph Jacob (University College London) · Olga Cicarrelli (Queen Square Multiple Sclerosis Centre) · Frederik Barkhof (University College London) · Daniel Alexander (University College London)

Goal-directed Generation of Discrete Structures with Conditional Generative Models
Maolaaisha Aminanmu (University of Geneva,University of Applied Sciences Western Switzerland) · Brooks Paige (University College London) · Alexandros Kalousis (University of Applied Sciences, Western Switzerland)

Deep active inference agents using Monte-Carlo methods
Zafeirios Fountas (University College London; Emotech Labs) · Noor Sajid (University College London) · Pedro Mediano (University of Cambridge ) · Karl Friston (University College London)

A kernel test for quasi-independence
Tamara Fernandez (University College London) · Wenkai Xu (Gatsby Unit, UCL) · Marc Ditzhaus (TU Dortmund University) · Arthur Gretton (Gatsby Unit, UCL)

Kernelized information bottleneck leads to biologically plausible 3-factor Hebbian learning in deep networks
Roman Pogodin (University College London) · Peter E Latham (Gatsby Unit, UCL)

On Second Order Behaviour in Augmented Neural ODEs
Alexander Norcliffe (University College London) · Cristian Bodnar (University of Cambridge) · Ben Day (University of Cambridge) · Nikola Simidjievski (University of Cambridge) · Pietro Lió (University of Cambridge)

The Advantage of Conditional Meta-Learning for Biased Regularization and Fine Tuning
Giulia Denevi (University College of London) · Massimiliano Pontil (IIT & UCL) · Carlo Ciliberto (Imperial College London)

Organizing recurrent network dynamics by task-computation to enable continual learning Lea
Duncker (Gatsby Unit, UCL) · Laura N Driscoll (Stanford) · Krishna V Shenoy (Stanford University) · Maneesh Sahani (Gatsby Unit, UCL) · David Sussillo (Stanford University)


Non-reversible Gaussian processes for identifying latent dynamical structure in neural data Virginia
Rutten (Gatsby Computational Neuroscience Unit (UCL)) · Alberto Bernacchia (MediaTek Research) · Maneesh Sahani (Gatsby Unit,

UCL) · Guillaume Hennequin (Cambridge)  [ORAL]

ICE-BeeM: Identifiable Conditional Energy-Based Deep Models Based on Nonlinear ICA Ilyes
Khemakhem (Gatsby Unit, UCL) · Ricardo Monti (Gatsby Unit, UCL) · Diederik P. Kingma (Google) · Aapo Hyvarinen (University of


COT-GAN: Generating Sequential Data via Causal Optimal Transport
Tianlin Xu (London School of Economics and Political Science) · Wenliang Li (Gatsby Unit, UCL) · Michael Munn (Google) · Beatrice Acciaio (London School of Economics)

The interplay between randomness and structure during learning in RNNs
Friedrich Schuessler (Technion) · Francesca Mastrogiuseppe (Gatsby Unit, UCL) · Alexis Dubreuil (ENS) · Srdjan Ostojic (Ecole Normale

Superieure) · Omri Barak (Technion - Israeli institute of technology)

PAC-Bayes Analysis Beyond the Usual
Bounds Omar Rivasplata (DeepMind & UCL) · Ilja Kuzborskij (DeepMind) · Csaba Szepesvari (DeepMind / University of Alberta) · John Shawe-Taylor


RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces
Sebastien Ehrhardt (University of Oxford) · Oliver Groth (Oxford Robotics Institute) · Aron Monszpart (Niantic) · Martin Engelcke (University of Oxford) · Ingmar Posner (Oxford University) · Niloy Mitra (UCL/Adobe) · Andrea Vedaldi (Facebook AI Research and University of Oxford)