Generative models and uncertainty quantification lie at the heart of Bayesian modelling and inference. At this small meeting, we discuss recent developments within the field. The meeting is deliberately kept small in order to ensure that discussion remains honest, lively and interesting. Attendance is, thus, mostly by invitation, but one can apply to join (see below).

Confirmed Speakers

Max Welling

Max Welling
Professor in Machine Learning at the University of Amsterdam and a VP Technologies at Qualcomm.

Aki Vehtari

Aki Vehtari
Associate professor in computational probabilistic modeling at Aalto University; visiting professor at Technical University of Denmark (DTU).

Will Grathwohl

Will Grathwohl
PhD student at the University of Toronto.

Mark van der Wilk

Mark van der Wilk
Machine Learning Researcher PROWLER.io.

José Miguel Hernández Lobato

José Miguel Hernández Lobato
University Lecturer in Machine Learning, University of Cambridge.

Rianne van den Berg

Rianne van den Berg
Research Scientist at Google AI, Amsterdam.

Sebastian Nowozin

Sebastian Nowozin
Machine learning researcher in the Brain team at Google AI, Berlin.

Daniel Hernández-Lobato

Daniel Hernández-Lobato
Lecturer of Computer Science at Universidad Autónoma de Madrid.

Andriy Mnih

Andriy Mnih
Research scientist at DeepMind.

Tom Rainforth

Tom Rainforth
Researcher in statistical machine learning at the University of Oxford.

Casper Kaae Sønderby

Casper Kaae Sønderby
Research Scientist in Google Brains Amsterdam Lab.

Julie Josse

Julie Josse
Professor of Statistics at Ecole Polytechnique (CMAP) and XPOP INRIA team.

Silvia Chiappa

Silvia Chiappa
Staff Research Scientist in Machine Learning at DeepMind.

Program

Wednesday (Oct 9) Thursday (Oct 10)
Session 1
Advances in deep generative models
Chair: Søren Hauberg
Session 3
Bayesian neural networks
Chair: Wouter Boomsma
8:55-9:00 Opening remarks
9:00-10:00 Max Welling
The Functional Process VAE
09:00-09:40 Sebastian Nowozin
Practical Bayesian Neural Networks: Three Obstacles
10:00-10:40 Andriy Mnih
Resampled Priors for Variational Autoencoders
09:40-10:40 Casper Kaae Sønderby
Bayesian Inference for Large Scale Image Classification
10:40-11:00 Coffee 10:40-11:00 Coffee
11:00-11:40 Rianne van den Berg
Normalizing flows for discrete data
Session 4
Generative models for downstream tasks
Chair: Ole Winther
11:40-12:40 Tom Rainforth
Disentangling Disentanglement in Variational Autoencoders
11:00-12:00 José Miguel Hernández-Lobato
Advances in Compression via Probabilistic Machine Learning
Session 2
Bayesian and causal inference
Chair: Pierre-Alexandre Mattei
12:00-13:00 Lunch
12:40-13:40 Lunch 13:00-14:00 Mark van der Wilk
Sampling for data augmentation: generative models vs invariances
13:40-14:40 Aki Vehtari
Bayesian workflow
14:00-15:00 Will Grathwohl
The many virtues of Incorporating energy-based generative models into discriminative learning
14:40-15:00 Coffee 14:40-15:00 Coffee
15:00-15:40 Daniel Hernández-Lobato
Adversarial Alpha Divergence Minimization for Bayesian Approximate Inference
15:40-16:20 Silvia Chiappa
Causal inference and fairness
16:20-18:00 Poster Session
18:30-22:00 Conference Dinner
at Toldboden (directions)

Registration

If you would like to attend this meeting, but have yet to recieve an invitation, then fill out the following Google form. Note that seating is limited, so we cannot guarantee a ticket.

Venue

The workshop takes place at:

Pier47
Langelinie Allé 47
2100 Copenhagen Ø, Denmark

Pier47

Organizers

The conference is jointly organized by:

We are grateful for funding from the Villum Foundation (grant 15334) and the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement number 757360).