The Cambridge Ellis Unit Summer School on Probabilistic Machine Learning is from 17-21 July 2023 at the Department of Computer Science and Technology

The Cambridge Ellis Unit Summer School on Probabilistic Machine Learning is a distinguished course offered to graduate students, researchers and professionals, featuring engaging experts in their respective field and/or world-recognized professionals speaking about advanced machine learning concepts.

Important Dates

Admission notification by: 15 May 2023

Admission notification by: 5 June 2023

We can no longer accept applications

Topics

  • Statistical Emulation
  • Uncertainty Quantification
  • Evaluation of Probablistic Models
  • Re-inforcement Learning
  • Application RL
  • Inference/PN
  • Application BO
  • Introduction to Diffusion Models
  • Normalising Flows
  • Deep Generative Models
  • Application of Diffusion Models
  • Variational Inference and Stein Discrepancy
  • Probablistic Models in Computer Vision and Graphics
  • Machine Learning and the Physical World

Lecturers (Confirmed)

Neill Campbell
Marc Deisenroth
David Ginsbourger
José Miguel Hernández Lobato
Yingzhen Li
Katja Hofmann
Henry Moss
Nicholas Krämer
Tony O'Hagan
Ian Osband
Carl Rasmussen
Arno Solin
Mike Tipping
Rich Turner
Mark van der Wilk
Francisco Vargas
Jonathan Wenger
Tian Xie

Schedule

Please see the proposed schedule below.

Items followed by * are arranged by the organisers but need to paid for separately.

Dates
M- 17 July 2023
T- 18 July 2023
W- 19 July 2023
Th- 20 July 2023
F- 21 July 2023

Themes

Introduction to Probablistic Modeling

Probabilistic Models/Sequential Decision Making

Probabilistic Numerics

Implicit Models/Diffusion Models

Further Probabilistic Modeling

09:00

Mark van der Wilk

Henry Moss- BayesOpt and Beyond: Optimization of Expensive Functions using Gaussian Processes

Rich Turner- Neural Processes for Environmental Research

09:30

10:00

Break

10:30

Break

Break

Break

Break

11:30

12:00

Lunch

Lunch

12:30

Lunch

Lunch

Tian Xie- Challenges and opportunities in accelerating materials design with geometric deep learning and generative models

Marc Deisenroth

13:30

14:00

Lunch

15:00

Break

Break

Neill Campbell- Probabilistic Generative and Compositional Models

16:00

16:30

17:15

Science Tour- 90 min*

19:00-22:00

Evening Dinner at Sidney Sussex College*

Venue

The Summer School will be in person and held at the Department of Computer Science and Technology.

You can see travel information here.

This is an inperson event but we will record talks when we can and put on our Youtube channel.

Fees

All attendees will need to cover their own accommodation and travel costs. 

Travel awards are available for attendees from under-represented backgrounds. Those selected to attend will then be given a chance to apply for the travel grant. Please email ellis-admin@eng.cam.ac.uk for more information.

There are no fees to attend the Summer School.

Lunch and tea/coffee will be provided.

Apply

Those wishing to attend the Cambridge Ellis Machine Learning Summer School will need to complete this form.

You will also have to supply:

  • 2 page CV
  • Letter of Reference: The writer should assess the qualities, characteristics, and capabilities of the person being recommended in terms of that individual’s abilities. The letter should address the applicant’s background and potential in Machine Learning, academic standing compared to other students, and how he or she would benefit from attending Cambridge Ellis Machine Learning Summer School. The referee should be a person that has some experience working together with the applicant. It can be for example a Ph.D. supervisor, a former employer or manager. 

Organisers

José Miguel Hernández-Lobato
Kim Cole
Carl Henrick Ek
Carl Rasmussen

Sponsors