2021 Gene Golub

SIAM Summer School

Theory and Practice of Deep Learning

The eleventh Gene Golub SIAM Summer School will now take place virtually from 19 - 30 July 2021 and will be hosted by the African Institute for Mathematical Sciences (AIMS) South Africa in Muizenberg.

The focus of the school will be the theory, implementation, and application of deep learning based on neural nets with many layers. Students will learn the mathematical underpinnings of deep learning using a combination of functional analysis and optimization theory. They will be introduced to applications that include computer vision, nonlinear programming, and forecasting, in addition to attending lectures by practitioners of deep learning in industry.

The summer school will include an important computing component, in which students will implement deep learning algorithms primarily using Python with TensorFlow and Keras. High performance computing resources will be provided by the AIMS supercomputing facility.

Note: all applicants who have already been accepted can still attend, however, as this will now be held virtually additional places have been added but these are only available for students studying at African Institutions.


Additional places have been made available and applications are open for participants from African institutions only.


20 February 2021

Learn more:

SIAM Web Site

Contact us : siam2020@aims.ac.za

The summer school is being organised by:

AIMS South Africa

University of Oxford

Tufts University

The intended audience is graduate students, meaning anyone studying beyond a three- or four-year undergraduate degree. Applicants are expected to have done a significant amount of mathematics and some computing in their studies in order to have the required mathematical and computational background for the summer school. Ideal candidates will be working on a research project that requires the use of deep learning methods.

Applicants selected to participate will pay no registration fee.