Home

Crash Lecture Series.

The University of Applied Sciences of Western Switzerland and University of Bordeaux are happy to present this lecture series as a preview of a longer course that may start in Autumn 2021. This course can be included to the AIDA curriculum for the International Artificial Intelligence Doctoral Academy.

Note: the course timing may be subject to variations.

Crash course Overview:

The “where and why” of interpretability. Reading of taxonomy papers and perspectives.

The “three dimensions” of interpretability. Activation maximizations, LIME surrogates, Class Activation maps and Concept Attribution methods seen in details with hands-on exercises.

Evaluation of interpretability methods. Reading of papers and hands-on exercises.

Prerequisites:

Confidence in linear algebra, probability, machine learning. Experience with Python, numpy, tensorflow.

Lecture 1: The Where and Why of Interpretability

Prof. Henning Müller and Mara Graziani (PhD student), Hes-so Valais and University of Geneva

(postponed)

Lect. 2: From attention models and saliency maps to the explainability of Deep Neural Networks. 

Prof. Jenny Benois-Pineau, University of Bordeaux

(postponed)

Lect. 3: The three dimensions of interpretability. Activation Maximization and Feature Attribution

Mara Graziani (PhD student), Hes-so Valais and University of Geneva

(postponed)

Lect. 4:  Concept-based interpretability

Mara Graziani (PhD student), Hes-so Valais and University of Geneva

(postponed)

Lect. 5: Evaluating your interpretability experiments

Mara Graziani (PhD student), Hes-so Valais and University of Geneva

(postponed)

This course is funded by AI4Media, a centre of excellence delivering next generation AI research and training at the service of the media, the society and democracy. The course is open to all and has no registration fees. It is part of the AIDA PhD programme and advertised by the Swiss AI Meetup community.

Crea il tuo sito web con WordPress.com
Crea il tuo sito