Probabilistic Concept Bottleneck Models

Tutorial to be held at UAI 2026
Amsterdam, Netherlands
August 17, 2026

You can find our tutorial slides here (pdf).

Schedule

Below is the expected (rough) schedule for this tutorial where we indicate next to each section who will be presenting that section’s material (GM for “Giuseppe Marra”, PB for “Pietro Barbiero”, and DD for “David Debot”).

Part Topic Presenter Main reference
Part 1 Interpretability Theory PB The Standard Interpretable Model
Part 2 Concept representations GM What's in the Bottle? A Survey and Roadmap of Concept Bottleneck Models
Part 3 Concept-based task predictions DD From statistical relational to neurosymbolic artificial intelligence: A survey
Part 4 PyTorch Concepts PB PyC: interpretable models in PyTorch

Required Background

Our material will assume a basic knowledge of ML (e.g., foundations of supervised learning, experimental design, basic probabilistic modelling, etc.), with particular emphasis on a solid Deep Learning foundation (e.g., tensor calculus, neural networks, backpropagation, etc.). Concepts that may require mathematical tools/expertise beyond those one would expect to be shared among the AI community will be (re)introduced in our tutorial.

Additional Material

If we get access to a recording of our presentation, we will include it in this section as soon as possible.

For a complete bibliography of the topics and works discussed in this tutorial, please refer to our resources section.

Presenters

Citing This Tutorial

If you found this tutorial useful for your research, blogs, or work, please cite it as follows:

Marra G., Barbiero P., and Debot D. (2026). Probabilistic Concept Bottleneck Models. Tutorial at the Conference on Uncertainty in Artificial Intelligence (UAI). https://interpretabledeeplearning.github.io/

Or use the following bibtex entry:

@misc{interpretabledl2026essai,
  title        = {Probabilistic Concept Bottleneck Models},
  author       = {Marra G. and Barbiero P. and Debot D.},
  year         = {2026},
  howpublished = {Conference on Uncertainty in Artificial Intelligence (UAI)},
  url          = {https://interpretabledeeplearning.github.io/}
}