Explainable AI
Large opaque models like neural networks require dedicated methods to study and interpret their behavior. In this series we review recent developments, analysing their relevance for business applications.

All content in this series

Pill

Mixture of Decision Trees for Interpretable Machine Learning

A linear gating function together with multiple expert decision trees is trained by expectation-maximization and results in a new, fully …

A picture of Michael
Explainable AI
Pill

Tesselation Filtering ReLu Networks

An analysis of the shape complexity of the decision surface of a ReLU neural network. A framework for decomposing a network along a specific …

A picture of Faried
Explainable AI
Training

Methods and issues in explainable AI

A two-day workshop for ML practitioners wishing to make their models better understandable for themselves and decision makers.

Pill

Towards relatable explainable AI

A novel contrastive explanation method for neural network classifiers is proposed and applied in the challenging realm of audio prediction. …

A picture of Faried
Explainable AI
Pill

Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)

A non-intrusive and user-input based method for interpreting decision-making of trained neural networks is introduced. It is cheap and has …

A picture of Michael
Explainable AI

Other series in Trustworthy and interpretable ML

Classifier calibration

For many applications of probabilistic classifiers it is important that the predicted confidence vectors reflect true probabilities (one says that the classifier is calibrated). Recently, it has been shown that common models fail to satisfy …

7 items

Bayesian ML and Probabilistic Programming

When the understanding of phenomena and interpretation of results are required in industrial or scientific applications, black-box models give way to hand-crafted (or automatically discovered) probabilistic models. Classical Bayesian …

10 items

Simulation-based Inference

Computer simulations provide high-fidelity models and allow to generate insights without the need of running real experiments. They are especially useful in areas where real experiments would be too costly or even impossible. However, …

4 items
Check all of our work