Robustness in ML
The robustness of a model is a measure of its stability with respect to perturbations of the input. We investigate and review recent developments in the field affecting the training, evaluation and deployment of ML systems.

All content in this series

Pill

Robust Solutions for Multi-Defender Stackelberg Security Games

Classical formulations of Multi-defender Stackelberg security games have the unpleasant property that the solutions are highly sensitive to …

A picture of Faried
Robustness in ML
Pill

Variational- and Metric-based Deep Latent Space for Out-of-Distribution Detection

A new method for detection of out-of-distribution samples in classification with neural networks which does not require OOD training data.

A picture of Miguel
Robustness in ML
Pill

A Fine-Grained Analysis on Distribution Shift

The topic of how different models behave under distribution shift is studied experimentally, after a framework for quantifying distribution …

A picture of Michael
Robustness in ML
Pill

Neural Collapse in deep classifiers during Terminal Phase of Training

Neural collapse refers to the observation that the last two layers of neural networks that were trained for a long time take a very simple, …

A picture of Michael
Robustness in ML
Pill

Boosting Certified Robustness of L Infinity Distance Nets

Recent architectural improvements for neural network classifiers make them certifiably robust against certain types of adversarial attacks - …

A picture of Miguel
Robustness in ML
Blog

Certified error rates for neural networks

Adversarial training has been shown to improve robustness of neural networks to certain classes of data perturbations. Despite constant …

A picture of Miguel
Robustness in ML

Other series in Safety and reliability in ML systems

Anomaly detection

Many industrial applications of automated decision-making involve the detection of anomalous behaviour. Precisely defining what this means is a complex problem hindered by strong class imbalances, lack of useful models or changing …

6 items

Formal verification under uncertainty

Being able to ensure that AI systems behave as intended is essential for their adoption in many areas. This series covers formal methods for the verification of safety, fairness, and other relevant properties of (probabilistic) ML systems. …

3 items

Safety in learned controllers

Safety constraints for learned controllers with known dynamics.

1 item
Check all of our work