290 citations · 1.2k across the 29 of their papers we have counts for
17 papers · 1 filter
Shapley Sets: Feature Attribution via Recursive Function Decomposition
Torty Sivill, Peter Flach
Despite their ubiquitous use, Shapley value feature attributions can be misleading due to feature interaction in both model and data. We propose an alternative attribution approach…
When the Ground Truth is not True: Modelling Human Biases in Temporal Annotations
Taku Yamagata, Emma L. Tonkin, Benjamin Arana Sanchez +5
In supervised learning, low quality annotations lead to poorly performing classification and detection models, while also rendering evaluation unreliable. This is particularly appa…
What and How of Machine Learning Transparency: Building Bespoke Explainability Tools with Interoperable Algorithmic Components
Kacper Sokol, Alexander Hepburn, Raul Santos-Rodriguez +1
Explainability techniques for data-driven predictive models based on artificial intelligence and machine learning algorithms allow us to better understand the operation of such sys…
FAT Forensics: A Python Toolbox for Implementing and Deploying Fairness, Accountability and Transparency Algorithms in Predictive Systems
Kacper Sokol, Alexander Hepburn, Rafael Poyiadzi +3
Predictive systems, in particular machine learning algorithms, can take important, and sometimes legally binding, decisions about our everyday life. In most cases, however, these s…
The Weak Supervision Landscape
Rafael Poyiadzi, Daniel Bacaicoa-Barber, Jesus Cid-Sueiro +3
Many ways of annotating a dataset for machine learning classification tasks that go beyond the usual class labels exist in practice. These are of interest as they can simplify or f…
Classifier Calibration: A survey on how to assess and improve predicted class probabilities
Telmo Silva Filho, Hao Song, Miquel Perello-Nieto +3
This paper provides both an introduction to and a detailed overview of the principles and practice of classifier calibration. A well-calibrated classifier correctly quantifies the…