Picture for Kristof Meding

Kristof Meding

Fairness Hacking: The Malicious Practice of Shrouding Unfairness in Algorithms

Add code
Nov 12, 2023
Viaarxiv icon

Trivial or impossible -- dichotomous data difficulty masks model differences (on ImageNet and beyond)

Add code
Oct 12, 2021
Figure 1 for Trivial or impossible -- dichotomous data difficulty masks model differences (on ImageNet and beyond)
Figure 2 for Trivial or impossible -- dichotomous data difficulty masks model differences (on ImageNet and beyond)
Figure 3 for Trivial or impossible -- dichotomous data difficulty masks model differences (on ImageNet and beyond)
Figure 4 for Trivial or impossible -- dichotomous data difficulty masks model differences (on ImageNet and beyond)
Viaarxiv icon

Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency

Add code
Jun 30, 2020
Figure 1 for Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency
Figure 2 for Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency
Figure 3 for Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency
Figure 4 for Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency
Viaarxiv icon

The Big Picture: Ethical Considerations and Statistical Analysis of Industry Involvement in Machine Learning Research

Add code
Jun 08, 2020
Figure 1 for The Big Picture: Ethical Considerations and Statistical Analysis of Industry Involvement in Machine Learning Research
Figure 2 for The Big Picture: Ethical Considerations and Statistical Analysis of Industry Involvement in Machine Learning Research
Figure 3 for The Big Picture: Ethical Considerations and Statistical Analysis of Industry Involvement in Machine Learning Research
Figure 4 for The Big Picture: Ethical Considerations and Statistical Analysis of Industry Involvement in Machine Learning Research
Viaarxiv icon