Search Results for author: Sebastian Scher

Found 6 papers, 2 papers with code

A conceptual model for leaving the data-centric approach in machine learning

no code implementations7 Feb 2023 Sebastian Scher, Bernhard Geiger, Simone Kopeinik, Andreas Trügler, Dominik Kowald

For a long time, machine learning (ML) has been seen as the abstract problem of learning relationships from data independent of the surrounding settings.

Fairness

Modelling the long-term fairness dynamics of data-driven targeted help on job seekers

no code implementations17 Aug 2022 Sebastian Scher, Simone Kopeinik, Andreas Trügler, Dominik Kowald

We conclude that in order to quantify the trade-off correctly and to assess the long-term fairness effects of such a system in the real-world, careful modeling of the surrounding labor market is indispensable.

Attribute Fairness

Testing robustness of predictions of trained classifiers against naturally occurring perturbations

no code implementations21 Apr 2022 Sebastian Scher, Andreas Trügler

We show both theoretically and with empirical examples that a method based on counterfactuals that was previously proposed for this is insufficient, as it is not a valid metric for determining the robustness against perturbations that occur ``naturally'', outside specific adversarial attack scenarios.

Adversarial Attack Adversarial Robustness +2

Ensemble methods for neural network-based weather forecasts

1 code implementation13 Feb 2020 Sebastian Scher, Gabriele Messori

However, the skill of the neural network forecasts is systematically lower than that of state-of-the-art numerical weather prediction models.

Weather Forecasting

WeatherBench: A benchmark dataset for data-driven weather forecasting

3 code implementations2 Feb 2020 Stephan Rasp, Peter D. Dueben, Sebastian Scher, Jonathan A. Weyn, Soukayna Mouatadid, Nils Thuerey

Data-driven approaches, most prominently deep learning, have become powerful tools for prediction in many domains.

Weather Forecasting

Cannot find the paper you are looking for? You can Submit a new open access paper.