← Search

John S Shawe-Taylor

2 accepted papers

2018

Empirical Risk Minimization Under Fairness Constraints

NeurIPS 2018poster

We address the problem of algorithmic fairness: ensuring that sensitive information does not unfairly influence the outcome of a classifier. We present an approach based on empirical risk minimization, which incorporates a fairness constraint into the learning problem. It encourages the conditional…

2018

PAC-Bayes bounds for stable algorithms with instance-dependent priors

NeurIPS 2018poster

PAC-Bayes bounds have been proposed to get risk estimates based on a training sample. In this paper the PAC-Bayes approach is combined with stability of the hypothesis learned by a Hilbert space valued algorithm. The PAC-Bayes setting is used with a Gaussian prior centered at the expected output. Th…

Cited by 66SourcePDFScholar