InfoGram and Admissible Machine Learning
It addresses the need for regulatory-compliant ML in high-stakes domains, representing an incremental advancement in trustworthy AI.
The paper tackles the problem of making machine learning algorithms deployable under regulatory constraints by introducing an information-theoretic framework and tools like InfoGram to redesign methods for compliance while maintaining accuracy, as demonstrated with real-data examples from finance, biomedicine, marketing, and criminal justice.
We have entered a new era of machine learning (ML), where the most accurate algorithm with superior predictive power may not even be deployable, unless it is admissible under the regulatory constraints. This has led to great interest in developing fair, transparent and trustworthy ML methods. The purpose of this article is to introduce a new information-theoretic learning framework (admissible machine learning) and algorithmic risk-management tools (InfoGram, L-features, ALFA-testing) that can guide an analyst to redesign off-the-shelf ML methods to be regulatory compliant, while maintaining good prediction accuracy. We have illustrated our approach using several real-data examples from financial sectors, biomedical research, marketing campaigns, and the criminal justice system.