Project Name: Human-Compatible Artificial Intelligence with Guarantees
Open Call Topic(s): Sustainable digitalisation
Description
In this proposal, we address the matter of transparency and explainability of AI using approaches inspired by control theory. Notably, we consider a comprehensive and flexible certification of properties of AI pipelines, certain closed loops and more complicated interconnections. At one extreme, one could consider risk-averse a priori guarantees via hard constraints on certain bias measures in the training process. At the other extreme, one could consider nuanced communication of the exact tradeoffs involved in AI pipeline choices and their effect on industrial and bias outcomes, post hoc. Both extremes offer little in terms of optimizing the pipeline and inflexibility in explaining the pipeline’s fairness-related qualities. Seeking the middle ground, we suggest a priori certification of fairness-related qualities in AI pipelines via modular compositions of pre-processing, training, inference, and post-processing steps with certain properties. Furthermore, we present an extensive programme in explainability of fairness-related qualities. We seek to inform both the developer and the user thoroughly in regard to the possible algorithmic choices and their expected effects. Overall, this will effectively support the development of AI pipelines with guaranteed levels of performance, explained clearly. Three use cases (in Human Resources automation, Financial Technology, and Advertising) will be used to assess the effectiveness of our approaches.
Reason for applying to HSbooster.eu services
Our project's stated impacts include "policy impact", which we believe would be best addressed by standardizing some of the outputs in standards mandated by the EC.
Main Standardisation Interests
Auditing of AI Fairness will be standardized. The European Commission (DC Connect) has issued a mandate for standardization, but the details are not completely clear, yet.
In the proposal, this concerns mostly WP3 (bias detection), WP4 (post-hoc guarantees of fairness), and WP5 (a priori guarantees of fairness). The activities are starting, so it's good to plan, but we also have some preliminary drafts we could share.
