Search for: All records

A Workshop on the Ethical Design of AIs was convened in September and October 2022. Workshop participants hailed from a wide range of disciplines and application domains, and expressed interest in establishing partnerships across academia, industry, and government agencies, to address the challenges that were identified during the event. One of the outcomes of the workshop was a recommendation for a 2023 Convergence Accelerator Track on the Ethical Design of AIs (EDAIs). Suggested recommendations of themes and goals for the EDAIs Track include the following: (1) Human Centered Design methodologies around Values and Measures and Incentives. (2) Proto Ethical AIs: Algorithms or Systems or Pipelines across multiple domains. (3) Best Practices for the design of ethical AIs. (4) Workforce development and education and training. This report documents the activities of the EDAIs Workshop 

In this study, we propose a post-hoc ex- plainability framework for deep learning models applied to quasi-periodic biomedical time-series classification. As a case study, we focus on the problem of atrial fibrillation (AF) detection from electrocardiography signals, which has strong clinical relevance. Starting from a state-of-the-art pretrained model, we tackle the problem from two differ- ent perspectives: global and local explanation. With global explanation, we analyze the model behavior by looking at entire classes of data, showing which regions of the input repetitive patterns have the most influence for a specific outcome of the model. Our explanation results align with the expectations of clinical experts, showing that features crucial for AF detection contribute heavily to the final decision. These features include R-R interval regularity, absence of the P-wave or presence of electrical activity in the isoelectric period. On the other hand, with local explanation, we analyze specific input signals and model outcomes. We present a comprehensive analysis of the network facing different conditions, whether the model has correctly classified the input signal or not. This enables a deeper understanding of the network’s behavior, showing the most informative regions that trigger the classification decision and highlighting possible causes of misbehavior.