Explainable AI for the healthcare domain

Date: 28th February 2020

Time: 15:00-16:00

Location: IF 2.33

Talks

Title:  Explainable AI for the healthcare domain
Speaker: Jorge Gaete
Abstract:

Different machine learning (ML) techniques have been used in the healthcare domain for tasks such as prognosis and diagnostic  of diseases. But as tasks in the medical domain are typically critical, medical staff have the need to understand the ML model and its outcomes. This is a problem since ML techniques are generally black boxes, giving rise to the the topic of Explainable AI. In this first year review talk I will introduce the topic of XAI in the medical domain, some of the main approaches for XAI, and my current work in the topic. I will put special attention to the use of Bayesian methods for XAI. Finally I will present some of the future research directions and a work plan for the next year