On our second meeting within the Trustworthiness Assessment with the Z-Inspection® methodology of the ExplainMe project, Jesmin Jahan Tithi provided an insightful talk about the phases of the Z-inspection process supported with illustrative examples and case studies.
What is an ethical tension?
How to measure the fairness of explanations?
What is the bias of experts conducting the assessment?
TWe started discussing first questions.
You can see the discussed slides HERE.
More insights about the Z-Inspection® methodology: https://z-inspection.org/).
Z-Inspection® is a holistic process to assess trustworthy AI.
For more details, see Z-Inspection®: A Process to Assess Trustworthy AI | IEEE Journals & Magazine | IEEE Xplore
Our aim is the co-design of a Trustworthy Explainable AI system for mental health monitoring using speech. The motivation behind ExplainMe stems from the observation that most state-of-the-art systems supporting remote mental health monitoring lack transparency in their reasoning and decision-making. At the same time, research confirms that acoustic features extracted from speech serve as valid markers for assessing the severity of manic and depressive symptoms. ExplainMe addressed this gap and aims to design an Explainable AI system for mental health monitoring using speech. Our assessment is co-lead by Jesmin Jahan Tithi, Ph.D, Hanna Sormunen and Megan Coffee with the support of Roberto V. Zicari.
The multi-disciplinary team has already grown to 31 experts, covering various disciplines, including computer science, explainable artificial intelligence, software engineering, psychiatry, law, ethics, philosophy, acoustic phonetics, and other areas of medicine and social sciences:
1 Katarzyna Kaczmarek-Majer (PI of ExplainMe project)
2 Jesmin Jahan Tithi (co-leader)
3 Megan Coffee (co-leader)
4 Hanna Sormunen (co-leader)
5 Małgorzata Sochacka
6 Anne Schwerk
7 Armin Grasnick
8 Manfred Jeusfeld 9 Marta Boratyn
10 Tana Pistorius
11 Filip Wichrowski
12 Leslie Salgado
13 Marcin Ostrowski
14 Adarsh Srivastava
15 Gokul Gawande
16 Markus Aniek
17 Jose Maria Alonso Moral
18 Valerie Bürger
19 Michał Chojnicki
20 Stella Namuganza
21 Martin Pereira
22 Roberto Zicari
23 Jose Maria Sousa
24 Marlena Sokół-Szawłowska
25 Katarzyna Chołdzyńska
26 Rupa Singh
27 Dorota Kamińska
28 Gabriella Casalino
29 Jacek Puczniewski
30 Vince Madai
31 Adarsh Srivastava
Big thank you to all experts for joining this assessment!
