Abriendo la caja negra de la IA: Avances en Inteligencia Artificial Explicable (XAI)
The thesis explores the black box problem in Artificial Intelligence (AI) and the need for more transparent and explainable AI systems. XAI seeks to open AI systems to scrutiny, making the decision-making process more understandable to users. This is crucial for building trust and ensuring fair and ethical implementation of AI across multiple sectors. The thesis hypothesizes that transparent AI models (white box) can be crucial in explaining more complex and opaque systems. The objectives include: 1) Understanding the XAI landscape, 2) Building interpretable models by design, 3) Generating post-hoc explanations for black box models, 4) Understanding the implications for industry.
keywords:
Publication: Thesis
1726561945535
September 17, 2024
/research/publications/abriendo-la-caja-negra-de-la-ia-avances-en-inteligencia-artificial-explicable-xai
The thesis explores the black box problem in Artificial Intelligence (AI) and the need for more transparent and explainable AI systems. XAI seeks to open AI systems to scrutiny, making the decision-making process more understandable to users. This is crucial for building trust and ensuring fair and ethical implementation of AI across multiple sectors. The thesis hypothesizes that transparent AI models (white box) can be crucial in explaining more complex and opaque systems. The objectives include: 1) Understanding the XAI landscape, 2) Building interpretable models by design, 3) Generating post-hoc explanations for black box models, 4) Understanding the implications for industry. - Ettore Mariotti - José María Alonso Moral
publications_en