AIMMES 2024

Fairness and Bias in Artificial Intelligence continue to be a dynamic field of research, prompting multiple questions and concerns. This is especially pertinent in light of the swift progress in generative AI and the widespread integration of AI solutions across diverse sectors, ranging from finance, recruitment, security, health, to public administration. The intricate interplay between technological progress and ethical considerations has become more evident, necessitating a thoughtful exploration of the implications. Concurrently, the evolving landscape of regulations and policies, exemplified by the EU AI Act, further intensifies the debates and raises crucial questions.

The workshop aims to delve into these multifaceted dimensions, fostering dialogue and understanding around the challenges and opportunities presented by the intersection of AI, fairness, and bias in our rapidly evolving technological landscape. Our workshop welcomes in particular contributions around three main areas and several related topics:

• Explaining Bias
  • new approaches for understanding, visualizing and communicating bias in datasets and AI systems
  • domain-specific approaches and use cases of bias explainability
  • XAI tools and software
• Measuring Bias
  • novel AI bias and fairness definitions, human factors in AI bias
  • role of social sciences, law and humanities in bias definition
  • involving vulnerable and under-represented communities in bias definition
  • measuring and studying bias in multi-attribute and multimodal settings
• Mitigating Bias
  • methods for exploring trade-offs across multiple notions of fairness and accuracy
  • use of synthetic data for bias mitigation
  • fairness-aware ranking and recommendation
  • application-specific mitigation approaches for hiring, credit scoring, face recognition, etc.