NLM DIR Seminar Schedule
UPCOMING SEMINARS
RECENT SEMINARS
-
May 2, 2025 Pascal Mutz
Characterization of covalently closed cirular RNAs detected in (meta)transcriptomic data -
May 2, 2025 Dr. Lang Wu
Integration of multi-omics data in epidemiologic research -
April 22, 2025 Stanley Liang, PhD
Large Vision Model for medical knowledge adaptation -
April 18, 2025 Valentina Boeva, Department of Computer Science, ETH Zurich
Decoding tumor heterogeneity: computational methods for scRNA-seq and spatial omics -
April 8, 2025 Jaya Srivastava
Leveraging a deep learning model to assess the impact of regulatory variants on traits and diseases
Scheduled Seminars on Oct. 31, 2023
Contact NLMDIRSeminarScheduling@mail.nih.gov with questions about this seminar.
Abstract:
Along with the dramatic growth of interest in artificial intelligence and deep learning is a growth in questions about such algorithms. For example, are they fair? A machine learning algorithm is behind PubMed search's Best Match algorithm. NIST's AI Risk Management Framework points out that fairness needs to be regularly measured and tracked across changes in algorithms. We measured fairness in PubMed search in the areas of article language and journal ranking. We also modified the search algorithm by changing which clicks are used to score articles and adding a dense retrieval feature. We measure the effect on fairness resulting from these changes. We conclude with a discussion of the implementation and implications of a common suggestion for balancing fairness and relevance of search results.