Graphs for Causal AI

Utkarshani Jaimini of the University of South Carolina’s Artificial Intelligence Institute explores how AI can move beyond correlation to better understand **cause and effect**. By combining knowledge graphs, neural networks, and domain expertise through **neurosymbolic AI**, her research aims to build more explainable and reliable AI systems for applications including healthcare and autonomous driving.

Guest

Utkarshani Jaimini: Utkarshani Jaimini is a Ph.D. candidate at the AI Institute, University of South Carolina. Her research focuses on improving explainability through causal knowledge graphs, with applications in autonomous driving, smart manufacturing, and healthcare. She, along with other co-organizers, organizes the workshop on causal NeSy AI at the top semantic web conference- Extended Semantic Web Conference (ESWC) 2025. She has contributed to tutorials, talks, and discussions at conferences like AAMAS, The Web Conference, Knoweldge Graph Conference (KGC) and International Semantic Web Conference (ISWC). She has taken on leadership roles that amplify her impact. As an area chair for Women in Machine Learning (WiML) at NeurIPS and a session chair at notable conferences, she has facilitated knowledge exchange and fostered inclusivity. Her proactive engagement with the broader AI community is evident in her involvement with diversity and inclusion initiatives such as the CRA Scholarship and the AnitaB Grace Hopper Celebration. Her contributions span academia and industry, emphasizing her commitment to fostering a diverse workforce, equitable academic environments, and bridging the gap between research and real-world applications