Conference: Enriching Statistical Inference with Artificial Intelligence

The conference "Enriching Statistical Inference with Artificial Intelligence" will be held at Purdue University May 12-14, 2025. During the past decade, deep learning has revolutionized data science, with transformative applications in fields such as computer vision, protein structure prediction, and natural language processing. These advancements underscore the immense potential of deep neural networks (DNNs) while revealing a gap in the statistical understanding of their mechanisms and successes. This conference seeks to address the gap by fostering a vibrant platform for exchanging ideas, advancing statistical theories to illuminate DNN performance, developing innovative artificial intelligence (AI) tools, and promoting interdisciplinary collaborations that harness the power of AI to solve real-world problems. This conference will significantly enrich statistical inference with AI, while also contributing to the evolution of AI by improving its robustness, interpretability, and uncertainty quantification. This improved inference will then impact society broadly by improving the experience of users interacting with the products of AI. This conference will delve into cutting-edge research at the intersection of AI and statistical inference. Key topics will include investigating fundamental phenomena in deep learning, such as benign overfitting, and developing robust methods for uncertainty quantification in DNN models. The conference will also emphasize practical applications, leveraging DNNs to address foundational scientific challenges like causal inference and variable selection in complex systems. Participants will acquire state-of-the-art AI tools to tackle the complexities of contemporary data science while contributing to the advancement of modern statistical theory. More information about this conference may be found at https://www.stat.purdue.edu/news/2024/bff9.html. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.