The application of artificial intelligence/machine learning (AI/ML) methods to experimental and computational sciences, in order to enable autonomy and accelerate discovery, has become an extremely exciting and rapidly-evolving field. In this context, we felt it was time to bring together researchers in this field to discuss recent developments and share best practices. This workshop will combine invited talks, moderated breakouts, and tutorial sessions; thus, it will be of interest both to experts, and newcomers to the field. We welcome all researchers, including students and other early-career scientists.
As our scientific questions become increasingly more complex, the underlying parameter spaces we have to explore and understand are growing in size and dimensionality beyond what a human is able to comprehend. Fortunately, ML/AI methods have emerged that provide comprehensive insights into these high-dimensional spaces, allowing scientists with deep domain expertise to focus on interpretations and decisions that affect the larger scope of their work rather than micromanaging measurements. This workshop will present a comprehensive overview of the emerging field of Autonomous Discovery, covering new theoretical methods, real-world applications, and tutorials of selected software packages. Breakout sessions will give attendees the chance to get first-hand experience and to directly engage with experts.