Fostering Collaborative Breakthroughs in Heritage Science through Machine Learning and Data Science

This proposed Experts Meeting intends to discuss new and emerging opportunities in the application of artificial intelligence, machine learning, data science, and visualization to current/emerging problems in Heritage Science. The panel of ~30 experts will span several important fields within the broader context of &heritage science and data science (AI/ML; Data Science; Imaging Science; …) in order to discuss and distill the following topics: ? Deep Learning Frameworks: What is the promise of Artificial Intelligence (e.g., deep learning) as a new framework for inquiry into problems in cultural heritage and with heritage materials/collections? Generalized Advances: How can researchers in machine learning, through applications in and data from heritage science, develop generalized breakthroughs and insights , such as methods for transfer learning, multi-modal learning, learning with background knowledge, and unsupervised learning? Data Science and imaging: what imaging modalities and other types of data measurements are showing the most promise in Heritage Science, and at what scale can they be applied? Visualization and analysis: where are the gaps, opportunities, and tools/techniques for the visualization and analysis of heritage data? Collections: Which types of heritage collections (e.g., large corpora of text; image collections of manuscripts) represent the frontier of challenges for machine learning, artificial intelligence, and data science with an associated likelihood of new discoveries and ensuing breakthrough results? Bridging Scholarly Cultures: How can academic fields/cultures become more closely aligned, encouraging more scholars experienced with heritage materials to actively engage with researchers who are technical specialists and vice versa? Brent Seales (UKy), together with Tom Learner (Getty), as meeting organizers, will write a proposal to the NSF for funding for the meeting, select and invite an organizing committee and attendees, and then plan and execute the meeting. The Getty Conservation Institute has agreed to co-sponsor the meeting, offering in-kind support and their Los Angeles based facility as the meeting location. We will seek to include experts from areas such as Machine Learning, Artificial Intelligence, Imaging and Visualization, Heritage Science, the Humanities, Computer Science / Data Science, and Industry. In addition to primary support from the NSF, we will work to secure co-sponsorship of the meeting from the AHRC, the NEH, and industry partners