Machine Learning and Data-Driven Approaches for Materials Synthesis, Characterization, and Understanding

Katherine Sytwu, Edward Barnard, David Prendergast

Recent advances in data-driven methodologies such as data mining, machine learning, and statistical modeling have enabled researchers to incorporate observations and technical intuition that may be difficult to analytically define into the experimentation process and extract out statistically-relevant behavior. Simultaneously, high-throughput synthesis and characterization have led to a deluge of scientific data that presents new challenges in both understanding information content and responding appropriately to focus on interesting regions or phenomena. This symposium will highlight the diverse opportunities and challenges in which data science can inform, augment, and improve experimentation and analysis for applications spanning autonomous materials discovery and synthesis, correlative and multimodal measurements, novel data analysis algorithms, and data infrastructure challenges that underpin implementation into experimental workflows.

Symposium Sponsor:

Symposium Location: B50 Auditorium

Symposium Schedule:

12:45 – 1:15 pm

Joseph Montoya, Toyota Research Institute

1:15 – 1:30 pm

Caravaggio Caniglia, Stanford University

1:30- 1:45 pm

Myriam Diatta, Berkeley Lab

1:45 – 2:15 pm

Aditi Krishnapriyan, University of California, Berkeley

2:15 – 2:45 pm
2:45 – 3:15 pm

Mitra Taheri, Johns Hopkins University

3:15 – 3:30 pm

Samuel Gleason, University of California, Berkeley

3:30- 3:45 pm

Xingzhi Wang, University of California, Berkeley

3:45 – 4:00 pm

Alexander Pattison, Berkeley Lab

4:00 – 4:15 pm

Ellis Kennedy, University of California, Berkeley

Abstracts