Multimedia Release

How to Make Machine Learning Work in A Data-Limited System (5 of 8)

Reports and Proceedings

American Association for the Advancement of Science (AAAS)

How to Make Machine Learning Work in A Data-Limited System (5 of 8)

image: A photograph of crumpled mylar, the corresponding laser scanned heightmap, and the resulting curvature map all overlayed. We show the data-collection pipeline imposed on a single image. In the upper right, we show a photograph of raw data. This fades into the heightmap that the laser scans (in the bottom, middle) and into the mean curvature map computed from the laser scan in the upper left. This material relates to a paper that appeared in the April 26th, 2019, issue of Science Advances, published by AAAS. The paper, by J. Hoffmann at Harvard University in Cambridge, MA, and colleagues was titled, "Machine learning in a data-limited regime: Augmenting experiments with synthetic data uncovers order in crumpled sheets." view more 

Credit: Jordan Hoffmann, Shruti Mishra, and Lisa Lee


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.