Abstract: Coherent X-ray diffraction imaging (CDI) is a powerful technique for operando materials characterization at the nanoscale. Uniquely, in this technique, the image resolution is not determined by the resolution of the optics used in the experiment but by the maximum angle through which the X-rays are scattered by the sample.  The challenge lies in recovering the phases of those scattered X-rays, which are not measured by the detector, to retrieve the image of the sample from the data. The phase retrieval process can be computationally expensive, and the current methods has inherent deficiencies. In this talk, I will describe a deep convolutional neural network approach to this inversion problem and report the current status of the approach.

Bio: Henry Chan is a postdoc at the Center for Nanoscale Materials at Argonne National Laboratory. He received his Ph.D. in computational chemistry from the University of Illinois at Chicago. He has extensive experience in high-performance computing and modeling of nanostructured materials. His current research focuses on the application of machine learning in molecular simulations and X-ray diffraction imaging. Some of these efforts include data-driven force fields derived using supervised learning, an invention for 3-D microstructure characterization, and phase retrieval in coherent diffraction imaging using deep convolutional neural networks.