Bio

Ahmad A. Rushdi, PhD, is the director of industry Research Programs, at Stanford’s Institute for Human-Centered AI (HAI). He works on translating cutting-edge AI research into applied, responsible, and deployable AI solutions for global enterprises in different domains, building durable bridges between Stanford scholars and industry researchers via research collaborations and executive education. As a research scientist, Ahmad's own research focuses on rigorous uncertainty quantification methods for trustworthy AI/ML models and systems. Previously, he held R&D roles at Sandia National Labs, Northrop Grumman, UC Davis, UT Austin, and Cisco. PhD: Electrical & Computer Engineering, UC Davis, and MS/BS: Electrical Engineering, Cairo University.

Additional Info

All Publications

  • Multifidelity data fusion in convolutional encoder/decoder networks JOURNAL OF COMPUTATIONAL PHYSICS Partin, L., Geraci, G., Rushdi, A. A., Eldred, M. S., Schiavazzi, D. E. 2023; 472

    View details for DOI 10.1016/j.jcp.2022.111666

    View details for Web of Science ID 000896020000012

  • Assessing the Fidelity of Explanations with Global Sensitivity Analysis Smith, M. R., Acquesta, E., Smutz, C., Rushdi, A., Moss, B. edited by Bui, T. X. HICSS. 2023: 1085-1094

    View details for Web of Science ID 001301786701017

  • A brief review on DNA storage, compression, and digitalization NANO COMMUNICATION NETWORKS Cevallos, Y., Nakano, T., Tello-Oquendo, L., Rushdi, A., Inca, D., Santillan, I., Shirazi, A., Samaniego, N. 2022; 31

    View details for DOI 10.1016/j.nancom.2021.100391

    View details for Web of Science ID 000776085500001

  • Multifidelity data fusion in convolutional encoder/decoder assembly networks for computational fluid dynamics Partin, L., Rushdit, A. A., Schiavazzi, D. E., AIAA AMER INST AERONAUTICS & ASTRONAUTICS. 2022

    View details for DOI 10.2514/6.2022-0803

    View details for Web of Science ID 001409636200203

  • Power Prediction of Airborne Wind Energy Systems Using Multivariate Machine Learning ENERGIES Rushdi, M. A., Rushdi, A. A., Dief, T. N., Halawa, A. M., Yoshida, S., Schmehl, R. 2020; 13 (9)

    View details for DOI 10.3390/en13092367

    View details for Web of Science ID 000535739300245

  • Spoke-Darts for High-Dimensional Blue-Noise Sampling ACM TRANSACTIONS ON GRAPHICS Mitchell, S. A., Ebeida, M. S., Awad, M. A., Park, C., Patney, A., Rushdi, A. A., Swiler, L. P., Manocha, D., Wei, L. 2018; 37 (2)

    View details for DOI 10.1145/3194657

    View details for Web of Science ID 000437485900011

  • All-quad meshing without cleanup Rushdi, A. A., Mitchell, S. A., Mahmoud, A. H., Bajaj, C. C., Ebeida, M. S. ELSEVIER SCI LTD. 2017: 83-98

    View details for DOI 10.1016/j.cad.2016.07.009

    View details for Web of Science ID 000395845100008

  • VPS: VORONOI PIECEWISE SURROGATE MODELS FOR HIGH-DIMENSIONAL DATA FITTING INTERNATIONAL JOURNAL FOR UNCERTAINTY QUANTIFICATION Rushdi, A. A., Swiler, L. P., Phipps, E. T., D'Elia, M., Ebeida, M. S. 2017; 7 (1): 1-21

    View details for DOI 10.1615/Int.J.UncertaintyQuantification.2016018697

    View details for Web of Science ID 000397511200001

  • POF-Darts: Geometric adaptive sampling for probability of failure RELIABILITY ENGINEERING & SYSTEM SAFETY Ebeida, M. S., Mitchell, S. A., Swiler, L. P., Romero, V. J., Rushdi, A. A. 2016; 155: 64-77

    View details for DOI 10.1016/j.ress.2016.05.001

    View details for Web of Science ID 000382420800007

  • Disk Density Tuning of a Maximal Random Packing Ebeida, M. S., Rushdi, A. A., Awad, M. A., Mahmoud, A. H., Yan, D., English, S. A., Owens, J. D., Bajaj, C. L., Mitchell, S. A. WILEY. 2016: 259-269

    Abstract

    We introduce an algorithmic framework for tuning the spatial density of disks in a maximal random packing, without changing the sizing function or radii of disks. Starting from any maximal random packing such as a Maximal Poisson-disk Sampling (MPS), we iteratively relocate, inject (add), or eject (remove) disks, using a set of three successively more-aggressive local operations. We may achieve a user-defined density, either more dense or more sparse, almost up to the theoretical structured limits. The tuned samples are conflict-free, retain coverage maximality, and, except in the extremes, retain the blue noise randomness properties of the input. We change the density of the packing one disk at a time, maintaining the minimum disk separation distance and the maximum domain coverage distance required of any maximal packing. These properties are local, and we can handle spatially-varying sizing functions. Using fewer points to satisfy a sizing function improves the efficiency of some applications. We apply the framework to improve the quality of meshes, removing non-obtuse angles; and to more accurately model fiber reinforced polymers for elastic and failure simulations.

    View details for DOI 10.1111/cgf.12981

    View details for Web of Science ID 000383444500025

    View details for PubMedID 27563162

    View details for PubMedCentralID PMC4994978

  • Recursive Spoke Darts: Local Hyperplane Sampling for Delaunay and Voronoi Meshing in Arbitrary Dimensions Ebeida, M. S., Rushdi, A. A. edited by Canann, S., Owen, S., Si, H. ELSEVIER SCIENCE BV. 2016: 110-122

    View details for DOI 10.1016/j.proeng.2016.11.033

    View details for Web of Science ID 000397997600009

  • All-Hex Meshing of Multiple-Region Domains without Cleanup Awad, M. A., Rushdi, A. A., Abbas, M. A., Mitchell, S. A., Mahmoud, A. H., Bajaj, C. L., Ebeida, M. S. edited by Canann, S., Owen, S., Si, H. ELSEVIER SCIENCE BV. 2016: 251-261

    Abstract

    In this paper, we present a new algorithm for all-hex meshing of domains with multiple regions without post-processing cleanup. Our method starts with a strongly balanced octree. In contrast to snapping the grid points onto the geometric boundaries, we move points a slight distance away from the common boundaries. Then we intersect the moved grid with the geometry. This allows us to avoid creating any flat angles, and we are able to handle two-sided regions and more complex topologies than prior methods. The algorithm is robust and cleanup-free; without the use of any pillowing, swapping, or smoothing. Thus, our simple algorithm is also more predictable than prior art.

    View details for DOI 10.1016/j.proeng.2016.11.055

    View details for Web of Science ID 000397997600020

    View details for PubMedID 28845204

    View details for PubMedCentralID PMC5568131

https://orcid.org/0000-0003-0438-2459