The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Workshop:SC21 SuperCompCloud: 5th International Workshop on Interoperability of Supercomputing and Cloud Technologies
Authors: Beth Plale (Indiana University)
Abstract: The integrity of data and AI models that use data are critical to the trustworthiness of the outcomes of AI research. Data and AI models could be corrupted by malicious actors on one hand and could be subject to restrictions on their use on the other. Both suggest the need for care. This problem takes on unique proportions when AI research requires large-scale HPC and cloud resources. In this talk I will speak to the issue of data integrity in HPC and cloud-based AI research, touching on current research at IU and within the context of the recently funded NSF AI Institute, ICICLE, Intelligent CI with Computational Learning in the Environment.