Topics of Interest

A more specific list of topics of interest for speakers and attendees to consider:

• Scientific data set creation, ingest, curation, labelling, and analysis with statistical models and inference

• Incorporating realtime and ad-hoc data analytics into applications and their deployment on supercomputing and cluster platforms

• Computational steering through machine learning models and related control theory approaches

• Meta-data and data metrics collection and generation for large data collections and output data sets of computational simulations

• Multi-precision training/inference methods and their use on modern hardware for simulation data

• Novel use of discriminative and generative machine learning approaches for scientific data sets including Adversarial and Reinforcement Learning with self-supervision

• Modern HPC storage issues when dealing with integration of computational simulation outputs with data analytics software

• Synchronous and asynchronous learning approaches at scale for methods related to deep neural network training, stochastic gradient descent, loss-function engineering, and related distributed optimization techniques

• Model derivation and training for scalable simulations and data sets

• Hyperparameter search and optimization incorporating recent advances in Bayesian optimization

• Deployment of statistical models and their implementations such as TensorFlow and PyTorch or application-specific tensor frameworks.

• Integration of models with large scale simulations code bases through containers (Kubernetes, Docker, Singularity, OpenShift), virtualizaiton, colocation, and workflow frameworks

• Pretrained models’ creation, use, and scaling for scientific simulations