Towards the Development of Entropy-Based Anomaly Detection in an Astrophysics Simulation
This work addresses a domain-specific problem for astrophysics researchers, focusing on an incremental application of existing ML methods to enhance simulation accuracy.
The paper tackles the problem of improving simulation accuracy in astrophysics by applying anomaly detection techniques to a core-collapse supernovae simulation, reporting early successes but without providing concrete numerical results.
The use of AI and ML for scientific applications is currently a very exciting and dynamic field. Much of this excitement for HPC has focused on ML applications whose analysis and classification generate very large numbers of flops. Others seek to replace scientific simulations with data-driven surrogate models. But another important use case lies in the combination application of ML to improve simulation accuracy. To that end, we present an anomaly problem which arises from a core-collapse supernovae simulation. We discuss strategies and early successes in applying anomaly detection techniques from machine learning to this scientific simulation, as well as current challenges and future possibilities.