AI/ML Methods for Simulations across HEP Frontiers

Simulations play a key role in high-energy physics research across the frontiers. They provide an important tool to explore new physics ideas and predict their signatures that can then be measured. Simulations are often essential for the proper analysis of experimental and observational measurements and in deriving error estimates. The accuracy requirements for large-scale simulations pose a major challenge, as many simulation parameters have to be tuned and the computational cost of the simulations can be very large. New AI/ML approaches can help with these challenges in many ways, for example by optimizing modeling approaches or speeding up simulations. However, at the same time, it is important to ensure that these new approaches do not introduce hidden biases into the predictions. In this talk I will provide an overview of how AI/ML methods have played important roles to support simulation approaches in the different frontiers. I will also discuss what challenges still lie ahead of us to bring AI/ML methods to full fruition in the area of high-energy physics simulations.