Recent advances in machine learning for nuclear and particle physics

Mirroring global trends, we now see use of machine learning (ML) throughout the phases of discovery in nuclear and particle physics. Providing a snapshot of recent technical trends and discussing evolving community dynamics, this talk aims to give an overview of the current state of ML in these physics domains. ML is currently being used from fast, online decision making in experiments to providing theoretical predictions with Bayesian uncertainty quantification. Methods range from traditional ML to modern deep learning architectures and foundation models. This talk will survey the wide array of ML methods in use and highlight recent novel ML work in experiment design, realtime systems, data analysis, and theory. As the use of ML becomes more ubiquitious, there are emerging community norms surrounding data access, model dissemination, and publication in some subfields. A summary of these trends will be provded with a look towards the future in community guidelines.