This talk explores the emerging interface of quantum computation and machine learning. I will introduce how physics-informed generative models for quantum states provide powerful tools for high-dimensional quantum simulations, enabling new advances in high-energy physics, quantum materials, and quantum information. I will also highlight recent efforts to develop ML and robotics methods inspired by physical and quantum principles, including generative models grounded in physical processes, quantum-inspired neural solvers for PDEs, and many-body-based approaches for optimizing multi-legged robot locomotion.

This talk will focus on AI meets High Energy Nuclear Physics from inverse problem solving and deep generative modelling, to introduce the methodologies of machine learning used in exploring QCD matter under extreme conditions. Related to the inverse problem solving for QCD matter studies, supervised learning, Bayesian Inference and automatic differentiation based discriminative AI methods will be discussed. For Deep Generative modelling, the flow-based models and diffusion models will be discussed for their usage in lattice field theory and heavy ion collisions simulations.