In the wake of rapid advancements in GPU-based computing and artificial intelligence, we're driven to see how AI can be leveraged to unpack complex physical problems. One of the foremost challenges that persists across research domains is securing high-quality, precise datasets, especially for supervised learning. In our lab, this challenge takes center stage. Equipped with powerful computing, bespoke in-house coding, and selectively chosen open-source tools, we're poised to address this data collection conundrum, facilitating the realization of detailed computer simulations.

While established AI methods like CNNs, MLPs, GNNs, and RNNs are prominent in research, we strike a balance. These foundational architectures are considered alongside our exploration of innovative territories, such as Transformer and diffusion models. Our neural network designs are carefully crafted, targeting the unique demands of our physical studies. Delving into physical phenomena is both a captivating challenge and a voyage of discovery. If this convergence of physics and AI sparks your curiosity or insights, we're eager to hear your perspective.