Hidden Physic

Hidden physics research focuses on uncovering and modeling physical phenomena not directly observable through traditional methods, aiming to improve predictions and understanding of complex systems. Current efforts leverage machine learning, particularly neural networks (including physics-informed neural networks and variations like Kolmogorov-Arnold networks) and neuro-symbolic frameworks, to infer hidden properties like mass and charge from limited data, often incorporating known physical laws into the models. This approach has significant implications for diverse fields, enabling more accurate simulations and predictions in areas such as fluid dynamics, plasma physics, and material science, and potentially leading to improved designs and control strategies in various engineering applications.