Internal Measurement

Internal measurement focuses on extracting information about a system's internal state from limited, often noisy, external observations. Current research emphasizes developing robust algorithms and models, including neural networks (e.g., deep learning architectures, neural fields) and techniques like the information bottleneck, to improve the accuracy and efficiency of these estimations across diverse applications. This field is crucial for advancing numerous scientific domains, from physics and materials science (e.g., refractive index mapping) to engineering (e.g., robotics, power systems monitoring) and artificial intelligence (e.g., language model evaluation, chaotic system analysis), by enabling more precise and reliable inference from incomplete data.