Unveil the unseen: Exploit information hidden in noise

Abstract

Noise and uncertainty are usually the enemy of machine learning, noise in training data leads to uncertainty and inaccuracy in the predictions. However, we develop a machine learning architecture that extracts crucial information out of the noise itself to improve the predictions. The phenomenology computes and then utilizes uncertainty in one target variable to predict a second target variable. We apply this formalism to PbZr${}_{0.7}$Sn${}_{0.3}$O${}_3$ crystal, using the uncertainty in dielectric constant to extrapolate heat capacity, correctly predicting a phase transition that otherwise cannot be extrapolated. For the second example -- single-particle diffraction of droplets -- we utilize the particle count together with its uncertainty to extrapolate the ground truth diffraction amplitude, delivering better predictions than when we utilize only the particle count. Our generic formalism enables the exploitation of uncertainty in machine learning, which has a broad range of applications in the physical sciences and beyond.

arXiv comment13 pages, 10 figures. Appl Intell (2022)

References in corpus (7)

• Less is more: sampling chemical space with active learning
• Prediction model of band-gap for AX binary compounds by combination of density functional theory calculations and machine learning techniques
• Anomaly Detection using Autoencoders in High Performance Computing Systems
• OPTIMADE, an API for exchanging materials data
• Unsupervised machine learning via transfer learning and k-means clustering to classify materials image data
• Materials data validation and imputation with an artificial neural network
• Shot noise variation within ensembles of gold atomic break junctions at room temperature

Cited by in corpus (1)

• Probabilistic selection and design of concrete using machine learning