Smart machine learning or discovering meaningful physical and chemical contributions through dimensional stacking

Abstract

Despite remarkable advances in characterization techniques of functional materials yielding an ever growing amount of data, the interplay between the physical and chemical phenomena underpinning materials' functionalities is still often poorly understood. Dimensional reduction techniques have been used to tackle the challenge of understanding materials' behavior, leveraging the very large amount of data available. Here, we present a method for applying physical and chemical constraints to dimensional reduction analysis, through dimensional stacking. Compared to traditional, uncorrelated techniques, this approach enables a direct and simultaneous assessment of behaviors across all measurement parameters, through stacking of data along specific dimensions as required by physical or chemical correlations. The proposed method is applied to the nanoscale electromechanical relaxation response in (1 − x)PMN-xPT solid solutions, enabling a direct comparison of electric fieldand chemical composition-dependent contributors. A poling-like, and a relaxation-like behavior with a domain glass state are identified, and their evolution is tracked across the phase diagram. The proposed dimensional stacking technique, guided by the knowledge of the underlying physics of correlated systems, is valid for the analysis of any multidimensional dataset, opening a spectrum of possibilities for multidisciplinary use.npj Computational Materials (2019) 5:85 ; https://doi.

Better, Faster, and Less Biased Machine Learning: Electromechanical Switching in Ferroelectric Thin Films

“…[51] Dimensional stacking was applied by simple concatenation of the contact frequency at resonance and piezoresponse data prior to analysis. [39] All raw data and analysis codes are provided in the form of Jupyter notebooks. [52] These notebooks perform the analyses discussed above as well as generate the plots used as figures in this work.…”

“…In order to impart additional physical constraints, dimensional stacking is used to concatenate the piezoresponse and contact resonance data prior to the DL analysis. [39] Such concatenation implies a fundamental correlation between the electromechanical response (as probed by the piezoresponse, PR, which combines A and φ information), and the elastic properties of the material (as tracked by the cantilever-tip resonance) to the applied waveform, as discussed above. We note that the piezoresponse data were also processed by mean removal from each PR curve, and subsequently scaled by a constant factor such that the resulting PR dataset has a variance comparable to that of the resonance curves (for details see the Supporting Information).…”

“…Traditional approaches to materials science indeed are based on interpretation of correlated datasets through the lens of physical and/or chemical behaviors-aspects fundamentally absent in ML methods. Recently, dimensional stacking [39] has been proposed as a method to introduce such correlations and constraints into ML techniques through the concatenation of datasets across relevant dimensions. Encoding such knowledge within the Machine-learning techniques are more and more often applied to the analysis of complex behaviors in materials research.…”

Artificial intelligence-enabled smart mechanical metamaterials: advent and future trends

“…Advantages of the novel materials and structures are obtained mainly due to the capability of AI models in finding the good balance between rationale microstructures and high accuracy, such that to effectively satisfy the performance requirements for those architected materials and structures [286][287][288][289]. Comparing with the applications in other fields, AI has not debuted in materials and structures until recent years [290][291][292][293][294][295][296][297][298][299][300]. Studies have been reported on using AI algorithms to emulate complex biological processes (e.g.…”

To switch or not to switch – a machine learning approach for ferroelectricity

“…In order to analyze and extract meaning from these data sets, machine learning has become increasingly important. [27][28][29][30][31][32][33][34] Dimensionality reduction without loss of important information, de-noising, clustering and identifying characteristic features in data sets have been achieved using supervised and unsupervised machine-learning algorithms.…”