Deep neural networks for understanding noisy data applied to physical property extraction in scanning probe microscopy

Abstract

The rapid development of spectral-imaging methods in scanning probe, electron, and optical microscopy in the last decade have given rise for large multidimensional datasets. In many cases, the reduction of hyperspectral data to the lower-dimension materialsspecific parameters is based on functional fitting, where an approximate form of the fitting function is known, but the parameters of the function need to be determined. However, functional fits of noisy data realized via iterative methods, such as least-square gradient descent, often yield spurious results and are very sensitive to initial guesses. Here, we demonstrate an approach for the reduction of the hyperspectral data using a deep neural network approach. A combined deep neural network/least-square approach is shown to improve the effective signal-to-noise ratio of band-excitation piezoresponse force microscopy by more than an order of magnitude, allowing characterization when very small driving signals are used or when a material's response is weak.

Machine learning at the (sub)atomic scale: next generation scanning probe microscopy

“…' Shortly before Zhang et al's work was published in July this year, a related approach to the automated extraction of 'buried' information from AFM, rather than STM, data was described in what is very likely to be a highly influential paper, by Benjamin Aldritt and his colleagues (and recently accepted in the journal Science Advances) [34]. Building on previous machine learning protocols developed by Sergei Kalinin and co-workers at Oak Ridge National Laboratories (among others) [29,36,37], Aldritt et al have developed what they describe as automated structure discovery AFM (ASD-AFM), a deep learning framework based on a similar methodology to that of Zhang et al, whereby a large set of simulated images is generated-in this case via a combination of density functional theory (DFT) optimisation of molecular structures and the probe-particle model of tip-sample interactions developed by Hapala et al [38,39] -and a convolutional neural net is used to determine the best match between experimental data and a molecular geometry. Figure 2 illustrates the general CNN methodology adopted by Aldritt et al, alongside Ziatdinov, Maksov, and Kalinin's earlier work on determining the rotational state of adsorbed molecules from STM data [35].…”

“…Atomic force microscopy also has its own variants of spectroscopy and spectromicroscopy 4 including single point force-distance curves (where 'distance' refers to the tip-sample separation), force-distance maps, potential energy landscapes, damping/dissipation variations, phase maps, Kelvin probe force microscopy, and higher harmonic signal variations as a function of both lateral and vertical position of the probe. Taken together, these various information channels provide an exceptionally rich multimodal (or hyperspectral) multidimensional dataset, which, either in isolation or combined with image data, can be mined via machine learning strategies to not only provide significant improvements in both post-experiment [37] and real-time [45] effective signal-to-noise ratio but, importantly, to classify and determine material properties at the nanoscale and below. Burzawa et al [46] have taken his strategy one step further and adopted machine learning not to extract materials properties but to determine which particular physical model/dynamics drives pattern formation in a system (in their case, the 2D Ising model).…”

Application of a long short-term memory for deconvoluting conductance contributions at charged ferroelectric domain walls

“…Recently, advances in large structured databases, efficient computation, and machine-learning algorithms have allowed the extraction of physically meaningful information based on statistical analysis 17 . This approach has become pervasive in scanning probe microscopy (SPM) measurements, where it has provided important insight into the switching processes of ferroelectric domain structures 18,19 , improved effective signal-to-noise ratios 20 , and was utilized to identify tip degradation artifacts 21 . These investigations showed the potential of combining multidimensional SPM with statistical methods of machine learning to better understand nanoscale functional responses and pathways.…”

Machine learning-based multidomain processing for texture-based image segmentation and analysis

“…In cases when such dataset can be generated, DNN performance in the classification and segmentation of the image is extremely efficient. 23 However, when such training set in unavailable, more traditional machine learning tools become necessary. Here, different types of feature spatial organization within the image are captured by algorithms targeted at specific structural descriptors such as symmetry, periodicity, directionality, etc.…”