PFM Social Network

Making DCNN work in real time on SPM - kudos to Yongtao Liu and Maxim Ziatdinov! We get computer to identify domain walls of interest and run hysteresis loop measurements only at these selected locations.

Exploring Physics of Ferroelectric Domain Walls in Real Time: Deep Learning Enabled Scanning Probe Microscopy The functionality of ferroelastic domain walls in ferroelectric materials is explored in real-time via the in-situ implementation of computer vision algorithms in scanning probe microscopy (SPM) experiment. The robust deep convolutional neural network (DCNN) is implemented based on a deep residual l...

If you are interested in Bayesian Active learning for scanning probe microscopy (or electron microscopy, or nanoindentation) - we have the grand overview out! GPax, GPim, and AtomAI libraries by are there if interested to apply for own systems and microscopes!

Bayesian Active Learning for Scanning Probe Microscopy: from Gaussian Processes to Hypothesis Learning Recent progress in machine learning methods, and the emerging availability of programmable interfaces for scanning probe microscopes (SPMs), have propelled automated and autonomous microscopies to the forefront of attention of the scientific community. However, enabling automated microscopy requires...

Can the machine learning algorithm run the microscope to learn physics of domain formation via exploring competing hypotheses?

Also, extension to other automated experiments is synthesis or scientific instrumentation is straightforward. Physics is all it takes.

Hypothesis-Driven Automated Experiment in Scanning Probe Microscopy: Exploring the Domain Growth Laws in Ferroelectric Materials We report the development and implementation of a hypothesis learning based automated experiment, in which the microscope operating in the autonomous mode identifies the physical laws behind the material's response. Specifically, we explore the bias induced transformations that underpin the function...

As an interesting development, looks like Bruker has implemented the commercial version of the SS-PFM mode with the option for complex spectroscopies. This is really great, since now systematic studies of polarization dynamics are becoming possible on commercial tools!

Exactly 15 years after the Rev. Sci. Instr. paper introducing this mode for the first time, https://aip.scitation.org/doi/abs/10.1063/1.2214699, and adding more complex capabilities as in https://aip.scitation.org/doi/abs/10.1063/1.2980031 and https://pubs.acs.org/doi/abs/10.1021/nn505176a

https://www.bruker.com/en/products-and-solutions/microscopes/materials-afm/afm-modes/ss-pfm.html

Switching Spectroscopy Piezoresponse Force Microscopy (SS-PFM) Switching Spectroscopy Piezoresponse Force Microscopy (SS-PFM) mode provides highly accurate ferroelectric hysteresis loop measurement by improving the sensitivity and accuracy of PFM

Very important paper for ferroelectric and PFM community - if we remember that there is water meniscus between tip and the surface.

Ferroelectric 2D ice under graphene confinement - Nature Communications Ferroelectric ordering of water has been at the heart of intense debates due to its importance in enhancing our understanding of the condensed matter. Here, the authors observe ferroelectric properties of water ice in a two dimensional phase under confinement between two graphene layers.

Lectures from ORNL workshop on Automated Experiment and Machine Learning in Scanning Probe Microscopy -> online. Featuring lectures by Mikhail Katsnelson , Leroy Cronin, Danilo J. Rezende , and many others, as well as tutorials on variational autoencoders for image and spectral analysis, Bayesian Optimization, and deep convolutional networks and deep kernel learning. See:

Machine Learning and Automated Experiments in Scanning Probe Microscopy Virtual School Recordings, October 4-7, 2021 | ORNL Organizers: Rama Vasudevan, Kyle Kelley, Maxim Ziatdinov, Josh Agar, and Sergei V. Kalinin This virtual school on ML and AE in SPM, held October 4-7, 2021, combined invited and contributed presentations at the forefront of ML applications in Scanning Probe Microscopy, including both atomically resol...

Some automated experiment in PFM - based on prior-known descriptors. Aka define what domain wall you want to study, find it, and measure polarization loops or current-voltage curves - right at the wall, or at some separation from it, as a function of curvature, or other characteristics.

Probing polarization dynamics at specific domain configurations: Computer-vision based automated experiment in piezoresponse force microscopy Topological defects in ferroelectric materials have attracted much attention due to the emergence of conductive, ferroic, and magnetic functionalities. However, many topological configurations dyna...

Can multi objective Bayesian Optimization help find right Ferro (or antiferro) electric for specific application? Especially with chemical (or interfacial) control?

Multi-objective Bayesian optimization of ferroelectric materials with interfacial control for memory and energy storage applications Optimization of materials performance for specific applications often requires balancing multiple aspects of materials functionality. Even for the cases where generative physical model of material behavior is known and reliable, this often requires search over multidimensional parameter space to ide...

We are excited to invite you to the virtual school “Machine Learning and Automated Experiment in Scanning Probe Microscopy” to be held at ORNL, October 4-7, 2021, organized by Rama Vasudevan, Kyle Kelley, Maxim Ziatdinov, Josh Agar, and Sergei V. Kalinin. Due to the close similarity between SPM and STEM, these topics may also be of interest for STEM community as well.

The registration deadline is August 25. Link to register:

https://www.surveymonkey.com/r/VDWFV5K

Machine learning (ML) has emerged as a powerful tool for data and image analysis and as an enabling component of autonomous systems in areas ranging from biological and medical imaging to self-driving cars. This rapid growth in ML applications poses the question as to which of these methods can be applied in scanning probe microscopy, both to derive insights into the physics and chemistry of real materials, enable optimization of imaging conditions, and engender transition to the automated and autonomous experiment. This virtual school on ML and AE for SPM, to be held virtually on October 4-7, 2021, will combine invited and contributed presentations at the forefront of ML applications in Scanning Probe Microscopy, including both atomically resolved Scanning Tunneling Microscopy and Spectroscopy and mesoscopic Scanning Probe Microscopy techniques.

Special emphasis will be made on necessary conditions for physically-meaningful machine learning analysis and especially automated experiments in SPM. It will feature tutorials on recent developments in ML analysis of mesoscopic and atomically resolved images and spectroscopy, including deep convolutional neural networks (DCNNs) for feature identifications, symmetry-invariant (variational) autoencoders ((V)AE), and Gaussian Processes and Deep Kernel Learning-based super-resolution imaging and image reconstruction, and reinforcement learning for image optimization and automated experiment. The presentations will be followed by hands-on tutorial sessions introducing the attendees to the AtomAI, GPim, PyroVED, and various Pycroscopy packages. All the technologies and workflows discussed during the tutorials will be open source. The attendees are encouraged to contact the organizers in advance to setup analysis of own datasets. The meeting will be free of charge. The final program will be available by September 2021.

If interested, please contact Sergei Kalinin ([email protected]) to submit the abstract.

Machine Learning and Automated Experiment in Scanning Probe Microscopy Take this survey powered by surveymonkey.com. Create your own surveys for free.