Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “interpretable machine learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 685 records · Page 38
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Constrained non-negative matrix factorization enabling real-time insights of in situ and high-throughput experiments

Non-negative matrix factorization (NMF) is an appealing class of methods for performing unsupervised learning on streaming spectral data, particularly in time-sensitive applications such as in situ characterization of materials. These methods seek to decompose a dataset into a small number of components and weights that can compactly represent the underlying signal while effectively reconstructing the observations with minimal error. However, canonical NMF methods have no underlying requirement that the reconstruction uses components or weights that are representative of the true physical processes. In this work, we demonstrate how constraining a subset of the NMF weights or components as rigid priors, provided as known or assumed values, can provide significant improvement in revealing true underlying phenomena. We present a PyTorch-based method for efficiently applying constrained NMF and demonstrate its application to several synthetic examples. Our implementation allows an expert researcher-in-the-loop to provide and dynamically adjust the constraints during a live experiment involving streaming spectral data. Such interactive priors allow researchers to specify known or identified independent components, as well as functional expectations about the mixing or transitions between the components. We further demonstrate the application of this method to measured synchrotron x-ray total scattering data from in situ beamline experiments. In such a context, constrained NMF can result in a more interpretive and scientifically relevant decomposition than canonical NMF or other decomposition techniques. As a result, the details of the method are provided, along with general guidance for employing constrained NMF in the extraction of critical information and insights during time-sensitive experimental applications.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Search for new particles in events with energetic jets and large missing transverse momentum in proton-proton collisions at $ \sqrt{s} $ = 13 TeV

A search is presented for new particles produced at the LHC in proton-proton collisions at s√ = 13 TeV, using events with energetic jets and large missing transverse momentum. The analysis is based on a data sample corresponding to an integrated luminosity of 101 fb -1 , collected in 2017–2018 with the CMS detector. Machine learning techniques are used to define separate categories for events with narrow jets from initial-state radiation and events with large-radius jets consistent with a hadronic decay of a W or Z boson. A statistical combination is made with an earlier search based on a data sample of 36 fb -1 , collected in 2016. No significant excess of events is observed with respect to the standard model background expectation determined from control samples in data. The results are interpreted in terms of limits on the branching fraction of an invisible decay of the Higgs boson, as well as constraints on simplified models of dark matter, on first-generation scalar leptoquarks decaying to quarks and neutrinos, and on models with large extra dimensions. Several of the new limits, specifically for spin-1 dark matter mediators, pseudoscalar mediators, colored mediators, and leptoquarks, are the most restrictive to date.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

ChemEcho v1.0

ChemEcho is a tool that converts tandem mass spectra into embeddings used to build machine learning (ML) models with fully explainable predictions. It provides an API for transforming raw tandem mass spectral data into embeddings, along with functions for training and validating ML models. Additionally, it includes utilities for retrieving and cleaning training data. ChemEcho is broadly applicable in ML pipelines that use tandem mass spectra for a variety of tasks, such as chemical classification or bioactivity mining. While there are existing methods to generate embeddings from fragmentation data, ChemEcho's approach ensures that predictions remain interpretable, enabling experts to evaluate results and generate hypotheses about the underlying data.

Harwood, Thomas [Lawrence Berkeley National Labora↗
DOE OSTI2022

Classification of Nuclear Reactor Operations Using Spatial Importance and Multisensor Networks

Distributed multisensor networks record multiple data streams that can be used as inputs to machine learning models designed to classify operations relevant to proliferation at nuclear reactors. The goal of this work is to demonstrate methods to assess the importance of each node (a single multisensor) and region (a group of proximate multisensors) to machine learning model performance in a reactor monitoring scenario. This, in turn, provides insight into model behavior, a critical requirement of data-driven applications in nuclear security. Using data collected at the High Flux Isotope Reactor at Oak Ridge National Laboratory via a network of Merlyn multisensors, two different models were trained to classify the reactor’s operational state: a hidden Markov model (HMM), which is simpler and more transparent, and a feed-forward neural network, which is less inherently interpretable. Traditional wrapper methods for feature importance were extended to identify nodes and regions in the multisensor network with strong positive and negative impacts on the classification problem. These spatial-importance algorithms were evaluated on the two different classifiers. The classification accuracy was then improved relative to baseline models via feature selection from 0.583 to 0.839 and from 0.811 ± 0.005 to 0.884 ± 0.004 for the HMM and feed-forward neural network, respectively. While some differences in node and region importance were observed when using different classifiers and wrapper methods, the nodes near the facility’s cooling tower were consistently identified as important—a conclusion further supported by studies on feature importance in decision trees. Node and region importance methods are model-agnostic, inform feature selection for improved model performance, and can provide insight into opaque classification models in the nuclear security domain.

Tibbetts, Jake↗
DOE OSTI2022

Fiber Bundle Morphisms as a Framework for Modeling Many-to-Many Maps

While it is not generally reflected in the `nice' datasets used for benchmarking machine learning algorithms, the real-world is full of processes that would be best described as many-to-many. That is, a single input can potentially yield many different outputs (whether due to noise, imperfect measurement, or intrinsic stochasticity in the process) and many different inputs can yield the same output (that is, the map is not injective). For example, imagine a sentiment analysis task where, due to linguistic ambiguity, a single statement can have a range of different sentiment interpretations while at the same time many distinct statements can represent the same sentiment. When modeling such a multivalued function $f: X \rightarrow Y$, it is frequently useful to be able to model the distribution on $f(x)$ for specific input $x$ as well as the distribution on fiber $f^{-1}(y)$ for specific output $y$. Such an analysis helps the user (i) better understand the variance intrinsic to the process they are studying and (ii) understand the range of specific input $x$ that can be used to achieve output $y$. Following existing work which used a fiber bundle framework to better model many-to-one processes we describe how morphisms of fiber bundles provide a template for building models which naturally capture the structure of many-to-many processes.

Coda, Elizabeth D.↗
DOE OSTI2024

Low-index mesoscopic surface reconstructions of Au surfaces using Bayesian force fields

Metal surfaces have long been known to reconstruct, significantly influencing their structural and catalytic properties. Many key mechanistic aspects of these subtle transformations remain poorly understood due to limitations of previous simulation approaches. Using active learning of Bayesian machine-learned force fields trained from ab initio calculations, we enable large-scale molecular dynamics simulations to describe the thermodynamics and time evolution of the low-index mesoscopic surface reconstructions of Au (e.g., the Au(111)-‘Herringbone,’ Au(110)-(1 × 2)-‘Missing-Row,’ and Au(100)-‘Quasi-Hexagonal’ reconstructions). This capability yields direct atomistic understanding of the dynamic emergence of these surface states from their initial facets, providing previously inaccessible information such as nucleation kinetics and a complete mechanistic interpretation of reconstruction under the effects of strain and local deviations from the original stoichiometry. We successfully reproduce previous experimental observations of reconstructions on pristine surfaces and provide quantitative predictions of the emergence of spinodal decomposition and localized reconstruction in response to strain at non-ideal stoichiometries. A unified mechanistic explanation is presented of the kinetic and thermodynamic factors driving surface reconstruction. Furthermore, we study surface reconstructions on Au nanoparticles, where characteristic (111) and (100) reconstructions spontaneously appear on a variety of high-symmetry particle morphologies.

36 MATERIALS SCIENCE↗
NASA NTRSDate not supplied

Evaluating the Efficacy of Conditional Variational Autoencoders in Generating Synthetic Single Nuclei RNA-Seq Data for Space Biology Research

Astronauts are subject to unique stressors during spaceflight, leading to changes in their cellular function. However, neither astronauts nor model organisms respond the same to spaceflight, and research implicates a contribution of omics components in differential responses. Understanding how gene expression affects astronaut health is critical for the success of long-term space missions, prompting interest in developing personalized predictive models leveraging artificial intelligence (AI) and machine learning (ML) techniques. Developing such models requires extensive data, which is challenging to obtain and share. This study explores the use of conditional variational autoencoders (CVAEs) to synthetically generate single-nuclei RNA-seq (snRNA-seq) data. CVAEs build on standard variational autoencoders (VAEs) by conditioning data generation on covariates like sample identity and mission parameters, enhancing the relevance of generated data for specific contexts. For our work, we built two CVAEs with varying degrees of sparsity to optimize both interpretability and generative power. We train and validate models on existing snRNA-seq data collected from the brain tissue of mice subjected to spaceflight conditions and their ground control counterparts. We evaluate model performance using statistical tests and visualizations to compare synthetic data to real data. We aim to demonstrate that these prototype CVAE architectures could be used in future space biology work and that this is a method worth further exploring.

Sarah Golts↗
DOE OSTI2021

Predicting Energetics Materials’ Crystalline Density from Chemical Structure by Machine Learning

To expedite new molecular compound development, a long-sought goal within the chemistry community has been to predict molecules’ bulk properties of interest a priori to synthesis from a chemical structure alone. In this work, we demonstrate that machine learning methods can indeed be used to directly learn the relationship between chemical structures and bulk crystalline properties of molecules, even in the absence of any crystal structure information or quantum mechanical calculations. We focus specifically on a class of organic compounds categorized as energetic materials called high explosives (HE) and predicting their crystalline density. An ongoing challenge within the chemistry machine learning community is deciding how best to featurize molecules as inputs into machine learning models—whether expert handcrafted features or learned molecular representations via graph-based neural network models—yield better results and why. We evaluate both types of representations in combination with a number of machine learning models to predict the crystalline densities of HE-like molecules curated from the Cambridge Structural Database, and we report the performance and pros and cons of our methods. Our message passing neural network (MPNN) based models with learned molecular representations generally perform best, outperforming current state-of-the-art methods at predicting crystalline density and performing well even when testing on a data set not representative of the training data. However, these models are traditionally considered black boxes and less easily interpretable. Here, to address this common challenge, we also provide a comparison analysis between our MPNN-based model and models with fixed feature representations that provides insights as to what features are learned by the MPNN to accurately predict density.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Transforming jet flavour tagging at ATLAS

Jet flavour tagging enables the identification of jets originating from heavy-flavour quarks in proton–proton collisions at the Large Hadron Collider, playing a critical role in its physics programmes. This paper presents GN2, a transformer-based flavour tagging algorithm deployed by the ATLAS Collaboration that represents a different methodology compared to previous approaches. Designed to classify jets based on the flavour of their constituent particles, GN2 processes low-level tracking information in an end-to-end architecture and incorporates physics-informed auxiliary training objectives to enhance both interpretability and performance. Its performance is validated in both simulation and collision data. The measured c-jet (light-jet) rejection in data is improved by a factor of 3.5 (1.8) for a 70% b-jet tagging efficiency, compared to the previous algorithm. GN2 provides substantial benefits for physics analyses involving heavy-flavour jets, such as measurements of Higgs boson pair production and the couplings of bottom and charm quarks to the Higgs boson, and demonstrates the impact of advanced machine learning methods in experimental particle physics.

Characterization and analytical techniques↗
DOE OSTI2022

Big Data Synchrophasor Monitoring and Analytics for Resiliency Tracking (BDSMART)

This report contains key findings from a project titled Big Data Synchrophasor Monitoring and Analytics for Resiliency Tracking (BDSMART), which was carried out through a collaborative effort of a team of researchers from Texas A&M Engineering Experiment Station, Temple University, and Quanta Technology, LLC. The in-kind support came from OSIsoft (acquired by AVEVA), which provided their PI Historian software to demonstrate the use case of streaming PMU data. The first section of the report describes the project goals and objectives related to the development of Machine Learning (ML) models capable of detecting and classifying events by processing phasor measurements captured in the field by Phasor Measurement Units (PMUs). The data for this study was contributed by the utilities/ISOs from the Western and Eastern interconnects and ERCOT, further referred to as Interconnect B (IC B), Interconnect A (IC A), and Interconnect C (IC C), respectively. The approach that the BDSMART Research Team proposed and the key research tasks defined by the team are outlined in this section. The next section describes the technical approach. We first discuss the data constraints related to the PMU measurements and data interpretation constraints imposed by the data contributors. They provided neither the topological information of the grid nor PMU placement locations and captured recorded data at very few locations in the system with the reporting rate of either 30 or 60 fps. The recordings are mostly positive sequence voltage, frequency, and ROCOF, and in some limited cases, three-phase voltages and currents. We then reflect on the bad data issues that stem from poor recording practices and vague definitions of the PMU status bits to supposedly be used for bad data identification. Finally, the data discovery points to imprecise time stamps with incomplete event start/end time, as well as inconsistent and incomplete event labeling, which combined make the implementation of the data models using supervising learning quite challenging. Following the data discovery study, we hypothesize that because the IC B data has the most complete labels, we should focus our model development on that data and then test it on data from other interconnects. We also define the common metrics used to evaluate the results from the ML algorithm tests. We concluded this section by summarizing the common ML models we used and explaining how we implemented and tested them. The issues from this section are expanded in the Training Dataset Report from this project. The final section of this report deals with the accomplishments and conclusions. As the accomplishments, we formulate the problem we are solving and what is achieved by solving the problem. We then reflect on each of the analytics tools we developed and point out the performance of each tool when applied to solving the mentioned problems. We reference this work for further details to the papers we published on each tool. In the conclusions, we give recommendations on how to improve future PMU recording practices to facilitate the ML algorithm implementation and guidance for the future standardization work aimed at clarifying the ambiguities associated with the PMU status bits. We finally list future tasks that can bring about further improvements in the proposed algorithms. The issues from this section are expanded in the Training, and Test Dataset Report filed at the project completion date.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Interpreting strain tensor data to characterize and monitor reservoirs for CO2 storage and other applications

Recent advances in instrumentation have made it feasible to measure the transient strain tensor caused by small changes in fluid volume or pressure in the subsurface and this has opened the door to new opportunities for characterization and monitoring. We have deployed strainmeters and then conducted injection well tests in an underlying reservoir at 530m depth. The resulting data indicated that the horizontal strain at shallow strainmeters (30 to 40m depth) was tensile and the vertical strain was compressive. The radial strain was less than the horizontal strain, and the strain rates decreased from 100 nanostrain/day to roughly 10 n/d over a few days. The signal at two strainmeters at shallow depth were consistent, although the magnitude of the horizontal strains were different reflecting the different radial directions from the well. The signal at a deep strainmeter deployed at reservoir depth was much different, with tensile vertical strains and compressive horizontal strains. These data can be interpreted by inverting poroelastic forward models developed using numerical and analytical methods. The average horizontal strain in the caprock resembles the transient pressure in the underlying reservoir and classic type-curve methods from transient well testing can be used for preliminary interpretations of strain data. We have developed closed-form analytical solutions to a pressurized poroelastic inclusion and inhomogeneity in a half-space. This model is fast and can be inverted to estimate reservoir stiffness and geometry. Numerical models developed using finite element methods allow more details of the subsurface to be included in the inversion, but they require much longer run times and this makes inversion cumbersome using standard methods. We have developed an inversion approach that uses a proxy model created using machine learning to do most of the forward calculations. The proxy model is periodically updated and refined using the finite element model to ensure accuracy. This approach significantly reduces the computational requirements and makes it feasible to use Bayesian inversion with large numerical models. We have shown that the strain tensor in the caprock is sensitive to pressure in the reservoir, boundaries in the reservoir, and pressure in the caprock caused by leaks. These results indicate that coupling strain tensor data with inversion has the potential to help evaluate reservoirs during initial characterization, and to monitor them during the CO2 injection and storage process.

Murdoch, Larry↗
DOE OSTI2025

Multi-fidelity equations of state and transport coefficient datasets for pulsed-power applications

Reliably simulating experiments relevant to the National Nuclear Security Administration (NNSA) requires a detailed description of material properties across a wide range of conditions. Such properties include the equations of state, charged-particle transport coefficients, and optical properties like the opacity. Together, these properties make up the material models used in radiation-magnetohydrodynamic simulations of nuclear fusion experiments. Many of these models do not incorporate uncertainties in the data used to produce them. It is unknown whether these uncertainties significantly impact the interpretation of simulation results and diagnostics. The purpose of this work is to quantify how such uncertainties impact simulations of pulsed-power experiments. We accomplished this task by first assessing discrepancies between approaches used to generate the data. This included bringing together members of the high-energy-density community spanning the three NNSA laboratories and multiple universities. Then, using these data, we developed a general framework that systematically incorporates physical uncertainties within the material models suitable for uncertainty quantification analyses. The framework utilizes machine learning, Bayesian inference, and incorporates multi-fidelity datasets. We demonstrated the framework by quantifying the impact that material model uncertainties have on simulations of pulsed-power experiments underway on Z at Sandia National Laboratories. As a result of this work, we discovered that modest uncertainties in material models (roughly 20%) correspond to significant uncertainties in the outputs from simulations. Our framework has enabled rapid construction of material models through an automated procedure and allows for the generation of material models of interest to the NNSA.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Supporting Risk-informed Decision-making During Reactor Accidents

Uncertainty in severe accident evolution and outcome is driven by event bifurcations that represent distinctive challenges to defensive layers and tend to promote the emergence of discrete classes of core damage and accident risk. This discrete set of "attractor" states arise from the complex networks of competing physical phenomena and conditional event cascades occurring as the overall system degrades – a process that yields increasing degrees of freedom and accident progression pathways. Characterization of these event spaces has proven elusive to more traditional data interrogation methods, but proves tractable by application of more advanced data collection and machine learning approaches. Through application of these approaches we demonstrate a conceptual framework that enables real-time/robust, risk-informed decision-making support to improve accident mitigation and encourage “graceful exits” during low probability, extreme events limiting accident consequences. In this analysis, we simulated over 8,000 short-term station blackout (STSBO) accidents with the state-of-the-art integral severe accident code, MELCOR, and demonstrate the potential for ML approaches to predict simulation outcomes. We chose to pair ML tools with interpretable and mechanistic event trees for the considered STSBO accident space to predict the likelihood of future event paths along the tree. In addition to the current state of the system, we use information from recent trajectories of temperature, pressure, and other physical features, combining both the current state and past trajectories to forecast future event paths. Finally, we simulate the random injection of variable amounts of water to quantify the efficacy of available actions at reducing risks along the many branches in the event tree. We identify scenarios and windows of opportunity to mitigate risk as well as scenarios in which such actions are unlikely to alter the accident end-state.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRS2009

Multi-Mission Automated Task Invocation Subsystem

Multi-Mission Automated Task Invocation Subsystem (MATIS) is software that establishes a distributed data-processing framework for automated generation of instrument data products from a spacecraft mission. Each mission may set up a set of MATIS servers for processing its data products. MATIS embodies lessons learned in experience with prior instrument- data-product-generation software. MATIS is an event-driven workflow manager that interprets project-specific, user-defined rules for managing processes. It executes programs in response to specific events under specific conditions according to the rules. Because requirements of different missions are too diverse to be satisfied by one program, MATIS accommodates plug-in programs. MATIS is flexible in that users can control such processing parameters as how many pipelines to run and on which computing machines to run them. MATIS has a fail-safe capability. At each step, MATIS captures and retains pertinent information needed to complete the step and start the next step. In the event of a restart, this information is retrieved so that processing can be resumed appropriately. At this writing, it is planned to develop a graphical user interface (GUI) for monitoring and controlling a product generation engine in MATIS. The GUI would enable users to schedule multiple processes and manage the data products produced in the processes. Although MATIS was initially designed for instrument data product generation,

Cheng, Cecilia S.↗
DOE OSTI2022

A review of computing-based automated fault detection and diagnosis of heating, ventilation and air conditioning systems

We report faults in Heating, Ventilation, and Air Conditioning (HVAC) systems of buildings result in significant energy waste in building operation. With fast-growing sensing data availability and advancement in computing, computational modeling has demonstrated strong capability to detect and diagnose HVAC system faults, hence, ensuring efficient building operation. This paper comprehensively reviews the state-of-the-art computing-based fault detection and diagnosis (FDD) for HVAC systems. Overall, the reviewed computing-based FDD methods are classified as two major approaches: knowledge-based and data-driven approaches. We then identify multiple important topics, including data availability, training data size, data quality, approach generality, capability, interpretability, and required modeling efforts, along with corresponding metrics to summarize the most updated FDD development. Generally, the knowledge-based approaches are further divided as physics-based modeling, Diagnostic Bayesian Network, and performance indicator-based methods while data-driven approaches include supervised learning, unsupervised learning, and regression and statistics-based methods. State-of-the-art FDD development, remaining challenges, and future research directions are further discussed to push forward FDD in practice. Availability of fault data, capability of existing methods to deal with complex fault situations (such as simultaneous faults), modeling interpretability for data-driven methods, and required engineering efforts for physics-based methods are identified as remaining challenges in FDD development. Improving modeling fidelity and reducing modeling efforts are essential for applying physics-based methods in real buildings. Meanwhile, addressing fault data availability, increasing algorithm adaptability, and handling multiple faults are essential to further enhance the applicability of data-driven FDD approaches.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Revealing Local Structures through Machine-Learning-Fused Multimodal Spectroscopy

Atomistic structures of materials offer valuable insights into their functionality. Determining these structures remains a fundamental challenge in materials science, especially for systems with defects. While both experimental and computational methods exist, each has limitations in resolving nanoscale structures. Core-level spectroscopies, such as X-ray absorption (XAS) or electron energy-loss spectroscopies (EELS), have been used to determine the local bonding environment and structure of materials. Recently, machine learning (ML) methods have been applied to extract structural and bonding information from XAS/EELS data. However, frameworks relying solely on a single data stream, defined as characterization data derived from a single element using one technique, are often insufficient because multiple local environments can yield similar spectral features, making it challenging to differentiate between competing structural hypotheses. Here, in this work, we address this challenge by integrating multimodal ab initio simulations, experimental data acquisition, and ML techniques for structure characterization. Our goal is to determine local structures and properties using EELS and XAS data from multiple elements and edges. To showcase our approach, we use various lithium nickel manganese cobalt (NMC) oxide compounds which are used for lithium ion batteries, including those with oxygen vacancies and antisite defects, as the sample material system. We successfully inferred local element content, ranging from lithium to transition metals, with quantitative agreement with experimental data. Beyond local element inference, we find that ML model based on multimodal spectroscopic data is able to determine whether local defects such as oxygen vacancy and antisites are present, a task which is impossible for single mode spectra or other experimental techniques. Furthermore, our framework is able to provide physical interpretability, bridging spectroscopy with the local atomic and electronic structures.

battery↗
DOE OSTI2025

Solving high-dimensional partial integral differential equations: The finite expression method

Partial integro-differential equations (PIDEs) have broad applications in the sciences, from electro-magnetism to options pricing. Here, in this paper, we introduce a new finite expression method (FEX) to solve PIDEs. This approach builds upon the original FEX and its inherent advantages with new advances: 1) A novel method of parameter grouping is proposed to reduce the number of coefficients in high-dimensional function approximation; 2) A Taylor series approximation method is implemented to significantly improve the computational efficiency and accuracy of the evaluation of the integral terms of PIDEs. The new FEX based method, denoted FEX-PG to indicate the addition of the parameter grouping (PG) step to the algorithm, provides both high accuracy and interpretable numerical solutions, with the outcome being an explicit equation that facilitates intuitive understanding of the underlying solution structures. These features are often absent in traditional methods, such as finite element methods (FEM) and finite difference methods, as well as in deep learning-based approaches. To benchmark our method against recent advances, we apply the new FEX-PG to solve benchmark PIDEs in the literature. In high-dimensional settings, FEX-PG exhibits strong and robust performance, achieving relative errors on the order of single precision machine epsilon, significantly outperforming existing approaches based on neural networks.

Combinatorial optimization↗
DOE OSTI2025

Multioutput Convolutional Neural Network for Improved Parameter Extraction in Time-Resolved Electrostatic Force Microscopy Data

Time-resolved scanning probe microscopy methods, like time-resolved electrostatic force microscopy (trEFM), enable imaging of dynamic processes ranging from ion motion in batteries to electronic dynamics in microstructured thin film semiconductors for solar cells. Reconstructing the underlying physical dynamics from these techniques can be challenging due to the interplay of cantilever physics with the actual transient kinetics of interest in the resulting signal. Previously, quantitative trEFM used empirical calibration of the cantilever or feed-forward neural networks trained on simulated data to extract the physical dynamics of interest. Both these approaches are limited by interpreting the underlying signal as a single exponential function, which serves as an approximation but does not adequately reflect many realistic systems. Here, we present a multi-branched, multi-output convolutional neural network (CNN) that uses the trEFM signal in addition to the physical cantilever parameters as input. The trained CNN accurately extracts parameters describing both single-exponential and bi-exponential underlying functions, and more accurately reconstructs real experimental data in the presence of noise. This article demonstrates an application of physics-informed machine learning to complex signal processing tasks, enabling more efficient and accurate analysis of trEFM.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗