Search NASA⌕ Search

SEARCH · Search NASA

Results for “compression algorithm”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Bit Grooming: statistically accurate precision-preserving quantization with compression, evaluated in the netCDF Operators (NCO, v4.4.8+)

Abstract. Geoscientific models and measurements generate false precision (scientifically meaningless data bits) that wastes storage space. False precision can mislead (by implying noise is signal) and be scientifically pointless, especially for measurements. By contrast, lossy compression can be both economical (save space) and heuristic (clarify data limitations) without compromising the scientific integrity of data. Data quantization can thus be appropriate regardless of whether space limitations are a concern. We introduce, implement, and characterize a new lossy compression scheme suitable for IEEE floating-point data. Our new Bit Grooming algorithm alternately shaves (to zero) and sets (to one) the least significant bits of consecutive values to preserve a desired precision. This is a symmetric, two-sided variant of an algorithm sometimes called Bit Shaving that quantizes values solely by zeroing bits. Our variation eliminates the artificial low bias produced by always zeroing bits, and makes Bit Grooming more suitable for arrays and multi-dimensional fields whose mean statistics are important. Bit Grooming relies on standard lossless compression to achieve the actual reduction in storage space, so we tested Bit Grooming by applying the DEFLATE compression algorithm to bit-groomed and full-precision climate data stored in netCDF3, netCDF4, HDF4, and HDF5 formats. Bit Grooming reduces the storage space required by initially uncompressed and compressed climate data by 25–80 and 5–65 %, respectively, for single-precision values (the most common case for climate data) quantized to retain 1–5 decimal digits of precision. The potential reduction is greater for double-precision datasets. When used aggressively (i.e., preserving only 1–2 digits), Bit Grooming produces storage reductions comparable to other quantization techniques such as Linear Packing. Unlike Linear Packing, whose guaranteed precision rapidly degrades within the relatively narrow dynamic range of values that it can compress, Bit Grooming guarantees the specified precision throughout the full floating-point range. Data quantization by Bit Grooming is irreversible (i.e., lossy) yet transparent, meaning that no extra processing is required by data users/readers. Hence Bit Grooming can easily reduce data storage volume without sacrificing scientific precision or imposing extra burdens on users.

58 GEOSCIENCES↗
DOE OSTI2017

Reconstructing three-dimensional protein crystal intensities from sparse unoriented two-axis X-ray diffraction patterns

Recently, there has been a growing interest in adapting serial microcrystallography (SMX) experiments to existing storage ring (SR) sources. For very small crystals, however, radiation damage occurs before sufficient numbers of photons are diffracted to determine the orientation of the crystal. The challenge is to merge data from a large number of such `sparse' frames in order to measure the full reciprocal space intensity. To simulate sparse frames, a dataset was collected from a large lysozyme crystal illuminated by a dim X-ray source. The crystal was continuously rotated about two orthogonal axes to sample a subset of the rotation space. With the EMC algorithm [expand–maximize–compress; Loh & Elser (2009). Phys. Rev. E , 80 , 026705], it is shown that the diffracted intensity of the crystal can still be reconstructed even without knowledge of the orientation of the crystal in any sparse frame. Moreover, parallel computation implementations were designed to considerably improve the time and memory scaling of the algorithm. The results show that EMC-based SMX experiments should be feasible at SR sources.

Chemistry↗
DOE OSTI2021

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multi-threaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0x and 6.8x on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3x over the multi-thread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗
DOE OSTI2021

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multithreaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0× and 6.8× on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3× over the multithread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗
DOE OSTI2013

Kalman-filtered compressive sensing for high resolution estimation of anthropogenic greenhouse gas emissions from sparse measurements.

The estimation of fossil-fuel CO2 emissions (ffCO2) from limited ground-based and satellite measurements of CO2 concentrations will form a key component of the monitoring of treaties aimed at the abatement of greenhouse gas emissions. The limited nature of the measured data leads to a severely-underdetermined estimation problem. If the estimation is performed at fine spatial resolutions, it can also be computationally expensive. In order to enable such estimations, advances are needed in the spatial representation of ffCO2 emissions, scalable inversion algorithms and the identification of observables to measure. To that end, we investigate parsimonious spatial parameterizations of ffCO2 emissions which can be used in atmospheric inversions. We devise and test three random field models, based on wavelets, Gaussian kernels and covariance structures derived from easily-observed proxies of human activity. In doing so, we constructed a novel inversion algorithm, based on compressive sensing and sparse reconstruction, to perform the estimation. We also address scalable ensemble Kalman filters as an inversion mechanism and quantify the impact of Gaussian assumptions inherent in them. We find that the assumption does not impact the estimates of mean ffCO2 source strengths appreciably, but a comparison with Markov chain Monte Carlo estimates show significant differences in the variance of the source strengths. Finally, we study if the very different spatial natures of biogenic and ffCO2 emissions can be used to estimate them, in a disaggregated fashion, solely from CO2 concentration measurements, without extra information from products of incomplete combustion e.g., CO. We find that this is possible during the winter months, though the errors can be as large as 50%.

Ray, Jaideep↗
DOE OSTI2020

Ghost imaging second harmonic generation microscopy

Second harmonic generation (SHG) microscopy is useful for visualizing interfaces and sub-structures within a wide range of materials due to the propensity for SHG to occur in non-centrosymmetric environments. However, since SHG is a nonlinear process generally necessitating small focal sizes for higher peak powers, a raster scanning approach is usually needed to build an SHG image over a significant sample size. While raster scanning is effective, there is a cost in terms of the time needed to acquire the image and, also, some materials cannot withstand the higher optical intensities within the small focal volume. As such, we describe a SHG microscopy approach based on ghost imaging (GI), which enables imaging data to be collected in parallel rather than sequentially as in raster scanning techniques. We experimentally demonstrate the approach and combine GI-SHG with compressive sensing to make further substantial gains in reducing the amount of sampling required for image reconstruction. Furthermore, GI-SHG is shown to have significant advantages for imaging in highly scattering environments, partly because GI is a background-free approach requiring spatial correlations between photons that travel two paths, with one path entirely devoid of sample interaction. This basic property of GI means that only the photons that travel unimpeded through the sample preserve the spatial correlations needed to reconstruct the image, while more scattered photons do not contribute to the overall GI signal. Finally, we compare the image quality and sampling properties of three different reconstruction algorithms used for compressive sensing.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

A hybrid finite volume method and smoothed particle hydrodynamics approach for efficient and accurate blast simulations

Modeling strong shock waves in fluids remains a persistent challenge in computational physics. Essential to research efforts in industry and defense, numerous methods have been devised to improve the accuracy and efficiency of shock simulations. A novel, hybrid Finite Volume Method (FVM)-Smoothed Particle Hydrodynamics (SPH) approach is capable of further improving efficiency and retaining accuracy by exploiting the favorable characteristics of each respective method. This hybrid approach is presented for shock capturing in compressible fluids. The Python framework Pyro2 is employed to simulate a coarse FVM mesh, while the Python framework PySPH is utilized to model the fluid in regions with high gradients through SPH particles. The performance of the hybrid FVM-SPH scheme, compared to the individual FVM and SPH methods, is assessed in 1 kt and 10 kt blast simulations. Our results indicate that the hybrid approach offers higher computational efficiency than SPH while preserving its accuracy and characteristics. The hybrid approach had a relative speedup of 11.3x and 22.3x over the FVM and SPH approaches for the 1 kt simulation and a relative speedup of 14.7x and 20.9x over the FVM and SPH approaches for the 10 kt simulation. The hybrid SPH algorithm enables future compressible fluid simulations with more extensive capabilities than grid-based methods alone, presenting potential applications in modeling fluid-structure interactions and solid deformation and fracturing in blast simulations.

Myers, Conner↗
DOE OSTI2020

Online adaptive basis refinement and compression for reduced-order models via vector-space sieving

In many applications, projection-based reduced-order models (ROMs) have demonstrated the ability to provide rapid approximate solutions to high-fidelity full-order models (FOMs). However, there is no a priori assurance that these approximate solutions are accurate; their accuracy depends on the ability of the low-dimensional trial basis to represent the FOM solution. As a result, ROMs can generate inaccurate approximate solutions, e.g., when the FOM solution at the online prediction point is not well represented by training data used to construct the trial basis. To address this fundamental deficiency of standard model-reduction approaches, this work proposes a novel online-adaptive mechanism for efficiently enriching the trial basis in a manner that ensures convergence of the ROM to the FOM, yet does not incur any FOM solves. The mechanism is based on the previously proposed adaptive $h$-refinement method for ROMs, but improves upon this work in two crucial ways. First, the proposed method enables basis refinement with respect to any orthogonal basis (not just the Kronecker basis), thereby generalizing the refinement mechanism and enabling it to be tailored to the physics characterizing the problem at hand. Second, the proposed method provides a fast online algorithm for periodically compressing the enriched basis via an efficient proper orthogonal decomposition (POD) method, which does not incur any operations that scale with the FOM dimension. We report these two features allow the proposed method to serve as (1) a failsafe mechanism for ROMs, as the method enables the ROM to satisfy any prescribed error tolerance online (even in the case of inadequate training), and (2) an efficient online basis-adaptation mechanism, as the combination of basis enrichment and compression enables the basis to adapt online while controlling its dimension.

42 ENGINEERING↗
DOE OSTI2013

Simultaneous reduction of radiation dose and scatter for CBCT by using collimators

Purpose: On-board cone-beam CT (CBCT) imaging has been widely available in radiotherapy clinic for target localization. However, the extra radiation dose from CBCT is always a concern for its frequent use. Additionally, the relatively large scatter in CBCT often degrades the image quality. By using collimators, some of the X-rays can be stopped from reaching the patient and the detectors, hence both the scatter and the patient doses are simultaneously reduced. The authors show in this work that the collimated CBCT data can be reconstructed without any noticeable artifacts for certain collimator blocking ratios and blocking patterns, and the focus of this work is to study the relationship between the image quality and these two collimator factors.Methods: A CBCT system with collimators was simulated following the typical geometry used in clinic. Different collimator designs were tested by varying the size and the number of the collimator slits, and at the same time, the ratio of transmitted beams to total beams was varied from 100% to 10%, resulting in hundreds of different simulation scenarios. Lung and pelvis phantoms created from patients CT images were used in the simulations, and an iterative reconstruction algorithm using the compressed sensing technique was adopted. The image quality was examined by root mean square errors (RMSEs) and compared with the conventional CBCT images.Results: The CBCT image quality increases as the amount of beams passing through the collimators increases, and decreases as the size of the collimator slits increases. With ultra-high resolution collimators, the RMSEs were comparable to the conventional CBCT image quality until the beam transmission rate is reduced below 25%.Conclusions: Collimators can reduce the scatters and radiation dose, however, the collimated CBCT image quality is strongly dependent on both the collimator blocking ratio and the blocking pattern. To achieve image quality comparable to the conventional CBCT, the amount of information and data format must be adequate.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2017

Scientific data interpolation with low dimensional manifold model

Here, we propose to apply a low dimensional manifold model to scientific data interpolation from regular and irregular samplings with a significant amount of missing information. The low dimensionality of the patch manifold for general scientific data sets has been used as a regularizer in a variational formulation. The problem is solved via alternating minimization with respect to the manifold and the data set, and the Laplace–Beltrami operator in the Euler–Lagrange equation is discretized using the weighted graph Laplacian. Various scientific data sets from different fields of study are used to illustrate the performance of the proposed algorithm on data compression and interpolation from both regular and irregular samplings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Recovering fine details from under-resolved electron tomography data using higher order total variation ℓ 1 regularization

Over the last decade or so, reconstruction methods using ℓ 1 regularization, often categorized as compressed sensing (CS) algorithms, have significantly improved the capabilities of high fidelity imaging in electron tomography. The most popular ℓ 1 regularization approach within electron tomography has been total variation (TV) regularization. In addition to reducing unwanted noise, TV regularization encourages a piecewise constant solution with sparse boundary regions. In this paper we propose an alternative ℓ 1 regularization approach for electron tomography based on higher order total variation (HOTV). Like TV, the HOTV approach promotes solutions with sparse boundary regions. In smooth regions however, the solution is not limited to piecewise constant behavior. We demonstrate that this allows for more accurate reconstruction of a broader class of images – even those for which TV was designed for – particularly when dealing with pragmatic tomographic sampling patterns and very fine image features. In conclusion, we develop results for an electron tomography data set as well as a phantom example, and we also make comparisons with discrete tomography approaches.

47 OTHER INSTRUMENTATION↗
DOE OSTI2017

Cryo-STEM Tomography with Inpainting

Traditionally, microscopists have worked with the Nyquist-Shannon theory of sampling, which states that to be able to reconstruct the image fully it needs to be sampled at a rate of at least twice the highest spatial frequency in the image. This sampling rate assumes that the image is sampled at regular intervals and that every pixel contains information that is crucial for the image (it even assumes that noise is important). Images in general, and especially low dose S/TEM images, contain significantly less information than can be encoded by a grid of pixels (which is why image compression works). Mathematically speaking, the image data has a low dimensional or sparse representation. Through the application of compressive sensing methods [1,2,3] this representation can be found using pre-designed measurements that are usually random for implementation simplicity. These measurements and the compressive sensing reconstruction algorithms have the added benefit of reducing noise. This reconstruction approach can be extended into higher dimensions, whereby the random sampling in each 2-D image can be extended into: a sequence of tomographic projections (i.e. tilt images); a sequence of video frames (i.e. incorporating temporal resolution and dynamics); spectral resolution (i.e. energy filtering an image to see the distribution of elements); and ptychography (i.e. sampling a full diffraction image at each location in a 2-D grid across the sample). This approach has been employed experimentally for materials science samples requiring low-dose imaging [2], and can be readily applied to biological samples. Figure 1 shows the resolution possible in a complex biological system, mouse pancreatic islet beta cells [4], when tomogram slices are reconstructed using subsampling. Reducing the number of pixels (1/6 pix and 1/3*1/3) shows minimal degradation compared to the reconstructions using all pixels (all data and 1/3 tilt). Although subsampling 1/6 of the tilts (1/6 of overall dose) degrades the reconstruction to the point that the cellular structures cannot be identified. Using 1/3 of both the pixels and the tilts provides a high quality image at 1/9 the overall dose even for this most basic and rapid demonstration of the CS methods. Figure 2 demonstrates the theoretical tomogram reconstruction quality (vertical axis) as undersampling (horizontal axis) is increased; we examined subsampling pixels and tilt-angles individually and a combined approach in which both pixels and tilts are sub-sampled. Note that subsampling pixels maintains high quality reconstructions (solid lines). Using the inpainting algorithm to obtain tomograms can automatically reduce the dose applied to the system by an order of magnitude. Perhaps the best way to understand the impact is to consider that by using inpainting (and with minimal hardware changes), a sample that can normally withstand a dose of ~10 e/Å2 can potentially be imaged with an “equivalent quality” to a dose level of 103 e/Å2. To put this in perspective, this is approaching the dose level used for the most advanced images, in terms of spatial resolution, for inorganic systems. While there are issues for biological specimens beyond dose (structural complexity being the most important one), this sampling approach allows the methods that are traditionally used for materials science to be applied to biological systems [5]. References: [1] A Stevens, H Yang, L Carin et al. Microscopy 63(1), (2014), pp. 41. [2] L Kovarik, A Stevens, A Liyu et al. Appl. Phys. Lett. 109, 164102 (2016) [3] A Stevens, L Kovarik, P Abellan et al. Adv. Structural and Chemical Imaging 1(10), (2015), pp. 1. [4] MD Guay, W Czaja, MA Aronova et al. Scientific Reports 6, 27614 (2016) [5] Supported by the Chemical Imaging, Signature Discovery, and Analytics in Motion Initiatives at PNNL. PNNL is operated by Battelle Memorial Inst. for the US DOE; contract DE-AC05-76RL01830.

Stevens, Andrew↗
DOE OSTI2014

Few-view cone-beam CT reconstruction with deformed prior image

Purpose: Prior images can be incorporated into the image reconstruction process to improve the quality of subsequent cone-beam CT (CBCT) images from sparse-view or low-dose projections. The purpose of this work is to develop a deformed prior image-based reconstruction (DPIR) strategy to mitigate the deformation between the prior image and the target image. Methods: The deformed prior image is obtained by a projection-based registration approach. Specifically, the deformation vector fields used to deform the prior image are estimated through iteratively matching the forward projection of the deformed prior image and the measured on-treatment projections. The deformed prior image is then used as the prior image in the standard prior image constrained compressed sensing (PICCS) algorithm. A simulation study on an XCAT phantom and a clinical study on a head-and-neck cancer patient were conducted to evaluate the performance of the proposed DPIR strategy. Results: The deformed prior image matches the geometry of the on-treatment CBCT more closely as compared to the original prior image. Consequently, the performance of the DPIR strategy from few-view projections is improved in comparison to the standard PICCS algorithm, based on both visual inspection and quantitative measures. In the XCAT phantom study using 20 projections, the average root mean squared error is reduced from 14% in PICCS to 10% in DPIR, and the average universal quality index increases from 0.88 in PICCS to 0.92 in DPIR. Conclusions: The present DPIR approach provides a practical solution to the mismatch problem between the prior image and target image, which improves the performance of the original PICCS algorithm for CBCT reconstruction from few-view or low-dose projections.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Equation of State of a Natural Chromian Spinel at Ambient Temperature

A natural chromian spinel with the composition (Mg0.48(3)Fe0.52(3))(Fe0.06(1)Al0.28(1)Cr0.66(2))2O4 was investigated up to 15 GPa via synchrotron X-ray diffraction with a diamond-anvil cell at room temperature. No phase transition was clearly observed up to the maximum experimental pressure. The pressure–volume data fitted to the third-order Birch–Murnaghan equation of state yielded an isothermal bulk modulus ( K T 0 ) of 207(5) GPa and its first pressure derivative ( K T 0 ' ) of 3.2(7), or K T 0 = 202(2) GPa with K T 0 ' fixed as 4. With this new experimental result and the results on some natural chromian spinels in the literature, a simple algorithm describing the relation between the K T 0 and the compositions of the natural chromian spinels was proposed. To examine this algorithm further, more compression experiments should be performed on natural chromian spinels with different chemical compositions.

Mi, Zhongying↗
DOE OSTI2024

A Framework for Compressing Unstructured Scientific Data via Serialization

We present a general framework for compressing unstructured scientific data with known local connectivity. A common application is simulation data defined on arbitrary finite element meshes. The framework employs a greedy topology preserving reordering of original nodes which allows for seamless integration into existing data processing pipelines. This reordering process depends solely on mesh connectivity and can be performed offline for optimal efficiency. However, the algorithm’s greedy nature also supports on-the-fly implementation. The proposed method is compatible with any compression algorithm that leverages spatial correlations within the data. The effectiveness of this approach is demonstrated on a large-scale real dataset using several compression methods, including MGARD, SZ, and ZFP.

Reshniak, Viktor [ORNL] (ORCID:0000000315454462)↗
DOE OSTI2022

An Algorithmic and Software Pipeline for Very Large Scale Scientific Data Compression with Error Guarantees

Efficient data compression is becoming increasingly critical for storing scientific data because many scientific applications produce vast amounts of data. This paper presents an end-to-end algorithmic and software pipeline for data compression that guarantees both error bounds on primary data (PD) and derived data, known as Quantities of Interest (QoI).We demonstrate the effectiveness of the pipeline by compressing fusion data generated by a large-scale fusion code, XGC, which produces tens of petabytes of data in a single day. We demonstrate that the compression is conducted by setting aside computational resources known as staging nodes, and does not impact the simulation performance. For efficient parallel I/O, the pipeline uses ADIOS2, which many codes such as XGC already use for their parallel I/O. We show that our approach can compress the data by two orders of magnitude while guaranteeing high accuracy on both the PD and the QoIs. Further, the amount of resources required by compression is a few percent of the resources required by simulation while ensuring that the compression time for each stage is less than the corresponding simulation time.This pipeline consists of three main steps. The first step decomposes the data using domain decomposition into small subdomains. Each subdomain is then compressed independently to achieve a high level of parallelism. The second step uses existing techniques that guarantee error bounds on the primary data for each subdomain. The third step uses a post-processing optimization technique based on Lagrange multipliers to reduce the QoI errors for data corresponding to each subdomain. The Lagrange multipliers generated can be further quantized or truncated to increase the compression level. All of the above characteristics of our approach make it highly practical to apply on-the-fly compression while guaranteeing errors on QoIs that are critical to the scientists.

Banerjee, Tania↗
DOE OSTI2023

Quantum mixed state compiling

The task of learning a quantum circuit to prepare a given mixed state is a fundamental quantum subroutine. We present a variational quantum algorithm (VQA) to learn mixed states which is suitable for near-term hardware. Our algorithm represents a generalization of previous VQAs that aimed at learning preparation circuits for pure states. We consider two different ansätze for compiling the target state; the first is based on learning a purification of the state and the second on representing it as a convex combination of pure states. In both cases, the resources required to store and manipulate the compiled state grow with the rank of the approximation. Thus, by learning a lower rank approximation of the target state, our algorithm provides a means of compressing a state for more efficient processing. As a byproduct of our algorithm, one effectively learns the principal components of the target state, and hence our algorithm further provides a new method for principal component analysis. We investigate the efficacy of our algorithm through extensive numerical implementations, showing that typical random states and thermal states of many body systems may be learnt this way. Additionally, we demonstrate on quantum hardware how our algorithm can be used to study hardware noise-induced states.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Maximizing efficiency of dataset compression for machine learning potentials with information theory

Machine learning interatomic potentials (MLIPs) balance high accuracy and lower costs compared to density functional theory calculations, but their performance often depends on the size and diversity of training datasets. Large datasets improve model accuracy and generalization but are computationally expensive to produce and train on, while smaller datasets risk discarding rare but important atomic environments and compromising MLIP accuracy/reliability. Here, we develop an information-theoretical framework to quantify the efficiency of dataset compression methods and propose an algorithm that maximizes this efficiency. By framing atomistic dataset compression as an instance of the minimum set cover (MSC) problem over atom-centered environments, our method identifies the smallest subset of structures that contains as much information as possible from the original dataset while pruning redundant information. The approach is extensively demonstrated on the GAP-20 and TM23 datasets and validated on 64 varied datasets from the ColabFit repository. Across all cases, MSC consistently retains outliers, preserves dataset diversity, and reproduces the long-tail distributions of forces even at high compression rates, outperforming other subsampling methods. Furthermore, MLIPs trained on MSC-compressed datasets exhibit reduced error for out-of-distribution data even in low-data regimes. We explain these results using an outlier analysis and show that such quantitative conclusions could not be achieved with conventional dimensionality reduction methods. The algorithm is implemented in the open-source QUESTS package and can be used for several tasks in atomistic modeling, from data subsampling, outlier detection, and training improved MLIPs at a lower cost.

36 MATERIALS SCIENCE↗