Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Neural network compression”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Dynamical Low-Rank Compression of Neural Networks with Robustness under Adversarial Attacks

Deployment of neural networks on resource-constrained devices demands models that are both compact and robust to adversarial inputs. However, compression and adversarial robustness often conflict. In this work, we introduce a dynamical low-rank training scheme enhanced with a novel spectral regularizer that controls the condition number of the low-rank core in each layer. This approach mitigates the sensitivity of compressed models to adversarial perturbations without sacrificing clean accuracy. The method is model- and data-agnostic, computationally efficient, and supports rank adaptivity to automatically compress the network at hand. Extensive experiments across standard architectures, datasets, and adversarial attacks show the regularized networks can achieve over 94 compression while recovering or improving adversarial accuracy relative to uncompressed baselines.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗
DOE OSTI2024

Dark Energy Survey Year 3 results: Simulation-based cosmological inference with wavelet harmonics, scattering transforms, and moments of weak lensing mass maps. Validation on simulations

Beyond-two-point statistics contain additional information on cosmological as well as astrophysical and observational (systematics) parameters. In this methodology paper we provide an end-to-end simulationbased analysis of a set of Gaussian and non-Gaussian weak lensing statistics using detailed mock catalogs of the Dark Energy Survey (DES). Here, we implement: 1) second and third moments; 2) wavelet phase harmonics (WPH); 3) the scattering transform (ST). Our analysis is fully based on simulations, it spans a space of seven $νw$CDM cosmological parameters, and it forward models the most relevant sources of systematics of the data (masks, noise variations, clustering of the sources, intrinsic alignments, and shear and redshift calibration). We implement a neural network compression of the summary statistics, and we estimate the parameter posteriors using a likelihood-free-inference approach. We validate the pipeline extensively, and we find that WPH exhibits the strongest performance when combined with second moments, followed by ST, and then by third moments. The combination of all the different statistics further enhances constraints with respect to second moments, up to 25 percent, 15 percent, and 90 percent for S 8 , Ω m , and the figure-of-merit FoM S8;Ωm , respectively. We further find that non-Gaussian statistics improve constraints on w and on the amplitude of intrinsic alignment with respect to second moments constraints. The methodological advances presented here are suitable for application to Stage IV surveys from Euclid, Rubin-LSST, and Roman with additional validation on mock catalogs for each survey. In a companion paper we present an application to DES Year 3 data.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Dark Energy Survey Year 3 results: Simulation-based cosmological inference with wavelet harmonics, scattering transforms, and moments of weak lensing mass maps. II. cosmological results

Here, we present a simulation-based cosmological analysis using a combination of Gaussian and non-Gaussian statistics of the weak lensing mass (convergence) maps from the first three years of the Dark Energy Survey. We implement the following: (1) second and third moments; (2) wavelet phase harmonics; (3) the scattering transform. Our analysis is fully based on simulations, spans a space of seven 𝑤 Cold Dark Matter (𝑤⁢ CDM) cosmological parameters, and forward models the most relevant sources of systematics inherent in the data: masks, noise variations, clustering of the sources, intrinsic alignments, and shear and redshift calibration. We implement a neural network compression of the summary statistics, and we estimate the parameter posteriors using a simulation-based inference approach. Including and combining different non-Gaussian statistics is a powerful tool that strongly improves constraints over Gaussian statistics (in our case, the second moments); in particular, the figure of merit (𝑆 8 , Ω m ) is improved by 70% (Λ ⁢CDM) and 90% (𝑤 ⁢CDM). When all the summary statistics are combined, we achieve a 2% constraint on the amplitude of fluctuations parameter 𝑆 8 ≡ 𝜎 8 ⁢(Ω m /0.3) 0.5 , obtaining 𝑆 8 = 0.794 ±0.017 (Λ⁢ CDM) and 𝑆 8 = 0.817 ±0.021 (𝑤 ⁢CDM), and a ∼10% constraint on Ω m , obtaining Ω m =0.259 ±0.025 (Λ ⁢CDM) and Ω m = 0.273 ±0.029 (𝑤⁢ CDM). In the context of the 𝑤⁢ CDM scenario, these statistics also strengthen the constraints on the parameter 𝑤, obtaining 𝑤 <−0.72. The constraints from different statistics are shown to be internally consistent (with a 𝑝-value>0.1 for all combinations of statistics examined). We compare our results to other weak lensing results from the first three years of the Dark Energy Survey data, finding good consistency; we also compare with results from external datasets, such as planck constraints from the cosmic microwave background, finding statistical agreement, with discrepancies no greater than <2.2⁢𝜎.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1993

Image and Video Compression with VLSI Neural Networks

An advanced motion-compensated predictive video compression system based on artificial neural networks has been developed to effectively eliminate the temporal and spatial redundancy of video image sequences and thus reduce the bandwidth and storage required for the transmission and recording of the video signal. The VLSI neuroprocessor for high-speed high-ratio image compression based upon a self-organization network and the conventional algorithm for vector quantization are compared. The proposed method is quite efficient and can achieve near-optimal results.

compression artificial neural networks VLSI↗
NASA NTRS1996

VLSI Neural Networks Help To Compress Video Signals

Advanced analog/digital electronic system for compression of video signals incorporates artificial neural networks. Performs motion-estimation and image-data-compression processing. Effectively eliminates temporal and spatial redundancies of sequences of video images; processes video image data, retaining only nonredundant parts to be transmitted, then transmits resulting data stream in form of efficient code. Reduces bandwidth and storage requirements for transmission and recording of video signal.

Fang, Wai-Chi↗
DOE OSTI2025

Dark Energy Survey Year 3 results: $w$CDM cosmology from simulation-based inference with persistent homology on the sphere

We present cosmological constraints from Dark Energy Survey Year 3 (DES Y3) weak lensing data using persistent homology, a topological data analysis technique that tracks how features like clusters and voids evolve across density thresholds. For the first time, we apply spherical persistent homology to galaxy survey data through the algorithm TopoS2, which is optimized for curved-sky analyses and HEALPix compatibility. Employing a simulation-based inference framework with the Gower Street simulation suite, specifically designed to mimic DES Y3 data properties, we extract topological summary statistics from convergence maps across multiple smoothing scales and redshift bins. After neural network compression of these statistics, we estimate the likelihood function and validate our analysis against baryonic feedback effects, finding minimal biases (under $0.3σ$) in the $Ω_\mathrm{m}-S_8$ plane. Assuming the $w$CDM model, our combined Betti numbers and second moments analysis yields $S_8 = 0.821 \pm 0.018$ and $Ω_\mathrm{m} = 0.304\pm0.037$-constraints 70% tighter than those from cosmic shear two-point statistics in the same parameter plane. Our results demonstrate that topological methods provide a powerful and robust framework for extracting cosmological information, with our spherical methodology readily applicable to upcoming Stage IV wide-field galaxy surveys.

Prat, J. [Nordita; Royal Inst. Tech., Sodertalje; ↗
NASA NTRS1993

Digital Image Compression Using Artificial Neural Networks

The problem of storing, transmitting, and manipulating digital images is considered. Because of the file sizes involved, large amounts of digitized image information are becoming common in modern projects. Our goal is to described an image compression transform coder based on artificial neural networks techniques (NNCTC). A comparison of the compression results obtained from digital astronomical images by the NNCTC and the method used in the compression of the digitized sky survey from the Space Telescope Science Institute based on the H-transform is performed in order to assess the reliability of the NNCTC.

Serra-Ricart, M.↗
NASA NTRS1997

Motion video compression system with neural network having winner-take-all function

A motion video data system includes a compression system, including an image compressor, an image decompressor correlative to the image compressor having an input connected to an output of the image compressor, a feedback summing node having one input connected to an output of the image decompressor, a picture memory having an input connected to an output of the feedback summing node, apparatus for comparing an image stored in the picture memory with a received input image and deducing therefrom pixels having differences between the stored image and the received image and for retrieving from the picture memory a partial image including the pixels only and applying the partial image to another input of the feedback summing node, whereby to produce at the output of the feedback summing node an updated decompressed image, a subtraction node having one input connected to received the received image and another input connected to receive the partial image so as to generate a difference image, the image compressor having an input connected to receive the difference image whereby to produce a compressed difference image at the output of the image compressor.

Fang, Wai-Chi↗
NASA NTRS1991

Video data compression using artificial neural network differential vector quantization

An artificial neural network vector quantizer is developed for use in data compression applications such as Digital Video. Differential Vector Quantization is used to preserve edge features, and a new adaptive algorithm, known as Frequency-Sensitive Competitive Learning, is used to develop the vector quantizer codebook. To develop real time performance, a custom Very Large Scale Integration Application Specific Integrated Circuit (VLSI ASIC) is being developed to realize the associative memory functions needed in the vector quantization algorithm. By using vector quantization, the need for Huffman coding can be eliminated, resulting in superior performance against channel bit errors than methods that use variable length codes.

Krishnamurthy, Ashok K.↗
NASA NTRS1989

Neural networks for data compression and invariant image recognition

An approach to invariant image recognition (I2R), based upon a model of biological vision in the mammalian visual system (MVS), is described. The complete I2R model incorporates several biologically inspired features: exponential mapping of retinal images, Gabor spatial filtering, and a neural network associative memory. In the I2R model, exponentially mapped retinal images are filtered by a hierarchical set of Gabor spatial filters (GSF) which provide compression of the information contained within a pixel-based image. A neural network associative memory (AM) is used to process the GSF coded images. We describe a 1-D shape function method for coding of scale and rotationally invariant shape information. This method reduces image shape information to a periodic waveform suitable for coding as an input vector to a neural network AM. The shape function method is suitable for near term applications on conventional computing architectures equipped with VLSI FFT chips to provide a rapid image search capability.

Gardner, Sheldon↗
NASA NTRS2020

Neural Network Repair of Lossy Compression Artifacts in the Sept 2015 – March 2016 Duration of the MMS/FPI Dataset

During the Sept 2015 –March 2016 duration (sometimes referred to as Phase 1A) of the Magnetospheric Multiscale Mission (MMS), the Dual Electron Spectrometers (DES) were configured to generously utilize lossy compression. While this maximized the number of velocity distribution functions downlinked, it cameat the expense of lost information content for a fractionof the frames. Following this period of lossy compression, the DES was re-configured in a way that allowed for 95% of the framesto arrive to the ground without loss. Using this high-quality set offrameson-orbit observations, we compressed and decompressed the frameson the ground to create a side-by-side record of the compression effect. This record was used to drive an optimization method that (a) derived basis functions capable of approximating the lossless sample space and with non-negative coefficients and (b) fitted a function which maps the lossy framesto basis weights that recreate the framewithout compression artifacts. This methodis introduced and evaluated in this paper.Data users should expect a higher level of confidence in the absolute scale of density/temperature measurements andnotice less sinusoidal bias in the velocity X and Y components(GSE).

Magnetosphere, Neural Networks, Image Processing, ↗
DOE OSTI2024

Spike-and-Slab Shrinkage Priors for Structurally Sparse Bayesian Neural Networks

Network complexity and computational efficiency have become increasingly significant aspects of deep learning. Sparse deep learning addresses these challenges by recovering a sparse representation of the underlying target function by reducing heavily overparameterized deep neural networks. Specifically, deep neural architectures compressed via structured sparsity (e.g., node sparsity) provide low-latency inference, higher data throughput, and reduced energy consumption. In this article, we explore two well-established shrinkage techniques, Lasso and Horseshoe, for model compression in Bayesian neural networks (BNNs). To this end, we propose structurally sparse BNNs, which systematically prune excessive nodes with the following: 1) spike-and-slab group Lasso (SS-GL) and 2) SS group Horseshoe (SS-GHS) priors, and develop computationally tractable variational inference, including continuous relaxation of Bernoulli variables. We establish the contraction rates of the variational posterior of our proposed models as a function of the network topology, layerwise node cardinalities, and bounds on the network weights. Furthermore, we empirically demonstrate the competitive performance of our models compared with the baseline models in prediction accuracy, model compression, and inference latency.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Bayesian tensorized neural networks with automatic rank selection

Tensor decomposition is an effective approach to compress over-parameterized neural networks and to enable their deployment on resource-constrained hardware platforms. However, directly applying tensor compression in the training process is a challenging task due to the difficulty of choosing a proper tensor rank. In order to address this challenge, this paper proposes a low-rank Bayesian tensorized neural network. Our Bayesian method performs automatic model compression via an adaptive tensor rank determination. We also present approaches for posterior density calculation and maximum a posteriori (MAP) estimation for the end-to-end training of our tensorized neural network. Here, we provide experimental validation on a two-layer fully connected neural network, a 6-layer CNN and a 110-layer residual neural network where our work produces 7.4x to 137x more compact neural networks directly from the training while achieving high prediction accuracy.

Low-rank tensor↗
DOE OSTI2026

Variable rate neural compression for sparse detector data

Particle colliders produce data at extraordinary rates, posing major challenges for transmission and storage. High-throughput compression algorithms are therefore essential. In the sPHENIX experiment taking data at the Relativistic Heavy Ion Collider, a time projection chamber records three-dimensional (3D) particle trajectories that are highly sparse, making conventional learning-free lossy compression ineffective. Convolutional neural networks have surpassed traditional methods in compression ratio and accuracy. However, they fail to exploit sparsity for efficiency. To address these gaps, we present BCAE-VS, a bicephalous convolutional autoencoder with variable compression ratio for sparse data, which adapts compression to input complexity through key-point identification and sparse convolution. BCAE-VS achieves higher accuracy and compression ratios than prior neural approaches while being orders of magnitude smaller. Moreover, its throughput increases with sparsity—a property not observed in other methods. Although it was developed for collider experiments, BCAE-VS readily extends to other sparse data domains, such as light detection and ranging (LiDAR) sensing and 3D microscopy.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗