Search NASA⌕ Search

SEARCH · Search NASA

Results for “pruning”

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.

29 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

The Effects of Compounded Model Size Reductions on Adversarial Robustness

Recent advances in Edge AI and Tiny Machine Learning (TinyML) have enabled the deployment of machine learning models on resource-constrained environments. However, deploying these models on edge devices, such as micro-controllers, requires significant model footprint reduction through a variety of techniques such as quantization, pruning, and clustering. While these optimization methods offer considerable advantages, they potentially introduce AI-related security vulnerabilities, particularly concerning model robustness with respect to adversarial AI attacks. Prior research has extensively examined the impact of quantization on adversarial robustness; however, the effects of alternative reduction techniques and their combinations remain understudied. This paper investigates the impact of model size reduction techniques on adversarial robustness, when applied individually and combined. We utilized Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks to generate adversarial perturbations for both training and testing data, and then evaluated the models' accuracy under adversarial training conditions. Our findings revealed that reduction techniques generally diminished robustness; although, combining techniques was not found to make robustness any worse than when applied individually. Moreover, specific techniques can potentially enhance resistance to small size perturbations. This research provides insights into the trade-offs between model size reduction and security, establishing a foundation for future investigations into improving adversarial training techniques and methodologies for maintaining robustness while preserving memory footprint benefits.

Austria, Phillipe [ORNL] (ORCID:0000000236223973)↗
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 OSTI2025

Eureka: Enabling Fine-Grained Access and Range Queries on Compressed Scientific Data via Data-Index Co-Compression

Handling large-scale scientific data in high-performance computing (HPC) environments poses significant challenges, including excessive I/O, high storage costs, and slow query performance. Traditional approaches often require full data decompression and scans, making them impractical for real-time or interactive analysis. To address these limitations, we introduce Eureka, a unified data-index co-compression framework that enables fine-grained access and efficient range queries on compressed scientific datasets. Eureka integrates spatial domain decomposition with block-wise error-bounded lossy compression to support selective decompression. It constructs a hierarchical AVL-tree index during compression to capture block-level value ranges, enabling fast pruning during query execution. To reduce metadata overhead, the index itself is also compressed while ensuring recall-preserving results. Experiments on six diverse HPC simulation datasets show that Eureka achieves up to 25x data compression and over 300x index compression, surpassing state-of-the-art compressors such as SZ3 and ZFP in rate-distortion performance. Additionally, Eureka delivers over 30x speedup for low-selectivity range queries, making it a scalable and efficient solution for modern scientific data analysis.

Yan, Ning↗
DOE OSTI2025

Using AMReX's Embedded Boundaries to Support MFIX-Exa’s Geometry Capabilities

MFIX-Exa is an AMReX-based application used for the simulation of chemically reacting multiphase flows inside reactors. The reactor geometry is mostly a combination of primitive shapes such as cylinders, cubes and spheres of varying dimensions connected together with potentially a lot of dead space between them. Additionally, the reactors may have particle-laden jets introduced at different parts of their geometry to enhance reactant mixing. This talk presents how MFIX-Exa uses OpenSCAD’s csg format to allow users to define the embedded boundary (EB) geometry and how it supports specifying inflow through certain sections of the EB. The talk will describe how MFIX-Exa leverages the EB features of grid-pruning, restart files and redistribution. Finally, the talk goes over some capabilities which are currently not available in MFIX-Exa but planned for the future, such as moving parts of the reactor geometry.

embedded boundaries↗
DOE OSTI2025

Scalable workflow for evaluating and optimizing large language models

This work describes the improved workflow for evaluating open-source large language models (LLMs) for trustworthiness. The workflow facilitates the acquisition of LLMs, the generation of LLM responses, and the evaluation of the responses for their trustworthiness. As a use case, the workflow is employed to evaluate dense, quantized, and pruned Meta Llama3.1 LLMs for their truthfulness. The outcome of the project could set the stage for understanding and developing trustworthy models in the future projects.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Low-latency Jet Tagging for HL-LHC Using Transformer Architectures

Transformers are the state-of-the-art model architectures and widely used in application areas of machine learning. However the performance of such architectures is less well explored in the ultra-low latency domains where deployment on FPGAs or ASICs is required. Such domains include the trigger and data acquisition systems of the LHC experiments. We present a transformer-based algorithm for jet tagging built with the HGQ2 framework, which is able to produce a model with heterogeneous bitwidths for fast inference on FPGAs, as required in the trigger systems at the LHC experiments. The bitwidths are acquired during training by minimizing the total bit operations as an additional parameter. By allowing a bitwidth of zero, the model is pruned in-situ during training. Using this quantization-aware approach, our algorithm achieves state-of-the-art performance while also retaining permutation invariance which is a key property for particle physics applications. Due to the strength of transformers in representation learning, our work also serves as a stepping stone for the development of a larger foundation model for trigger applications.

Laatu, Lauri [Imperial Coll., London]↗
DOE OSTI2026

Geographic_Distribution_of_Populus_trichocarpa_Genotypes_by_DBSCAN_Cluster

Aninteractive mapshowingPopulus trichocarpaGWAS sub-population structure identified by DBSCAN clustering, which were derived from a UMAP projection of the top 8 PCs of LD-pruned pangenome SNP data. Geographic origins are searchable by genotype or river system using the search bar.

09 BIOMASS FUELS↗
DOE OSTI2025

Entropy-Assisted Quality Pattern Identification in Finance

Short-term patterns in financial time series form the cornerstone of many algorithmic trading strategies, yet extracting these patterns reliably from noisy market data remains a formidable challenge. In this paper, we propose an entropy-assisted framework for identifying high-quality, non-overlapping patterns that exhibit consistent behavior over time. We ground our approach in the premise that historical patterns, when accurately clustered and pruned, can yield substantial predictive power for short-term price movements. To achieve this, we incorporate an entropy-based measure as a proxy for information gain: patterns that lead to high one-sided movements in historical data yet retain low local entropy are more “informative” in signaling future market direction. Compared to conventional clustering techniques such as K-means and Gaussian Mixture Models (GMMs), which often yield biased or unbalanced groupings, our approach emphasizes balance over a forced visual boundary, ensuring that quality patterns are not lost due to over-segmentation. By emphasizing both predictive purity (low local entropy) and historical profitability, our method achieves a balanced representation of Buy and Sell patterns, making it better suited for short-term algorithmic trading strategies. This paper offers an in-depth illustration of our entropy-assisted framework through two case studies on Gold vs. USD and GBPUSD. While these examples demonstrate the method’s potential for extracting high-quality patterns, they do not constitute an exhaustive survey of all possible asset classes.

Physics↗
DOE OSTI2024

autoGEMM: Pushing the Limits of Irregular Matrix Multiplication on Arm Architectures

This paper presents an open-source library that pushes the limits of performance portability for irregular General Matrix Multiplication (GEMM) on the widely-used Arm architectures. Our library, autoGEMM, is designed to support a wide range of Arm processors: from edge devices to HPC-grade CPUs. autoGEMM generates optimized kernels for various hardware configurations by auto-combining fragments of autogenerated micro-kernels that employ hand-written optimizations to maximize computational efficiency. We optimize the kernel pipeline by tuning the register reuse and the data load/store overlapping. In addition, we use a dynamic tiling scheme to generate balanced tile shapes. Finally, we position autoGEMM on top of the TVM framework where our dynamic tiling scheme prunes the search space for TVM to identify the optimal combination of parameters for code optimization. Evaluations on five different classes of Arm chips demonstrate the advantages of autoGEMM. For small matrices, autoGEMM achieves 98% of peak and up to 2.0x speedup over state-of-the-art libraries such as LIBXSMM and LibShalom. For irregular matrices (i.e. tall skinny and long rectangles), autoGEMM is 1.3-2.0x faster than widely-used libraries such as OpenBLAS and Eigen. autoGEMM is publicly available at: https://github.com/wudu98/autoGEMM.

Wu, Du↗
DOE OSTI2025

MoE-Inference-Bench: Performance Evaluation of Mixture of Expert Large Language and Vision Models

Mixture of Experts (MoE) models have enabled the scaling of Large Language Models (LLMs) and Vision Language Models (VLMs) by achieving massive parameter counts while maintaining computational efficiency. However, MoEs introduce several inference-time challenges, including load imbalance across experts and the additional routing computational overhead. To address these challenges and fully harness the benefits of MoE, a systematic evaluation of hardware acceleration techniques is essential. We present MoE-Inference-Bench, a comprehensive study to evaluate MoE performance across diverse scenarios. We analyze the impact of batch size, sequence length, and critical MoE hyperparameters such as FFN dimensions and number of experts on throughput. We evaluate several optimization techniques on Nvidia H100 GPUs, including pruning, Fused MoE operations, speculative decoding, quantization, and various parallelization strategies. Our evaluation includes MoEs from the Mixtral, DeepSeek, OLMoE and Qwen families. The results reveal performance differences across configurations and provide insights for the efficient deployment of MoEs.

Chitty-Venkata, Krishna Teja↗
DOE OSTI2026

Surrogate Neural Architecture Codesign Package (SNAC-Pack)

Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and latency. We present the Surrogate Neural Architecture Codesign Package (SNAC-Pack), an open-source AutoML framework for hardware-aware neural architecture codesign and end-to-end FPGA deployment. SNAC-Pack runs a multi-objective global search with Optuna and NSGA-II, loading trials to a shared SQLite store that enables parallel workers across compute nodes. A hardware surrogate model outputs per-trial resource and latency estimates, avoiding the synthesis cost that would otherwise dominate the search loop. A local search stage then applies quantization-aware training (QAT) together with iterative magnitude pruning in a combined compression loop, after which the final model is synthesized to FPGA firmware via the hls4ml Python library. A YAML configuration and an optional agentic frontend let users run the pipeline on new datasets without modifying the framework. We demonstrate SNAC-Pack on jet classification at the Large Hadron Collider and superconducting qubit readout, discovering compact architectures that match or exceed strong baselines on the task metric while reducing FPGA resource utilization and, in the qubit readout case, reducing the design space exploration process from months of manual fine-tuning to hours of automated search.

Weitz, Jason [UC, San Diego]↗