paper-with-me

홈 › Papers

Neural Architecture Codesign for Fast Bragg Peak Analysis

2023-12-10 · Luke McDermott, Jason Weitz, Dmitri Demler, Daniel Cummings, Nhan Tran, Javier Duarte

We develop an automated pipeline to streamline neural architecture codesign for fast, real-time Bragg peak analysis in high-energy diffraction microscopy. Traditional approaches, notably pseudo-Voigt fitting, demand significant computational resources, prompting interest in deep learning models for more efficient solutions. Our method employs neural architecture search and AutoML to enhance these models, including hardware costs, leading to the discovery of more hardware-efficient neural architectures. Our results match the performance, while achieving a 13$\times$ reduction in bit operations compared to the previous state-of-the-art. We show further speedup through model compression techniques such as quantization-aware-training and neural network pruning. Additionally, our hierarchical search space provides greater flexibility in optimization, which can easily extend to other tasks and domains.

📄 PDF Abstract BibTeX arXiv:2312.05978

Code (0)

등록된 구현이 없습니다.

Tasks

AutoMLModel CompressionNetwork PruningNeural Architecture SearchQuantization

Similar Papers 제목 키워드 기반

Neural Architecture Codesign for Fast Physics Applications

2025-01-09 · Jason Weitz, Dmitri Demler, Luke McDermott, Nhan Tran 외

We develop a pipeline to streamline neural architecture codesign for physics applications to reduce the need for ML expertise when designing models for novel tasks. Our method employs neural architecture search and netwo…

High-Level SynthesisModel CompressionNetwork PruningNeural Architecture Search+1

BraggNN: Fast X-ray Bragg Peak Analysis Using Deep Learning

2020-08-18 · Zhengchun Liu, Hemant Sharma, Jun-Sang Park, Peter Kenesei 외

X-ray diffraction based microscopy techniques such as High Energy Diffraction Microscopy rely on knowledge of the position of diffraction peaks with high precision. These positions are typically computed by fitting the o…

3D ReconstructionDeep LearningGPU

Network architecture search of X-ray based scientific applications

2024-04-16 · Adarsha Balaji, Ramyad Hadidi, Gregory Kollmer, Mohammed E. Fouda 외

X-ray and electron diffraction-based microscopy use bragg peak detection and ptychography to perform 3-D imaging at an atomic resolution. Typically, these techniques are implemented using computationally complex tasks su…

Neural Architecture Search

PeakNet: An Autonomous Bragg Peak Finder with Deep Neural Networks

2023-03-24 · Cong Wang, Po-Nan Li, Jana Thayer, Chun Hong Yoon

Serial crystallography at X-ray free electron laser (XFEL) and synchrotron facilities has experienced tremendous progress in recent times enabling novel scientific investigations into macromolecular structures and molecu…

GPUSemantic Segmentation

Discovering Features in Sr$_{14}$Cu$_{24}$O$_{41}$ Neutron Single Crystal Diffraction Data by Cluster Analysis

2018-09-13 · Yawei Hui, Yaohua Liu, Byung-Hoon Park

To address the SMC'18 data challenge, "Discovering Features in Sr$_{14}$Cu$_{24}$O$_{41}$", we have used the clustering algorithm "DBSCAN" to separate the diffuse scattering features from the Bragg peaks, which takes int…

ClusteringOpen-Ended Question Answering