Duc Hoang
YOU?
Author Swipe
View article: Advancing the CMS Level-1 Trigger: Jet Tagging with DeepSets at the HL-LHC
Advancing the CMS Level-1 Trigger: Jet Tagging with DeepSets at the HL-LHC Open
At the High Luminosity LHC, selecting important physics processes such as (di-) Higgs production will be a high priority. The Phase-2 Upgrade of the CMS Level-1 Trigger will reconstruct particle candidates and use pileup mitigation for the…
View article: Stand-alone Hybrid Solar-Wind Microgrid for Providing Clean and Reliable Energy to Disadvantaged Communities
Stand-alone Hybrid Solar-Wind Microgrid for Providing Clean and Reliable Energy to Disadvantaged Communities Open
The need for a tailored energy solution for disadvantaged communities, such as mobile homes and home clusters, motivates the development of a stand-alone hybrid solar-wind microgrid. Such communities often have access to the grid, but the …
View article: Building Machine Learning Challenges for Anomaly Detection in Science
Building Machine Learning Challenges for Anomaly Detection in Science Open
Scientific discoveries are often made by finding a pattern or object that was not predicted by the known rules of science. Oftentimes, these anomalous events or objects that do not conform to the norms are an indication that the rules of s…
View article: Do Compressed LLMs Forget Knowledge? An Experimental Study with Practical Implications
Do Compressed LLMs Forget Knowledge? An Experimental Study with Practical Implications Open
Compressing Large Language Models (LLMs) often leads to reduced performance, especially for knowledge-intensive tasks. In this work, we dive into how compression damages LLMs' inherent knowledge and the possible remedies. We start by propo…
View article: Current Status and Future Prospects for the Light Dark Matter eXperiment
Current Status and Future Prospects for the Light Dark Matter eXperiment Open
The constituents of dark matter are still unknown, and the viable possibilities span a vast range of masses. The physics community has established searching for sub-GeV dark matter as a high priority and identified accelerator-based experi…
View article: Compressing deep neural networks on FPGAs to binary and ternary precision with HLS4ML
Compressing deep neural networks on FPGAs to binary and ternary precision with HLS4ML Open
We present the implementation of binary and ternary neural networks in the hls4ml library, designed to automatically convert deep neural network models to digital circuits with FPGA firmware. Starting from benchmark models trained with flo…
View article: DarkQuest: A dark sector upgrade to SpinQuest at the 120 GeV Fermilab Main Injector
DarkQuest: A dark sector upgrade to SpinQuest at the 120 GeV Fermilab Main Injector Open
Expanding the mass range and techniques by which we search for dark matter is an important part of the worldwide particle physics program. Accelerator-based searches for dark matter and dark sector particles are a uniquely compelling part …
View article: Current Status and Future Prospects for the Light Dark Matter eXperiment
Current Status and Future Prospects for the Light Dark Matter eXperiment Open
The constituents of dark matter are still unknown, and the viable possibilities span a vast range of masses. The physics community has established searching for sub-GeV dark matter as a high priority and identified accelerator-based experi…
View article: Applications and Techniques for Fast Machine Learning in Science
Applications and Techniques for Fast Machine Learning in Science Open
In this community review report, we discuss applications and techniques for fast machine learning (ML) in science -- the concept of integrating power ML methods into the real-time experimental data processing loop to accelerate scientific …
View article: hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devices
hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devices Open
Accessible machine learning algorithms, software, and diagnostic tools for energy-efficient devices and systems are extremely valuable across a broad range of application domains. In scientific domains, real-time near-sensor processing can…
View article: Intelliquench: An Adaptive Machine Learning System for Detection of Superconducting Magnet Quenches
Intelliquench: An Adaptive Machine Learning System for Detection of Superconducting Magnet Quenches Open
In superconducting magnets, the irreversible transition of a portion of the conductor to resistive state is called a “quench.” Having large stored energy, magnets can be damaged by quenches due to localized heating, high voltage, or large …
View article: hls-fpga-machine-learning/hls4ml: aster
hls-fpga-machine-learning/hls4ml: aster Open
What's new: Support for GarNet layer (see paper) Input layer precision added to config generator utility New 'SkipOptimizers' config option. Now you can run all Optimizers by default (as in v0.3.0) but subtract any specified by 'SkipOptimi…
View article: MM-Hand: 3D-Aware Multi-Modal Guided Hand Generative Network for 3D Hand Pose Synthesis
MM-Hand: 3D-Aware Multi-Modal Guided Hand Generative Network for 3D Hand Pose Synthesis Open
Estimating the 3D hand pose from a monocular RGB image is important but challenging. A solution is training on large-scale RGB hand images with accurate 3D hand keypoint annotations. However, it is too expensive in practice. Instead, we ha…
View article: Fast inference of Boosted Decision Trees in FPGAs for particle physics
Fast inference of Boosted Decision Trees in FPGAs for particle physics Open
We describe the implementation of Boosted Decision Trees in the hls4ml\nlibrary, which allows the translation of a trained model into FPGA firmware\nthrough an automated conversion process. Thanks to its fully on-chip\nimplementation, hls4…
View article: Inferring Convolutional Neural Networks' Accuracies from Their Architectural Characterizations
Inferring Convolutional Neural Networks' Accuracies from Their Architectural Characterizations Open
Convolutional Neural Networks (CNNs) have shown strong promise for analyzing scientific data from many domains including particle imaging detectors. However, the challenge of choosing the appropriate network architecture (depth, kernel sha…
View article: A Comparison of Fast Kurtogram and SFLA based-ED Techniques for Bearing Fault Detection
A Comparison of Fast Kurtogram and SFLA based-ED Techniques for Bearing Fault Detection Open
The paper proposes a method using Shuffled Frog Leaping Algorithm (SFLA) to identify the optimal frequencies (center frequency and bandwith) of the bandpass filter.Addtion, fast kurtogram is also used to find the optimal bandpass filter.Si…
View article: Development of an ENVISAT Altimetry Processor Providing Sea Level Continuity Between Open Ocean and Arctic Leads
Development of an ENVISAT Altimetry Processor Providing Sea Level Continuity Between Open Ocean and Arctic Leads Open
Over the Arctic regions, current conventional altimetry products suffer from a lack of
\ncoverage or from degraded performance due to the inadequacy of the standard process-
\ning applied in the ground segments. This paper presents a set o…