Summary of Investigating Resource-efficient Neutron/gamma Classification Ml Models Targeting Efpgas, by Jyothisraj Johnson et al.
Investigating Resource-efficient Neutron/Gamma Classification ML Models Targeting eFPGAs
by Jyothisraj Johnson, Billy Boxer, Tarun Prakash, Carl Grace, Peter Sorensen, Mani Tripathi
First submitted to arxiv on: 19 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: High Energy Physics – Experiment (hep-ex); Nuclear Experiment (nucl-ex); Instrumentation and Detectors (physics.ins-det)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper explores the implementation of machine learning (ML) models in hardware using open-source embedded FPGA (eFPGA) frameworks. Building on previous work with Python package hls4ml, this study focuses on custom eFPGA fabrics for ML model deployment. The authors investigate resource-efficient implementations of fully-connected neural networks (fcNNs) and boosted decision trees (BDTs) for neutron/gamma classification using AmBe sealed source data. They examine input features, bit-resolution, sampling rate, hyperparameters, and trade-offs while tracking total resource usage. The performance metric is calculated neutron efficiency at a gamma leakage of 10^-3. This work aims to inform the specification of an eFPGA fabric for test chip integration. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper talks about using special computer chips (called FPGAs) to run machine learning models. It’s like a shortcut to make these models work faster and more efficiently. The authors tried different ways to do this with two types of models: one that works like a neural network in the brain, and another that’s like a decision tree. They used special data collected from a source called AmBe to train and test their models. They wanted to see how well these models worked while using as few resources (like memory or power) as possible. This research will help make better computer chips for testing new ideas. |
Keywords
» Artificial intelligence » Classification » Decision tree » Machine learning » Neural network » Tracking
