Pretraga

Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 1217056

Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification

(CMS HGCAL Collaboration) Burazin Mišura, Arijana; Musić, Josip; Ožegović, Julije; Lelas, Damir
Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification // 2022 International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2022)
Split, 2022. 1570819047, 6 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)

CROSBI ID: 1217056 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification

Autori
Burazin Mišura, Arijana ; Musić, Josip ; Ožegović, Julije ; Lelas, Damir

Kolaboracija
CMS HGCAL Collaboration

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
2022 International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2022) / - Split, 2022

Skup
30th International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2022)

Mjesto i datum
Split, Hrvatska, 22.09.2022. - 24.09.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
CMS ; trigger ; neural network ; quantization

Sažetak
Since the beginning of the Large Hadron Collider (LHC) project, one of the biggest problems faced by scientists is dealing with the enormous amount of data produced by detectors. For the High Luminosity LHC phase, a new calorimeter endcap named High Granularity Calorimeter (HGCAL) has been developed for the upgrade of the Compact Muon Solenoid (CMS) detector. High granularity together with increased pile-up will result in a huge increase in data rate. Therefore, efficient real time analysis methods are required to select data coming from events of interest from tremendous background production. The development of specialized libraries, like QKeras, enables the quantization of neural networks (NNs) so far used mostly in the offline analysis due to their high processing requirements. The reduction of NN size together with input quantization makes possible their usage in limited resources as a particle classification strategy. We present a comparison of fully connected and convolutional NNs used for the potential real-time signal/background classification method. Results show that convolutional models slightly outperform fully connected architectures in both generic and quantized cases.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Temeljne tehničke znanosti



POVEZANOST RADA

Ustanove:
Fakultet elektrotehnike, strojarstva i brodogradnje, Split,
Sveučilište u Splitu Sveučilišni odjel za stručne studije

Profili:

Avatar Url Damir Lelas (autor)

Avatar Url Josip Musić (autor)

Avatar Url Julije Ožegović (autor)

Avatar Url Arijana Burazin Misura (autor)

Citiraj ovu publikaciju:

(CMS HGCAL Collaboration) Burazin Mišura, Arijana; Musić, Josip; Ožegović, Julije; Lelas, Damir
Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification // 2022 International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2022)
Split, 2022. 1570819047, 6 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
(CMS HGCAL Collaboration) (CMS HGCAL Collaboration) Burazin Mišura, A., Musić, J., Ožegović, J. & Lelas, D. (2022) Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification. U: 2022 International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2022).
@article{article, author = {Burazin Mi\v{s}ura, Arijana and Musi\'{c}, Josip and O\v{z}egovi\'{c}, Julije and Lelas, Damir}, year = {2022}, pages = {6}, chapter = {1570819047}, keywords = {CMS, trigger, neural network, quantization}, title = {Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification}, keyword = {CMS, trigger, neural network, quantization}, publisherplace = {Split, Hrvatska}, chapternumber = {1570819047} }
@article{article, author = {Burazin Mi\v{s}ura, Arijana and Musi\'{c}, Josip and O\v{z}egovi\'{c}, Julije and Lelas, Damir}, year = {2022}, pages = {6}, chapter = {1570819047}, keywords = {CMS, trigger, neural network, quantization}, title = {Performance Comparison of Generic and Quantized Fully Connected and Convolutional Neural Networks for Real-Time Signal/Background Classification}, keyword = {CMS, trigger, neural network, quantization}, publisherplace = {Split, Hrvatska}, chapternumber = {1570819047} }

Podijeli:





Contrast
Increase Font
Decrease Font
Dyslexic Font
CROSBI koristi kolačiće (cookies) kako bi poboljšao funkcionalnost stranice.