Comparison between Adam, AdaMax and Adam W optimizers to implement a Weather Forecast based on Neural Networks for the Andean city of Quito

The main function of an optimizer is to determine in what measure to change the weights and the learning rate of the neural network to reduce losses. One of the best known optimizers is Adam, which main advantage is the invariance of the magnitudes of the parameter updates with respect to the change...

Full description

Saved in:
Bibliographic Details
Published in:2021 IEEE Fifth Ecuador Technical Chapters Meeting (ETCM) pp. 1 - 6
Main Authors: Llugsi, Ricardo, Yacoubi, Samira El, Fontaine, Allyx, Lupera, Pablo
Format: Conference Proceeding
Language:English
Published: IEEE 12.10.2021
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract The main function of an optimizer is to determine in what measure to change the weights and the learning rate of the neural network to reduce losses. One of the best known optimizers is Adam, which main advantage is the invariance of the magnitudes of the parameter updates with respect to the change of scale of the gradient. However, other optimizers are often chosen because they generalize in a better manner. AdamW is a variant of Adam where the weight decay is performed only after controlling the parameter-wise step size. In order to present a comparative scenario for optimizers in the present work, a Temperature Forecast for the Andean city of Quito using a neural network structure with uncertainty reduction was implemented and three optimizers (Adam, AdaMax and AdamW) were analyzed. In order to do the comparison three error metrics were obtained per hour in order to determine the effectiveness of the prediction. From the analysis it can be seen that Adam and AdaMax behave similarly reaching a maximum MSE per hour of 2.5°C nevertheless AdamW allows to reduce this error around 1.3°C.
AbstractList The main function of an optimizer is to determine in what measure to change the weights and the learning rate of the neural network to reduce losses. One of the best known optimizers is Adam, which main advantage is the invariance of the magnitudes of the parameter updates with respect to the change of scale of the gradient. However, other optimizers are often chosen because they generalize in a better manner. AdamW is a variant of Adam where the weight decay is performed only after controlling the parameter-wise step size. In order to present a comparative scenario for optimizers in the present work, a Temperature Forecast for the Andean city of Quito using a neural network structure with uncertainty reduction was implemented and three optimizers (Adam, AdaMax and AdamW) were analyzed. In order to do the comparison three error metrics were obtained per hour in order to determine the effectiveness of the prediction. From the analysis it can be seen that Adam and AdaMax behave similarly reaching a maximum MSE per hour of 2.5°C nevertheless AdamW allows to reduce this error around 1.3°C.
Author Fontaine, Allyx
Lupera, Pablo
Yacoubi, Samira El
Llugsi, Ricardo
Author_xml – sequence: 1
  givenname: Ricardo
  surname: Llugsi
  fullname: Llugsi, Ricardo
  email: ricardo.llugsi@epn.edu.ec
  organization: Telecommunications and Information Networks (DETRI),National Polytechnic School Department of Electronics,Quito,Ecuador
– sequence: 2
  givenname: Samira El
  surname: Yacoubi
  fullname: Yacoubi, Samira El
  email: yacoubi@univ-perp.fr
  organization: Université de Perpignan Via Domitia,UMR Espace-Dev,Perpignan,France
– sequence: 3
  givenname: Allyx
  surname: Fontaine
  fullname: Fontaine, Allyx
  email: allyx.fontaine@univ-guyane.fr
  organization: Université de Guyane,UMR Espace-Dev,Cayenne,French Guiana
– sequence: 4
  givenname: Pablo
  surname: Lupera
  fullname: Lupera, Pablo
  email: pablo.lupera@epn.edu.ec
  organization: National Polytechnic School,Faculty of Electrical and Electronics Engineering,Quito,Ecuador
BookMark eNotkNFKAzEQRSPog9Z-gSDzAbZms8nu5rEUq4JVhEIfy2R3QoPdZMmm1PoL_rSrloG5l-Fy4M4VO_fBE2O3GZ9mGdf3D6v5UuWFzKeCi2yqleZFlZ2xsS6rrCiUlNkwl-x7HtoOo-uDB0PpQORh1mB797uX-Anom78DrCF0ybXui2IPKYBrux215BMgrAnTliIsQqQa-wQGe2pgYL7SPuJukHQI8aMHGyIMUZj5htBD7dIRgoX3vUvhml1Y3PU0PumIrRZDj6fJy9vj83z2MtlWgk9saVBpLauSUCoqBc-lriquy8Ji3ehClFQPXgppjVFWoG6UMLUgjaZUlI_YzT_WEdGmi67FeNycPpT_AFlIYdc
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ETCM53643.2021.9590681
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781665441414
1665441410
EndPage 6
ExternalDocumentID 9590681
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-h820-f7ba599487ea45e720349880976facd9627ec76f424fbb5f2a9d52bc2e9ab75e3
IEDL.DBID RIE
IngestDate Thu Jun 29 18:38:10 EDT 2023
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-h820-f7ba599487ea45e720349880976facd9627ec76f424fbb5f2a9d52bc2e9ab75e3
PageCount 6
ParticipantIDs ieee_primary_9590681
PublicationCentury 2000
PublicationDate 2021-Oct.-12
PublicationDateYYYYMMDD 2021-10-12
PublicationDate_xml – month: 10
  year: 2021
  text: 2021-Oct.-12
  day: 12
PublicationDecade 2020
PublicationTitle 2021 IEEE Fifth Ecuador Technical Chapters Meeting (ETCM)
PublicationTitleAbbrev ETCM
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.7729363
Snippet The main function of an optimizer is to determine in what measure to change the weights and the learning rate of the neural network to reduce losses. One of...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Adam
Adam W
AdaMax
Bayesian
Convolutional
Dropout
Encoder-Decoder Neural Network
Neural networks
Optimizer
Temperature distribution
Time Series
Time series analysis
Uncertainty
Urban areas
Walk-Forward validation
Weather forecasting
Weight measurement
Title Comparison between Adam, AdaMax and Adam W optimizers to implement a Weather Forecast based on Neural Networks for the Andean city of Quito
URI https://ieeexplore.ieee.org/document/9590681
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LawIxEA5WeuipLVr6Zg49urqPrDFHkUoPVSxI9SZ5zFKh7hZdS-lf6J_uJC6WQi-97IYQsjBhZ74k38zH2J3VGTdG8CDUiQo45yKQCqOAHGaopY4S4eV8nh_FeNybz-Wkxlr7XBhE9OQzbLumv8u3hdm6o7KOTGXYdXnWB0J0d7laVdJvFMrO_XQwShOKsLTri6N2NfiXaooPGsPj_33uhDV_su9gso8rp6yGeYN9DfaCgVBxq6Bv1arlniP1ASq3vgNmUJAbWC0_CdhBWcByVVHEQcFsB_jACXIatSnBBTELNKcr0qFe6eVZ4RsgLAs0FPq5RZWDIbAORQZPW_IATTYdkhkegkpIIXihAB9kQqtUStqaoOIpuotXLum_JSSSKWOd_A4aavOYZ1qnWaykTWNtYpRKixSTM1bPixzPGSQSLSahyXhXcUxEzzqBq0wbV0dOYHzBGs6Oi7ddqYxFZcLLv7uv2JFbqsATRK5ZvVxv8YYdmvdyuVnf-vX9BgA1qaM
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Na8JAEF3EFtpTW7T0u3Po0WhMNq57FKlYqmJBqjfZjwkValI0ltK_0D_d2RgshV56SZZl2cAsmXm7-2YeY3dWx9wYwT1fh8rjnAtPKmx65DB9LXUzFLmcz_NAjEbt2UyOS6y2y4VBxJx8hnXXzO_ybWo27qisISPpt1ye9V7EeeBvs7WKtN-mLxv3k-4wCinG0r4vaNaL4b90U_Kw0Tv63wePWfUn_w7Gu8hywkqYVNhXdycZCAW7CjpWLWvuOVQfoBKbd8AUUnIEy8UnQTvIUlgsC5I4KJhuIR84SU6j1hm4MGaB5nRlOtQrvXJe-BoIzQINhU5iUSVgCK5DGsPThnxAlU16ZIa-V0gpeC8U4r1YaBVJSZsTVDxCd_XKJf25hEViZawT4EFDbR7wWOsoDpS0UaBNgFJpEWF4yspJmuAZg1CixdA3MW8pjqFoWydxFWvjKskJDM5Zxdlx_rYtljEvTHjxd_ctO-hPhoP54GH0eMkO3bJ5OV3kipWz1Qav2b55zxbr1U2-1t_8qazq
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2021+IEEE+Fifth+Ecuador+Technical+Chapters+Meeting+%28ETCM%29&rft.atitle=Comparison+between+Adam%2C+AdaMax+and+Adam+W+optimizers+to+implement+a+Weather+Forecast+based+on+Neural+Networks+for+the+Andean+city+of+Quito&rft.au=Llugsi%2C+Ricardo&rft.au=Yacoubi%2C+Samira+El&rft.au=Fontaine%2C+Allyx&rft.au=Lupera%2C+Pablo&rft.date=2021-10-12&rft.pub=IEEE&rft.spage=1&rft.epage=6&rft_id=info:doi/10.1109%2FETCM53643.2021.9590681&rft.externalDocID=9590681