Toward an Adaptive Artificial Neural Network–Based Postprocessor
Abstract We introduce an adaptive form of postprocessor where algorithm structures are neural networks where the number of hidden nodes and the network training features evolve. Key potential advantages of this system are the flexible, nonlinear mapping capabilities of neural networks and, through b...
Saved in:
Published in | Monthly weather review Vol. 149; no. 12; pp. 4045 - 4055 |
---|---|
Main Author | |
Format | Journal Article |
Language | English |
Published |
Washington
American Meteorological Society
01.12.2021
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Abstract
We introduce an adaptive form of postprocessor where algorithm structures are neural networks where the number of hidden nodes and the network training features evolve. Key potential advantages of this system are the flexible, nonlinear mapping capabilities of neural networks and, through backpropagation, the ability to rapidly establish capable predictors in an algorithm population. The system can be implemented after one initial training process and future changes to postprocessor inputs (new observations, new inputs, or model upgrades) are incorporated as they become available. As in prior work, the implementation in the form of a predator–prey ecosystem allows for the ready construction of ensembles. Computational requirements are minimal, and the use of a moving data window means that data storage requirements are constrained. The system adds predictive skill to a demonstration dynamical model representing the hemispheric circulation, with skill competitive with or exceeding that obtainable from multiple linear regression and standard artificial neural networks constructed under typical operational limitations. The system incorporates new information rapidly and the dependence of the approach on the training data size is similar to multiple linear regression. A loss of performance occurs relative to a fixed neural network architecture in which only the weights are adjusted after training, but this loss is compensated for by gains from the ensemble predictions. While the demonstration dynamical model is complex, current numerical weather prediction models are considerably more so, and thus a future step will be to apply this technique to operational weather forecast data. |
---|---|
AbstractList | Abstract
We introduce an adaptive form of postprocessor where algorithm structures are neural networks where the number of hidden nodes and the network training features evolve. Key potential advantages of this system are the flexible, nonlinear mapping capabilities of neural networks and, through backpropagation, the ability to rapidly establish capable predictors in an algorithm population. The system can be implemented after one initial training process and future changes to postprocessor inputs (new observations, new inputs, or model upgrades) are incorporated as they become available. As in prior work, the implementation in the form of a predator–prey ecosystem allows for the ready construction of ensembles. Computational requirements are minimal, and the use of a moving data window means that data storage requirements are constrained. The system adds predictive skill to a demonstration dynamical model representing the hemispheric circulation, with skill competitive with or exceeding that obtainable from multiple linear regression and standard artificial neural networks constructed under typical operational limitations. The system incorporates new information rapidly and the dependence of the approach on the training data size is similar to multiple linear regression. A loss of performance occurs relative to a fixed neural network architecture in which only the weights are adjusted after training, but this loss is compensated for by gains from the ensemble predictions. While the demonstration dynamical model is complex, current numerical weather prediction models are considerably more so, and thus a future step will be to apply this technique to operational weather forecast data. We introduce an adaptive form of postprocessor where algorithm structures are neural networks where the number of hidden nodes and the network training features evolve. Key potential advantages of this system are the flexible, nonlinear mapping capabilities of neural networks and, through backpropagation, the ability to rapidly establish capable predictors in an algorithm population. The system can be implemented after one initial training process and future changes to postprocessor inputs (new observations, new inputs, or model upgrades) are incorporated as they become available. As in prior work, the implementation in the form of a predator–prey ecosystem allows for the ready construction of ensembles. Computational requirements are minimal, and the use of a moving data window means that data storage requirements are constrained. The system adds predictive skill to a demonstration dynamical model representing the hemispheric circulation, with skill competitive with or exceeding that obtainable from multiple linear regression and standard artificial neural networks constructed under typical operational limitations. The system incorporates new information rapidly and the dependence of the approach on the training data size is similar to multiple linear regression. A loss of performance occurs relative to a fixed neural network architecture in which only the weights are adjusted after training, but this loss is compensated for by gains from the ensemble predictions. While the demonstration dynamical model is complex, current numerical weather prediction models are considerably more so, and thus a future step will be to apply this technique to operational weather forecast data. |
Author | Roebber, Paul J |
Author_xml | – sequence: 1 givenname: Paul surname: Roebber middlename: J fullname: Roebber, Paul J |
BookMark | eNotkMtKAzEYRoNUsK2u3Q64TvvnPrPsxapQL0jFZYiZBKbWSU1mLO58B9_QJ3FqXZ3Nx3fgDFCvDrVD6JzAiBAlxrfPj3iOKcEAeTEiR6hPBAUMvGA91AegCoPk_AQNUloDgJSc9tF0FXYmlpmps0lptk314bJJbCpf2cpssjvXxj80uxBff76-pya5MnsIqdnGYF1KIZ6iY282yZ39c4ieFper2TVe3l_dzCZLbCllDRYFFaAsl1KqnHlrS2ULMPJFEKXKgjrCQeTMUe_LnBlTdLDCUWuEYVx4NkQXh9_O_N661Oh1aGPdKTWVShLOcpJ3q_FhZWNIKTqvt7F6M_FTE9D7ULoLpeeaEr0PpQn7Bd6JXeE |
Cites_doi | 10.1175/1520-0434(1995)010<0689:DCOSTS>2.0.CO;2 10.1175/MWR-D-18-0187.1 10.1142/S0218127403007904 10.1175/2010JTECHA1449.1 10.1175/MWR-D-17-0084.1 10.1175/JAMC-D-15-0100.1 10.1175/MWR-D-19-0346.1 10.1175/MWR-D-18-0420.1 10.1038/353241a0 10.1038/333545a0 10.1175/WAF-D-11-00022.1 10.1111/j.1600-0870.1984.tb00230.x 10.1038/323533a0 10.1175/MWR-D-14-00096.1 10.3402/tellusa.v42i3.11884 10.1175/MWR-D-19-0344.1 10.1155/2014/296279 10.1175/1520-0434(2003)018<0288:TCUMOS>2.0.CO;2 10.1175/MWR-D-19-0063.1 10.1175/MWR-D-15-0096.1 10.1175/1520-0450(1972)011<1203:TUOMOS>2.0.CO;2 10.1175/1520-0434(2002)017<0206:TCUMOS>2.0.CO;2 10.1007/3-540-49430-8_2 10.1175/1520-0493(2003)131<2510:SEPOMT>2.0.CO;2 10.1175/MWR-D-14-00095.1 |
ContentType | Journal Article |
Copyright | Copyright American Meteorological Society Dec 2021 |
Copyright_xml | – notice: Copyright American Meteorological Society Dec 2021 |
DBID | AAYXX CITATION 7QH 7TG 7TN 7UA 8FD C1K F1W H8D H96 KL. L.G L7M |
DOI | 10.1175/MWR-D-21-0089.1 |
DatabaseName | CrossRef Aqualine Meteorological & Geoastrophysical Abstracts Oceanic Abstracts Water Resources Abstracts Technology Research Database Environmental Sciences and Pollution Management ASFA: Aquatic Sciences and Fisheries Abstracts Aerospace Database Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Meteorological & Geoastrophysical Abstracts - Academic Aquatic Science & Fisheries Abstracts (ASFA) Professional Advanced Technologies Database with Aerospace |
DatabaseTitle | CrossRef Aerospace Database Aquatic Science & Fisheries Abstracts (ASFA) Professional Meteorological & Geoastrophysical Abstracts Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Oceanic Abstracts Technology Research Database ASFA: Aquatic Sciences and Fisheries Abstracts Aqualine Advanced Technologies Database with Aerospace Meteorological & Geoastrophysical Abstracts - Academic Water Resources Abstracts Environmental Sciences and Pollution Management |
DatabaseTitleList | CrossRef Aerospace Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Meteorology & Climatology |
EISSN | 1520-0493 |
EndPage | 4055 |
ExternalDocumentID | 10_1175_MWR_D_21_0089_1 |
GroupedDBID | -~X .4S .DC 123 186 29M 2WC 3V. 4.4 6KP 6TJ 7XC 88I 8AF 8FE 8FG 8FH 8G5 8R4 8R5 AAMDK AAYXX ABCQX ABDBF ABPPZ ABTAH ABUWG ACGFO ACGOD ACIHN ADOJD AEAQA AELAQ AENEX AFFNX AFKRA AFRAH AGCDD AIKKD ALMA_UNASSIGNED_HOLDINGS ALQLQ ARAPS ARCSS ATCPS AZQEC BCR BCU BEC BENPR BES BGLVJ BHPHI BKSAR BLC BPHCQ BQZYQ C1A CAG CCPQU CITATION COF CS3 D1K DU5 DWQXO E3Z EAD EAP EBS EDH EDO EJD EMK EPL EST ESX F8P FRP GNUQQ GUQSH H13 HCIFZ H~9 I-F K6- KQ8 L7B LK5 M1Q M2O M2P M2Q M7R MV1 MVM OHT OK1 P2P P62 PATMY PCBAR PQQKQ PROAC PYCSY Q2X QF4 QM1 QN7 QO4 ROL RWA RWE RWL RXW S0X SJFOW TAE TN5 TR2 TUS U5U UHB UNMZH WH7 WHG ZY4 ~02 7QH 7TG 7TN 7UA 8FD C1K F1W H8D H96 KL. L.G L7M |
ID | FETCH-LOGICAL-c223t-592507c4666783fccd7c90a6b5177d92e140583e2ffd83aa9fd8c5e2ca5a345f3 |
ISSN | 0027-0644 |
IngestDate | Fri Sep 13 08:32:53 EDT 2024 Fri Aug 23 02:30:44 EDT 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 12 |
Language | English |
LinkModel | OpenURL |
MergedId | FETCHMERGED-LOGICAL-c223t-592507c4666783fccd7c90a6b5177d92e140583e2ffd83aa9fd8c5e2ca5a345f3 |
PQID | 2676143818 |
PQPubID | 41865 |
PageCount | 11 |
ParticipantIDs | proquest_journals_2676143818 crossref_primary_10_1175_MWR_D_21_0089_1 |
PublicationCentury | 2000 |
PublicationDate | 2021-12-00 20211201 |
PublicationDateYYYYMMDD | 2021-12-01 |
PublicationDate_xml | – month: 12 year: 2021 text: 2021-12-00 |
PublicationDecade | 2020 |
PublicationPlace | Washington |
PublicationPlace_xml | – name: Washington |
PublicationTitle | Monthly weather review |
PublicationYear | 2021 |
Publisher | American Meteorological Society |
Publisher_xml | – name: American Meteorological Society |
References | Schaffer (ref191) 2020; 148 Roebber (ref171) 2019; 147 Tsonis (ref481) 1988; 333 Eckel (ref281) 2016; 144 Lorenz (ref101) 1991; 353 Lorenz (ref361) 1990; 42A Lorenz (ref371) 1991; 353 Bremnes (ref271) 2019; 147 Volterra (ref231) 1931 Hill (ref311) 2020; 148 Lorenz (ref351) 1984; 36A Kim (ref61) 2016; 55 Kim (ref331) 2016; 55 Rasp (ref411) 2018; 146 Wilson (ref531) 2003; 18 Felker (ref21) 2011; 28 Nipen (ref121) 2011; 26 Glahn (ref31) 1972; 11 Lotka (ref381) 1925 Eckel (ref11) 2016; 144 Wang (ref511) 2014; 2014 Wilson (ref261) 2003; 18 Tsonis (ref211) 1988; 333 Pelosi (ref131) 2017; 145 Schaffer (ref461) 2020; 148 Wilson (ref521) 2002; 17 Stensrud (ref471) 2003; 131 Homleid (ref321) 1995; 10 Roebber (ref151) 2015a; 143 LeCun (ref71) 1998 Nipen (ref391) 2011; 26 Wilson (ref251) 2002; 17 Roebber (ref421) 2015a; 143 LeCun (ref341) 1998 Pelosi (ref401) 2017; 145 Roebber (ref441) 2019; 147 Wang (ref241) 2014; 2014 Roebber (ref431) 2015b; 143 Stensrud (ref201) 2003; 131 Glahn (ref301) 1972; 11 Lorenz (ref91) 1990; 42A Bremnes (ref01) 2019; 147 Felker (ref291) 2011; 28 Rumelhart (ref451) 1986; 323 Lotka (ref111) 1925 van Veen (ref491) 2003; 13 van Veen (ref221) 2003; 13 Hill (ref41) 2020; 148 Rasp (ref141) 2018; 146 Volterra (ref501) 1931 Roebber (ref161) 2015b; 143 Rumelhart (ref181) 1986; 323 Homleid (ref51) 1995; 10 Lorenz (ref81) 1984; 36A |
References_xml | – volume: 10 start-page: 689 year: 1995 ident: ref321 article-title: Diurnal corrections of short-term temperature forecasts using the Kalman filter publication-title: Wea. Forecasting doi: 10.1175/1520-0434(1995)010<0689:DCOSTS>2.0.CO;2 contributor: fullname: Homleid – volume: 146 start-page: 3885 year: 2018 ident: ref141 article-title: Neural networks for postprocessing ensemble weather forecasts publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-18-0187.1 contributor: fullname: Rasp – volume: 13 start-page: 2117 year: 2003 ident: ref221 article-title: Baroclinic flow and the Lorenz-84 model publication-title: Int. J. Bifurcation Chaos doi: 10.1142/S0218127403007904 contributor: fullname: van Veen – volume: 28 start-page: 709 year: 2011 ident: ref291 article-title: Forecasting post-extratropical transition outcomes for tropical cyclones using support vector machine classifiers publication-title: J. Atmos. Oceanic Technol. doi: 10.1175/2010JTECHA1449.1 contributor: fullname: Felker – volume: 145 start-page: 4837 year: 2017 ident: ref401 article-title: Adaptive Kalman filtering for postprocessing ensemble numerical weather predictions publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-17-0084.1 contributor: fullname: Pelosi – volume: 146 start-page: 3885 year: 2018 ident: ref411 article-title: Neural networks for postprocessing ensemble weather forecasts publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-18-0187.1 contributor: fullname: Rasp – volume: 55 start-page: 773 year: 2016 ident: ref331 article-title: Nearest neighbor–genetic algorithm for downscaling of climate change data from GCMs publication-title: J. Appl. Meteor. Climatol. doi: 10.1175/JAMC-D-15-0100.1 contributor: fullname: Kim – volume: 148 start-page: 1951 year: 2020 ident: ref191 article-title: Development and evaluation of an evolutionary programming-based tropical cyclone intensity model publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-19-0346.1 contributor: fullname: Schaffer – volume: 148 start-page: 1951 year: 2020 ident: ref461 article-title: Development and evaluation of an evolutionary programming-based tropical cyclone intensity model publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-19-0346.1 contributor: fullname: Schaffer – volume: 147 start-page: 1769 year: 2019 ident: ref01 article-title: Constrained quantile regression splines for ensemble postprocessing publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-18-0420.1 contributor: fullname: Bremnes – volume: 353 start-page: 241 year: 1991 ident: ref101 article-title: Dimension of weather and climate attractors publication-title: Nature doi: 10.1038/353241a0 contributor: fullname: Lorenz – volume: 333 start-page: 545 year: 1988 ident: ref211 article-title: The weather attractor over very short timescales publication-title: Nature doi: 10.1038/333545a0 contributor: fullname: Tsonis – start-page: 495 volume-title: Elements of Physical Biology year: 1925 ident: ref381 contributor: fullname: Lotka – volume: 26 start-page: 564 year: 2011 ident: ref391 article-title: Updating short-term probabilistic weather forecasts of continuous variables using recent observations publication-title: Wea. Forecasting doi: 10.1175/WAF-D-11-00022.1 contributor: fullname: Nipen – volume: 145 start-page: 4837 year: 2017 ident: ref131 article-title: Adaptive Kalman filtering for postprocessing ensemble numerical weather predictions publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-17-0084.1 contributor: fullname: Pelosi – volume: 36A start-page: 98 year: 1984 ident: ref351 article-title: Irregularity: A fundamental property of the atmosphere publication-title: Tellus doi: 10.1111/j.1600-0870.1984.tb00230.x contributor: fullname: Lorenz – volume: 323 start-page: 533 year: 1986 ident: ref451 article-title: Learning representations by back-propagating errors publication-title: Nature doi: 10.1038/323533a0 contributor: fullname: Rumelhart – volume: 333 start-page: 545 year: 1988 ident: ref481 article-title: The weather attractor over very short timescales publication-title: Nature doi: 10.1038/333545a0 contributor: fullname: Tsonis – volume: 143 start-page: 1506 year: 2015b ident: ref431 article-title: Using evolutionary programs to maximize minimum temperature forecast skill publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-14-00096.1 contributor: fullname: Roebber – volume: 42A start-page: 378 year: 1990 ident: ref91 article-title: Can chaos and intransitivity lead to interannual variability? publication-title: Tellus doi: 10.3402/tellusa.v42i3.11884 contributor: fullname: Lorenz – volume: 28 start-page: 709 year: 2011 ident: ref21 article-title: Forecasting post-extratropical transition outcomes for tropical cyclones using support vector machine classifiers publication-title: J. Atmos. Oceanic Technol. doi: 10.1175/2010JTECHA1449.1 contributor: fullname: Felker – start-page: 9 year: 1931 ident: ref501 article-title: Variations and fluctuations of the number of individuals in animal species living together publication-title: Animal Ecology contributor: fullname: Volterra – volume: 148 start-page: 2135 year: 2020 ident: ref311 article-title: Forecasting severe weather with random forests publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-19-0344.1 contributor: fullname: Hill – volume: 2014 year: 2014 ident: ref511 article-title: Dynamical analysis of the Lorenz-84 atmospheric circulation model publication-title: J. Appl. Math. doi: 10.1155/2014/296279 contributor: fullname: Wang – volume: 18 start-page: 288 year: 2003 ident: ref261 article-title: The Canadian Updateable Model Output Statistics (UMOS) system: Validation against perfect prog publication-title: Wea. Forecasting doi: 10.1175/1520-0434(2003)018<0288:TCUMOS>2.0.CO;2 contributor: fullname: Wilson – volume: 13 start-page: 2117 year: 2003 ident: ref491 article-title: Baroclinic flow and the Lorenz-84 model publication-title: Int. J. Bifurcation Chaos doi: 10.1142/S0218127403007904 contributor: fullname: van Veen – volume: 147 start-page: 4241 year: 2019 ident: ref171 article-title: Using a coevolutionary post-processor to improve skill for both forecasts of surface temperature and nowcasts of convection occurrence publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-19-0063.1 contributor: fullname: Roebber – volume: 144 start-page: 897 year: 2016 ident: ref281 article-title: A hybrid NWP—Analog ensemble publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-15-0096.1 contributor: fullname: Eckel – volume: 11 start-page: 1203 year: 1972 ident: ref301 article-title: The use of Model Output Statistics (MOS) in objective weather forecasting publication-title: J. Appl. Meteor. Climatol. doi: 10.1175/1520-0450(1972)011<1203:TUOMOS>2.0.CO;2 contributor: fullname: Glahn – volume: 2014 year: 2014 ident: ref241 article-title: Dynamical analysis of the Lorenz-84 atmospheric circulation model publication-title: J. Appl. Math. doi: 10.1155/2014/296279 contributor: fullname: Wang – volume: 11 start-page: 1203 year: 1972 ident: ref31 article-title: The use of Model Output Statistics (MOS) in objective weather forecasting publication-title: J. Appl. Meteor. Climatol. doi: 10.1175/1520-0450(1972)011<1203:TUOMOS>2.0.CO;2 contributor: fullname: Glahn – volume: 17 start-page: 206 year: 2002 ident: ref521 article-title: The Canadian Updateable Model Output Statistics (UMOS) system: Design and development tests publication-title: Wea. Forecasting doi: 10.1175/1520-0434(2002)017<0206:TCUMOS>2.0.CO;2 contributor: fullname: Wilson – volume: 144 start-page: 897 year: 2016 ident: ref11 article-title: A hybrid NWP—Analog ensemble publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-15-0096.1 contributor: fullname: Eckel – volume: 323 start-page: 533 year: 1986 ident: ref181 article-title: Learning representations by back-propagating errors publication-title: Nature doi: 10.1038/323533a0 contributor: fullname: Rumelhart – start-page: 9 year: 1998 ident: ref71 article-title: andK Efficient backprop of the Trade publication-title: Neural Networks Tricks doi: 10.1007/3-540-49430-8_2 contributor: fullname: LeCun – volume: 18 start-page: 288 year: 2003 ident: ref531 article-title: The Canadian Updateable Model Output Statistics (UMOS) system: Validation against perfect prog publication-title: Wea. Forecasting doi: 10.1175/1520-0434(2003)018<0288:TCUMOS>2.0.CO;2 contributor: fullname: Wilson – volume: 42A start-page: 378 year: 1990 ident: ref361 article-title: Can chaos and intransitivity lead to interannual variability? publication-title: Tellus doi: 10.3402/tellusa.v42i3.11884 contributor: fullname: Lorenz – volume: 147 start-page: 4241 year: 2019 ident: ref441 article-title: Using a coevolutionary post-processor to improve skill for both forecasts of surface temperature and nowcasts of convection occurrence publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-19-0063.1 contributor: fullname: Roebber – volume: 17 start-page: 206 year: 2002 ident: ref251 article-title: The Canadian Updateable Model Output Statistics (UMOS) system: Design and development tests publication-title: Wea. Forecasting doi: 10.1175/1520-0434(2002)017<0206:TCUMOS>2.0.CO;2 contributor: fullname: Wilson – volume: 131 start-page: 2510 year: 2003 ident: ref201 article-title: Short-range ensemble predictions of 2-m temperature and dewpoint temperature over New England publication-title: Mon. Wea. Rev. doi: 10.1175/1520-0493(2003)131<2510:SEPOMT>2.0.CO;2 contributor: fullname: Stensrud – volume: 36A start-page: 98 year: 1984 ident: ref81 article-title: Irregularity: A fundamental property of the atmosphere publication-title: Tellus doi: 10.1111/j.1600-0870.1984.tb00230.x contributor: fullname: Lorenz – start-page: 495 volume-title: Elements of Physical Biology year: 1925 ident: ref111 contributor: fullname: Lotka – volume: 131 start-page: 2510 year: 2003 ident: ref471 article-title: Short-range ensemble predictions of 2-m temperature and dewpoint temperature over New England publication-title: Mon. Wea. Rev. doi: 10.1175/1520-0493(2003)131<2510:SEPOMT>2.0.CO;2 contributor: fullname: Stensrud – volume: 148 start-page: 2135 year: 2020 ident: ref41 article-title: Forecasting severe weather with random forests publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-19-0344.1 contributor: fullname: Hill – volume: 26 start-page: 564 year: 2011 ident: ref121 article-title: Updating short-term probabilistic weather forecasts of continuous variables using recent observations publication-title: Wea. Forecasting doi: 10.1175/WAF-D-11-00022.1 contributor: fullname: Nipen – start-page: 9 year: 1998 ident: ref341 article-title: andK Efficient backprop of the Trade publication-title: Neural Networks Tricks doi: 10.1007/3-540-49430-8_2 contributor: fullname: LeCun – volume: 143 start-page: 1497 year: 2015a ident: ref421 article-title: Adaptive evolutionary programming publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-14-00095.1 contributor: fullname: Roebber – volume: 147 start-page: 1769 year: 2019 ident: ref271 article-title: Constrained quantile regression splines for ensemble postprocessing publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-18-0420.1 contributor: fullname: Bremnes – volume: 55 start-page: 773 year: 2016 ident: ref61 article-title: Nearest neighbor–genetic algorithm for downscaling of climate change data from GCMs publication-title: J. Appl. Meteor. Climatol. doi: 10.1175/JAMC-D-15-0100.1 contributor: fullname: Kim – volume: 143 start-page: 1506 year: 2015b ident: ref161 article-title: Using evolutionary programs to maximize minimum temperature forecast skill publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-14-00096.1 contributor: fullname: Roebber – volume: 10 start-page: 689 year: 1995 ident: ref51 article-title: Diurnal corrections of short-term temperature forecasts using the Kalman filter publication-title: Wea. Forecasting doi: 10.1175/1520-0434(1995)010<0689:DCOSTS>2.0.CO;2 contributor: fullname: Homleid – start-page: 9 year: 1931 ident: ref231 article-title: Variations and fluctuations of the number of individuals in animal species living together publication-title: Animal Ecology contributor: fullname: Volterra – volume: 353 start-page: 241 year: 1991 ident: ref371 article-title: Dimension of weather and climate attractors publication-title: Nature doi: 10.1038/353241a0 contributor: fullname: Lorenz – volume: 143 start-page: 1497 year: 2015a ident: ref151 article-title: Adaptive evolutionary programming publication-title: Mon. Wea. Rev. doi: 10.1175/MWR-D-14-00095.1 contributor: fullname: Roebber |
SSID | ssj0006642 |
Score | 2.4063447 |
Snippet | Abstract
We introduce an adaptive form of postprocessor where algorithm structures are neural networks where the number of hidden nodes and the network... We introduce an adaptive form of postprocessor where algorithm structures are neural networks where the number of hidden nodes and the network training... |
SourceID | proquest crossref |
SourceType | Aggregation Database |
StartPage | 4045 |
SubjectTerms | Algorithms Artificial neural networks Back propagation Back propagation networks Computer applications Computer architecture Data storage Dynamic models Forecasting data Information storage Modelling Neural networks Numerical prediction Numerical weather forecasting Predators Prediction models Prey Regression analysis Storage requirements Training Weather forecasting |
Title | Toward an Adaptive Artificial Neural Network–Based Postprocessor |
URI | https://www.proquest.com/docview/2676143818/abstract/ |
Volume | 149 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3NbtNAEF6V9sIFQQsiUKo9IIRUOdi7Xq99TJqmoaVRVVrR22p3vVaRqiRKXVXtiXfgDXkSZn8cJwUh4GI7tuRY83n-PDPfIvQ25rIisVZRpSsa2cpSpMCPWb2yhFKgfrGdHT4eZ6Pz9PCCXTTLu4fpklp19f1v50r-B1U4B7jaKdl_QHZxUzgBx4AvbAFh2P4dxq7ndRdUtFfKmWsC6s1d84_9Dm55N9zONXo3XQ20D36rdIv0zvyQwHS-HKGCkteXV3e7tz42DLMtbV3GKOVRtj2FoagUvhqQ5GEHRlsOqs10vrCzoVV02WIS2x7nORq7JhhJYltC_cqGCyvqmUeb14UsGcU09oyRv1prboktjr-cRoMIHhHCkaKbtI6pKcaPep_FyWAoPn0cHz1CG4QXzKbZ-fBg4XOzLA288P5pA4kT_MGHB7dfjT9W3a-LKc6eoichGcA9j-wztGYmm6jTCusOv8N7V18hqXC_tlDfI47lBDeI4xZx7BHHAfEf3747rPEK1s_R-XD_bG8UhVUwIg2hWx2xAqJUrlPIM3lOK61LrotYZoolnJcFMZAis5waUlVlTqUsYKeZIVoySVNW0RdofTKdmJcIs1RllqGQ8kSnRUlVoQBGBkmqplLlsoPeN7IRM092IlySyJkAMYqBIImwYhRJB203shNBI64FyXjmVDx_9efLr9Hj9qXcRuv1_Ma8geCuVjsO1x200d8fn5z-BNDWUBI |
link.rule.ids | 315,786,790,27957,27958,33408,33779 |
linkProvider | ProQuest |
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%3Ajournal&rft.genre=article&rft.atitle=Toward+an+Adaptive+Artificial+Neural+Network%E2%80%93Based+Postprocessor&rft.jtitle=Monthly+weather+review&rft.au=Roebber%2C+Paul+J&rft.date=2021-12-01&rft.pub=American+Meteorological+Society&rft.issn=0027-0644&rft.eissn=1520-0493&rft.volume=149&rft.issue=12&rft.spage=4045&rft_id=info:doi/10.1175%2FMWR-D-21-0089.1&rft.externalDBID=HAS_PDF_LINK |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0027-0644&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0027-0644&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0027-0644&client=summon |