Introduction to Nonlinear Statistics and Neural Networks：介绍非线性统计和神经网络精选.pptVIP

下载本文档

1
0
约2.53万字
约 55页
2018-04-30 发布于湖北
举报
版权申诉

Introduction to Nonlinear Statistics and Neural Networks：介绍非线性统计和神经网络精选.ppt

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Introduction to Nonlinear Statistics and Neural Networks：介绍非线性统计和神经网络精选

3/11/2009 Meto 630; V.Krasnopolsky, Nonlinear Statistics and NNs * NN Emulation of Input/Output Dependency: Input/Output Dependency: The Magic of NN Performance Xi Original Parameterization Yi Y = F(X) Xi NN Emulation Yi YNN = FNN(X) Mathematical Representation of Physical Processes Numerical Scheme for Solving Equations Input/Output Dependency: {Xi,Yi}I = 1,..N 3/11/2009 Meto 630; V.Krasnopolsky, Nonlinear Statistics and NNs * Neural Networks for NCAR (NCEP) LW Radiation NN characteristics 220 (612 for NCEP) Inputs: 10 Profiles: temperature; humidity; ozone, methane, cfc11, cfc12, N2O mixing ratios, pressure, cloudiness, emissivity Relevant surface characteristics: surface pressure, upward LW flux on a surface - flwupcgs 33 (69 for NCEP) Outputs: Profile of heating rates (26) 7 LW radiation fluxes: flns, flnt, flut, flnsc, flntc, flutc, flwds Hidden Layer: One layer with 50 to 300 neurons Training: nonlinear optimization in the space with dimensionality of 15,000 to 100,000 Training Data Set: Subset of about 200,000 instantaneous profiles simulated by CAM for the 1-st year Training time: about 1 to several days (SGI workstation) Training iterations: 1,500 to 8,000 Validation on Independent Data: Validation Data Set (independent data): about 200,000 instantaneous profiles simulated by CAM for the 2-nd year 3/11/2009 Meto 630; V.Krasnopolsky, Nonlinear Statistics and NNs * Neural Networks for NCAR (NCEP) SW Radiation NN characteristics 451 (650 NCEP) Inputs: 21 Profiles: specific humidity, ozone concentration, pressure, cloudiness, aerosol mass mixing ratios, etc 7 Relevant surface characteristics 33 (73 NCEP) Outputs: Profile of heating rates (26) 7 LW radiation fluxes: fsns, fsnt, fsdc, sols, soll, solsd, solld Hidden Layer: One layer with 50 to 200 neurons Training: nonlinear optimization in the space with dimensionality of 25,000 to 130,000 Training Data Set: Subset of about 200,000 instantaneous profiles simulated by CAM for the 1-st year Training ti