Variance Reduction Techniques for the Identification of Noisy Systems

用于识别噪声系统的方差减少技术

• 批准号: 9626406
• 负责人: John Moody
• 金额: $ 10万
• 依托单位: Oregon Graduate Institute of Science & Technology
• 依托单位国家: 美国
• 项目类别: Continuing grant
• 财政年份: 1997
• 资助国家: 美国
• 起止时间: 1997-02-15 至 2001-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9626406&HistoricalAwards=false
• 关键词: Variance; Reduction; Techniques; Identification; Noisy

The identification of systems with high noise levels is very challenging, since models of such systems are typically plagued by high model variance. This lead to higher expected prediction errors. This project will investigate two promising new approaches for reducing model variance and thus the variance of prediction errors: (1) New classes of smoothing regularizes for both feedforward and recurrent networks for reducing model variance while imposing desirable model biases. The PI expect, that his new smoothing regularizes will outperform standard quadratic weight decay, and ad hoc methods, in many cases of interest. (2) New committee bootstrap methods for reducing the prediction errors due to model variance. These include independent bootstrapping of training and validation sets within the committee, mutual training and model selection methods, and robust adaptive committees. The PI expects that his new committee bootstrap methods will achieve better training, better model selection, and greater variance reduction than is attainable be individual networks or by conventional committee averaging methods. The research will involve new analytical work, algorithm development, and extensive empirical testing of the algorithms on noisy time series prediction problems n macroeconomics, physiology, and engineering.

具有高噪声水平的系统的识别是非常具有挑战性的，因为这种系统的模型通常受到高模型方差的困扰。这导致更高的预期预测误差。 本项目将研究两种有前途的新方法，以减少模型方差，从而预测误差的方差：（1）新类的平滑正则化的前馈和递归网络，以减少模型方差，同时施加理想的模型偏差。 PI期望，他的新平滑正则化将优于标准的二次权重衰减，以及在许多感兴趣的情况下的特设方法。 (2)新的委员会引导方法，以减少由于模型方差的预测误差。 这些包括委员会内部训练和验证集的独立自举，相互训练和模型选择方法，以及强大的自适应委员会。 PI希望他的新委员会自助方法将实现更好的训练，更好的模型选择，以及比单个网络或传统委员会平均方法更大的方差减少。 这项研究将涉及新的分析工作，算法开发，以及对噪声时间序列预测问题的算法进行广泛的实证测试，包括宏观经济学，生理学和工程学。