检索结果-内蒙古大学图书馆

Influence of model complexity, training collinearity, collinearity shift, predictor novelty and their interactions on ecological forecasting

引用

GLOBAL ECOLOGY AND BIOGEOGRAPHY 2024年第3期33卷 371-384页

作者： Chen, Xin Liang, Ye Feng, Xiao Florida State Univ Dept Geog Tallahassee FL 32306 USA Univ Maryland Ctr Environm Sci Appalachian Lab Frostburg MD USA Oklahoma State Univ Dept Stat Stillwater OK USA

Aim: Ecological forecasting is critical in understanding of ecological responses to climate change and is increasingly used in climate mitigation plans. The forecasts from correlative models can be challenged by model complexity, training collinearity, collinearity shift and novel conditions of predictors that are common during model extrapolation. The individual effect of these four factors has been investigated, but it is still unclear how these four factors interactively affect forecasting. To fill this gap, we conducted a comprehensive simulation experiment to quantify how the four factors interactively influence model forecasting. Location: Simulated regions. Time Period: Simulated ***: We modelled three response variables commonly used in ecological forecasting following normal, Poisson and binomial distributions as a function of three functional relationships that represented model complexity under three levels of training collinearity using generalized linear models. By calculating prediction error under 3,780,000 testing scenarios, we partitioned its variance to model complexity, training collinearity, collinearity shift, predictor novelty and their interactions. Results: We found that increased predictor novelty and collinearity shift degraded model performance, leading up to double prediction errors when a predictor's range increased by similar to 22% or when the correlation r between two predictors changed >similar to 0.8 for the combination of high training collinearity and interaction functional relationship. Predictor novelty reduced the influence of collinearity shift on model performance, suggesting a negative interaction between them. This pattern was more pronounced under high model complexity and high training collinearity. Main Conclusions: The accuracy of ecological forecasting using correlative models depends on model complexity, training collinearity, collinearity shift, predictor novelty and their interactions. Besides the cons

关键词： collinearity shift ecological forecasting extrapolation model complexity model performance multi-collinearity predictor novelty training collinearity

来源：评论

学校读者我要写书评

暂无评论

Finite segment model complexity of an Euler-Bernoulli beam

引用

IFAC-PapersOnLine 2015年第1期28卷 334-340页

作者： Louca, Loucas S. University of Cyprus Department of Mechanical and Manufacturing Engineering Cyprus

A common approach for modeling the dynamic behavior of distributed parameter systems is the approximation through finite-segment models. These models are able to accurately predict the dynamic behavior of the system given that "adequate" segments are included in the model. Frequency-based methodologies can be used to address the complexity of such models. The purpose of the current work is to address the complexity of distributed parameter using the previously developed activity metric. More specifically the complexity of an Euler-Bernoulli beam model is considered. Bond graph models of this system already exist in the literature and the objective is to identify the necessary complexity (number of segments). A new modeling procedure is proposed for this type of systems where the model starts from simple and the number of segments is increased until an activity based criterion is satisfied. An illustrative example is provided to demonstrate the effectiveness of this methodology. © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

关键词： Graph theory Bond graph Distributed parameter Distributed parameter systems Energy based models Euler Bernoulli beams Euler bernoulli beam models Finite segments model complexity

来源：评论

学校读者我要写书评

暂无评论

On model complexity and selection

引用

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION 2003年第1期73卷 45-57页

作者： Carapeto, M Holt, W Refenes, APN Cass Business Sch Fac Finance London EC1Y 8TZ England London Business Sch Dept Decis Sci London NW1 4SA England

In the past, practitioners and researchers have compared the performance of neural networks with other model classes based on the multiple correlation coefficient or empirical validation. Such comparisons are biased towards neural networks as. such performance metrics do not account for model complexity. model complexity metrics are essential for parameter significance (e.g., F-test) and model mis-specification tests (e.g., autocorrelation). The estimation of degrees of freedom from the projection matrix of regression is therefore vitally important in all phases of the model building process for neural regression models. Degrees of freedom are used to measure model complexity and thus adjust statistics so that they may be meaningfully applied to regression models. In this paper we derive expressions for the influence matrix in linear and non-linear regression models and non-linear models with regularization, including neural networks. We show that they can be obtained within the same framework. In particular, we demonstrate that previous results obtained for neural networks hold for models without regularization terms or large sample sizes. We show how these results are used to adjust the multiple correlation coefficient by the degrees of freedom. The methodology is demonstrated using simulated data from a Cobb-Douglas type function.

关键词： degrees of freedom model complexity non-linear regression neural networks adjusted R-2

来源：评论

学校读者我要写书评

暂无评论

Reducing model complexity of the General Markov model of Evolution

引用

MOLECULAR BIOLOGY AND EVOLUTION 2011年第11期28卷 3045-3059页

作者： Jayaswal, Vivek Ababneh, Faisal Jermiin, Lars S. Robinson, John CSIRO Div Ecosyst Sci Canberra ACT Australia Univ Sydney Sch Math & Stat Sydney NSW 2006 Australia Univ Sydney Ctr Math Biol Sydney NSW 2006 Australia Al Hussein Bin Talal Univ Dept Math & Stat Maan Jordan Univ Sydney Sch Biol Sci Sydney NSW 2006 Australia Univ Roma La Sapienza Dept Biochem Sci Rossi Fanelli Rome Italy

The selection of an optimal model for data analysis is an important component of model-based molecular phylogenetic studies. Owing to the large number of Markov models that can be used for data analysis, model selection is a combinatorial problem that cannot be solved by performing an exhaustive search of all possible models. Currently, model selection is based on a small subset of the available Markov models, namely those that assume the evolutionary process to be globally stationary, reversible, and homogeneous. This forces the optimal model to be time reversible even though the actual data may not satisfy these assumptions. This problem can be alleviated by including more complex models during the model selection. We present a novel heuristic that evaluates a small fraction of these complex models and identifies the optimal model.

关键词： phylogenetics general Markov models model complexity model selection parameter estimation

来源：评论

学校读者我要写书评

暂无评论

On the complexity of model complexity: Viewpoints across the geosciences

引用

CATENA 2020年 186卷

作者： Baartman, Jantiene E. M. Melsen, Lieke A. Moore, Demie van der Ploeg, Martine J. Wageningen Univ Soil Phys & Land Management Grp Wageningen Netherlands Wageningen Univ Hydrol & Quantitat Water Management Grp Wageningen Netherlands POB 47 NL-6700 AA Wageningen Netherlands Droevendaalsesteeg 4 NL-6708 PB Wageningen Netherlands

It is the core task of geoscientists to gain insight into the complex systems of nature. Yet, complexity may be perceived very differently and a plethora of models with different degrees of complexity is available. How do we, geoscientists, decide what model complexity is warranted? Does this differ among disciplines? And, how do we even define model complexity? We developed a short questionnaire to investigate the geoscientific community's views on complexity in models. The response was overwhelming, with 618 completed responses. The results show that the number of processes explicitly included and the number of interactions / feedbacks incorporated were seen as important determinants of complexity. Confidence was not per se higher in the simulations of a complex model compared to a simple one. Interestingly, neither gender, the discipline within the geosciences, nor career stage or work sector, explained the characterization of model complexity. The results of the questionnaire demonstrate that there is no general consensus on how model complexity is perceived or should be defined, and that formal definitions are not broadly or generally accepted. In an environment seeking greater collaboration and interdisciplinarity, these results indicate the need for conscious dialogue about this topic among different model users.

关键词： model complexity Geosciences Perception Questionnaire

来源：评论

学校读者我要写书评

暂无评论

Controlling model complexity in Probabilistic model-Based Dynamic Optimization of Neural Network Structures 28th

Controlling Model Complexity in Probabilistic Model-Based Dy...

引用

28th International Conference on Artificial Neural Networks (ICANN)

作者： Saito, Shota Shirakawa, Shinichi Yokohama Natl Univ Yokohama Kanagawa Japan SkillUp Co Ltd Tokyo Japan

ISBN: (纸本)9783030304843;9783030304836

A method of simultaneously optimizing both the structure of neural networks and the connection weights in a single training loop can reduce the enormous computational cost of neural architecture search. We focus on the probabilistic model-based dynamic neural network structure optimization that considers the probability distribution of structure parameters and simultaneously optimizes both the distribution parameters and connection weights based on gradient methods. Since the existing algorithm searches for the structures that only minimize the training loss, this method might find overly complicated structures. In this paper, we propose the introduction of a penalty term to control the model complexity of obtained structures. We formulate a penalty term using the number of weights or units and derive its analytical natural gradient. The proposed method minimizes the objective function injected the penalty term based on the stochastic gradient descent. We apply the proposed method in the unit selection of a fully-connected neural network and the connection selection of a convolutional neural network. The experimental results show that the proposed method can control model complexity while maintaining performance.

关键词： Neural networks Structure optimization Stochastic natural gradient model complexity Stochastic relaxation

来源：评论

学校读者我要写书评

暂无评论

Does model complexity improve pricing accuracy? The case of CoCos

引用

REVIEW OF DERIVATIVES RESEARCH 2021年第3期24卷 261-284页

作者： Koziol, Christian Weitz, Sebastian Univ Tubingen Dept Finance Nauklerstr 47 D-72074 Tubingen Germany

In this study, we analyze whether model complexity improves accuracy of CoCo pricing models. We compare the out-of-sample pricing ability of four models using a broad dataset that contains all CoCos which were issued between January 1, 2013 and May 31, 2016 in euros. The regarded models include the standard model from De Spiegeleer and Schoutens (J Deriv 20:27-36, 2012), a modified version enriched by credit risk, an extended model that accounts for the effective lifetime of the CoCo, and a trading model, solely based on historic market prices but no pricing theory at all. For a normal market environment, the simple trading model provides a higher pricing accuracy than the theory-based models. Under distress, however, a theory-based model with a sufficiently high complexity is required.

关键词： Contingent convertible bond CoCo bond CoCo pricing Continuous-time derivatives pricing model complexity Test of pricing models

来源：评论

学校读者我要写书评

暂无评论

Impact of model complexity on CO₂ plume modeling at Sleipner 12

Impact of Model Complexity on CO<sub>2</sub> plume modeling ...

引用

12th International Conference on Greenhouse Gas Control Technologies (GHGT)

作者： Bandilla, Karl W. Celia, Michael A. Leister, Evan Princeton Univ Dept Civil & Environm Engn Princeton NJ 08544 USA

The goal of geologic carbon sequestration (GCS) is to store carbon dioxide (CO2) in the subsurface for time periods on the order of thousands of years. Mathematical modeling is an important tool to predict the migration of both CO2 and brine to ensure safe and permanent storage. Many modeling approaches with different levels of complexity have been applied to the problem of GCS ranging from simple analytic solutions to fully-coupled three-dimensional reservoir simulators. The choice of modeling approach is often a function of the spatial and temporal scales of the problem, reservoir properties, data availability, available computational resources, and the familiarity of the modeler with a specific modeling approach. In this study we apply a series of models with different levels of model complexity to the 9th layer of the Utsira Formation. The list of modeling approaches includes (from least complex to most complex): numerical vertical-equilibrium model with sharp-interface, numerical vertical-equilibrium model with capillary transition zone, vertically-integrated model with dynamic vertical pressure and saturation reconstruction, and fully-coupled three-dimensional model. The model domain consists of a 3 x 6 km section of the 9th layer, as described in the IEAGHG benchmark dataset. The layer thickness varies in space, ranging from 5 to 30 m, while porosity and permeability are close to constant at 0.36 and 1.8 Darcy, respectively. The models are all based on the same input data, and initial and boundary conditions are chosen in a way that ensures the different models are comparable. In addition, a simple box model is used for preliminary simulations. The models are compared based on the predicted CO2 plume footprints and saturation cross-sections. The predicted CO2 plumes are also compared to the actual CO2 plume footprint from seismic surveys to determine the ability of the different models to predict the actual CO2 plume footprint. The results show that vertical-

关键词： carbon sequestration modeling model complexity Sleipner vertical-equilibrium modeling

来源：评论

学校读者我要写书评

暂无评论

modelling a Crawler Robot Using Wheels as Pseudo-Tracks: model complexity vs Performance 7

Modelling a Crawler Robot Using Wheels as Pseudo-Tracks: Mod...

引用

IEEE 7th International Conference on Industrial Engineering and Applications (ICIEA)

作者： Moskvin, Ilya Lavrenov, Roman Magid, Evgeni Svinin, Mikhail Kazan Fed Univ Intelligent Robot Dept Lab Intelligent Robot Syst Kazan Russia Ritsumeikan Univ Coll Informat Sci & Engn Robot Dynam & Control Lab Kyoto Japan

ISBN: (纸本)9781728167855

Crawler mobile robots have a large variety of applications, including urban search and rescue. Before new algorithms and concepts could be integrated into a real robot control system and tested in field experiments, extensive virtual simulations should be performed in order to verify them carefully. This requires a proper modelling of a robot in a virtual environment, but a specific construction of crawler robot crawlers complicates the modelling. In this paper, in order to provide a high level of physical interaction similarity of a crawler with a supporting surface we model crawlers of mobile robot Servosila Engineer with a set of virtual wheels and study a relationship between a model complexity and a simulation performance. The model complexity is reflected by a number of virtual wheels that approximate each crawler of the robot. Verification experiments were performed in ROS/Gazebo simulator.

关键词： ROS Gazebo modelling mobile robot crawler robot simulation performance model complexity virtual wheels

来源：评论

学校读者我要写书评

暂无评论

Evaluation of model complexity in model predictive control within an exercise-enabled artificial pancreas

引用

IFAC-PapersOnLine 2017年第1期50卷 7756-7761页

作者： Resalat N. Youssef J.E. Reddy R. Jacobs P.G. Oregon Health & Science University Department of Biomedical Engineering 3303 SW Bond Ave Portland OR 97239 United States Oregon Health & Science University Harold Schnitzer Diabetes Center Portland OR 97239 United States

model predictive control (MPC) algorithms have been used often within artificial pancreas control systems both in-silico and in clinical studies. Increasingly complex models in the controller can more accurately predict the glycemic response, but they introduce increased computational complexity which can be challenging to implement especially within an embedded environment where computational resources are limited. Less complex models are also preferable in that they can be evaluated in silico against more complex plant models. There has not yet been an evaluation of how the complexity of models used within an MPC impacts performance within an artificial pancreas. A model within an artificial pancreas MPC algorithm should be as complex as necessary to accurately predict a glycemic response to meals, exercise, stress, and other disturbances, but not overly complex. In this paper, we evaluate four glucoregulatory models used within an MPC, starting with a 4-state model and increasing in complexity up to six states. We evaluate the complexity using an in-silico population derived from a more complex glucoregulatory model (9 state variables). We assess how complexity of the model impacts performance both in terms of standard control metrics such as settling time and overshoot as well as clinically relevant metrics such as percent time in euglycemia (glucose between 70 and 180 mg/dl), percent time in hypoglycemia (70 mg/dl) and percent time in hyperglycemia (>180 mg/dl). We find that model complexity matters far less than how well the model parameters match the individual subjects. When the simplest model is used, but fit to an individual subject's data, it performed comparably with more complex models. We selected a middle-complexity model and integrated it into our previously published exercise-enabled MPC model and evaluated it in a virtual patient population both with and without the exercise model present. We found that increasing complexity by modeling exercise is

关键词： Artificial Pancreas Glucoregulatory model model complexity model Predictive Control Type-1 diabetes

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：