检索结果-内蒙古大学图书馆

PLS modelling of detergency performance for some technical nonionic surfactants

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS 1996年第1期32卷 111-124页

作者： Lindgren, A Sjostrom, M Wold, S Research Group for Chemometrics Ume? Uniuersity 901 87 Ume? Sweden

The detergency effect of a series of technical nonionic surfactants has been examined. The detergency experiments have been performed according to different central composite circumscribed statistical designs. The data from the differently designed investigations were combined, and a response surface model was estimated for each surfactant with the use of partial least squares of latent structures. The combination of the designed series gave a new insight into the importance of a variable not considered before. The effects of different transformations and model complexities were investigated and a new approach to compare the efficiency of the different surfactants was found In the evaluation of the individual designs all replicates were used in the calculations, while the evaluation of the combined designs was made with the use of the mean values of each experimental point. The reason for this is presented.

关键词： nonionic surfactants detergency modelling statistical designs PLS MLR replicates transformations model complexity

来源：评论

学校读者我要写书评

暂无评论

Basin-scale modeling of CO₂ storage using models of varying complexity

引用

INTERNATIONAL JOURNAL OF GREENHOUSE GAS CONTROL 2014年第0期20卷 73-86页

作者： Huang, Xinwo Bandilla, Karl W. Celia, Michael A. Bachu, Stefan Princeton Univ Dept Civil & Environm Engn Princeton NJ 08544 USA Alberta Innovates Technol Futures Edmonton AB Canada

Geological carbon storage can significantly contribute to climate-change mitigation only if it is deployed at a very large scale. This means that injection scenarios must occur, and be analyzed, at the basin scale. Various mathematical models of different complexity may be used to assess the fate of injected CO2 and/or resident brine. These models span the range from multi-dimensional, multi-phase numerical simulators to simple single-phase analytical solutions. In this study, we consider a range of models, all based on vertically integrated governing equations, to predict the basin-scale pressure response to specific injection scenarios. The Canadian section of the mid-continent Basal Aquifer is used as a test site to compare the different modeling approaches. The model domain covers an area of approximately 811,000 km(2), and the total injection rate is 63 Mt/yr, corresponding to 9 locations where large point sources have been identified. Predicted areas of critical pressure exceedance are used as a comparison metric among the different modeling approaches. Comparison of the results shows that single-phase numerical models may be good enough to predict the pressure response over a large aquifer;however, a simple superposition of semi-analytical or analytical solutions is not sufficiently accurate because spatial variability of formation properties plays an important role in the problem, and these variations are not captured properly with simple superposition. We consider two different injection scenarios: injection at the source locations and injection at locations with more suitable aquifer properties. Results indicate that in formations with significant spatial variability of properties, strong variations in injectivity among the different source locations can be expected, leading to the need to transport the captured CO2 to suitable injection locations, thereby necessitating development of a pipeline network. (C) 2013 Elsevier Ltd. All rights reserved.

关键词： Geological carbon sequestration Large-scale pressure response model complexity Injection scenarios Basal Aquifer

来源：评论

学校读者我要写书评

暂无评论

Design considerations for knowledge source representations of a stochastically-based natural language understanding component

引用

SPEECH COMMUNICATION 1999年第2期28卷 141-154页

作者： Minker, W CNRS LIMSI Spoken Language Proc Grp F-91403 Orsay France

We describe a stochastic component for spoken natural language understanding in an application for train travel information retrieval in French. The aim is to discuss the design considerations for representing knowledge sources appropriately in such a parsing component. The development focuses on the design of a stochastic model topology that is optimally adapted in quality and complexity to the task model and the available training data. Another important issue concerns the iterative semantic labeling of large data amounts used for the component training, The parser has been evaluated on both corrected and uncorrected speech recognizer output transcriptions. (C) 1999 Elsevier Science B.V. All rights reserved.

关键词： Case grammar Hidden Markov models model complexity model topology Semantic representation

来源：评论

学校读者我要写书评

暂无评论

Bayesian Factorization and Learning for Monaural Source Separation

引用

IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING 2016年第1期24卷 185-195页

作者： Chien, Jen-Tzung Yang, Po-Kai Natl Chiao Tung Univ Dept Elect & Comp Engn Hsinchu 30010 Taiwan

This paper presents a new Bayesian nonnegative matrix factorization (NMF) for monaural source separation. Using this approach, the reconstruction error based on NMF is represented by a Poisson distribution, and the NMF parameters, consisting of the basis and weight matrices, are characterized by the exponential priors. A variational Bayesian inference procedure is developed to learn variational parameters and model parameters. The randomness in separation process is faithfully represented so that the system robustness to model variations in heterogeneous environments could be achieved. Importantly, the exponential prior parameters are used to impose sparseness in basis representation. The variational lower bound of log marginal likelihood is adopted as the objective to control model complexity. The dependencies of variational objective on model parameters are fully characterized in the derived closed-form solution. A clustering algorithm is performed to find the groups of bases for unsupervised source separation. The experiments on speech/music separation and singing voice separation show that the proposed Bayesian NMF (BNMF) with adaptive basis representation outperforms the NMF with fixed number of bases and the other BNMFs in terms of signal-to-distortion ratio and the global normalized source to distortion ratio.

关键词： Bayesian learning model complexity monaural source separation nonnegative matrix factorization

来源：评论

学校读者我要写书评

暂无评论

Stochastic cross validation

引用

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS 2018年 175卷 74-81页

作者： Xu, Lu Fu, Hai-Yan Goodarzi, Mohammad Cai, Chen-Bo Yin, Qiao-Bo Wu, Ya Tang, Bang-Cheng She, Yuan-Bin Zhejiang Univ Technol Coll Chem Engn Hangzhou 310014 Zhejiang Peoples R China Tongren Univ Coll Mat & Chem Engn Tongren 554300 Guizhou Peoples R China South Cent Univ Nationalities Sch Pharmaceut Sci Modernizat Engn Technol Res Ctr Ethn Minor Med Hu Wuhan 430074 Hubei Peoples R China Univ Texas Southwestern Med Ctr Dallas Dept Biochem Dallas TX 75390 USA Chuxiong Normal Univ Dept Chem & Life Sci Chuxiong 675000 Peoples R China

Cross validation (CV) is by far one of the most commonly used methods to estimate model complexity for partial least squares (PLS). In this study, stochastic cross validation (SCV) was proposed as a novel CV strategy, where the percent of left-out objects (PLOO) was defined as a changeable random number. We proposed two SCV strategies, namely, SCV with uniformly distributed PLOD (SCV-U) and SCV with normally distributed PLOD (SCV-N). SCV-U is actually a hybrid of leave-one-out CV (LOOCV), k-fold CV and Monte Carlo CV (MCCV). The rationale behind SCV-N is that the probability of large perturbations of the original training set will be small. SCV is expected to provide more flexibility for data splitting to explore and learn from the data set and evaluate internally a built model. SCV-U and SCV-N were used for PLS calibrations of three real data sets as well as a simulated data set and they were compared with LOOCV, k-fold CV and MCCV. Given a training and external validation set, different CV techniques were repeatedly used to evaluate the optimal model complexity and the prediction results were compared. The results indicate that SCV-U and SCV-N could provide useful alternatives to the traditional CV methods and SCV is less sensitive to the values of PLOO.

关键词： Multivariate calibration model complexity Partial least squares (PLS) Cross validation (CV) Stochastic cross validation (SCV)

来源：评论

学校读者我要写书评

暂无评论

UNIFIED APPROACH FOR EARLY-PHASE IMAGE UNDERSTANDING USING A GENERAL DECISION CRITERION

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 1989年第4期11卷 357-371页

作者： JEONG, DS LAPSA, PM PENN STATE UNIV DEPT ELECT ENGNUNIVERSITY PKPA 16802

Two types of approaches for computer vision are combined to model images or portions thereof, parametrically. These approaches, namely those based on polynomial models and those based on random-field models, are combined based on a general decision criterion for dealing with a variety of modeling strategies. Selection among alternative model structures is in accordance with the tradeoff between sample size and model complexity. Experiments with synthesized images and natural images such as Brodatz textures illustrate some identification and segmentation uses of this unified approach. The implemented segmentation algorithm achieves early-phase region extraction without relying on any contextual or high-level assumptions. A natural result of this is a list of regions, suitable as input for higher-level stages of image understanding in addition to a pixel-labeled image.

关键词： Polynomials Computer vision Image segmentation Pixel Parametric statistics Face detection Testing Layout Application software Image recognition INSPEC: Controlled Indexing polynomials computer vision computerised pattern recognition decision theory INSPEC: Non-Controlled Indexing pixel-labeled image computerised pattern recognition early-phase image understanding decision criterion computer vision polynomial models random-field models sample size model complexity Brodatz textures segmentation

来源：评论

学校读者我要写书评

暂无评论

model description of an alternately operated wastewater treatment plant - evaluation of the applicability Of SimpleTreat

引用

CHEMOSPHERE 2003年第3期50卷 283-292页

作者： Fauser, P Sorensen, PB Carlsen, L Vikelsoe, J Natl Environm Res Inst Dept Policy Anal & Environm Chem DK-4000 Roskilde Denmark Awareness Ctr DK-4000 Roskilde Denmark Natl Environm Res Inst Dept Environm Chem & Microbiol DK-4000 Roskilde Denmark

Alternately operated wastewater treatment plants (WWTPs) are fundamentally different compared to conventional activated sludge WWTPs with respect to flow patterns and aeration in the biological reactors. Several model applications exist for conventional WWTPs, e.g. SirnpleTreat, and in this study the effect of substituting a complex discontinuous operation, involving alternating degradation and flow conditions between two reactors, with one single bioreactor with continuos flow (SimpleTreat) has been investigated by setting up two models representing the respective operation schemes. The discontinuous operation induces fluctuations in the outlet concentrations that are not modelled with the single bioreactor model, however, the fluctuations and the associated uncertainties were found to be insignificant compared to the influence of the input parameter uncertainties on the model results. An empirical relationship between an aggregate pseudo- 1st order degradation rate for the single bioreactor model and realistic aerobic and anoxic Is( order degradation rates, respectively, has been established. When using this aggregate degradation rate in the single bioreactor model an outlet concentration can be calculated that deviates no more than 2% from the mean outlet concentration from the alternating operation model. For substances with aerobic half-lives longer than approximate to2 h, which is valid for many chemical substances, the aggregate 1st order degradation rate can be set equal to the aerobic 1st order degradation rate. (C) 2002 Elsevier Science Ltd. All rights reserved.

关键词： wastewater treatment plant alternating operation SimpleTreat model complexity continuous flow

来源：评论

学校读者我要写书评

暂无评论

A Survey of Neurodynamic Optimization

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2024年第4期8卷 2677-2696页

作者： Xia, Youshen Liu, Qingshan Wang, Jun Cichocki, Andrzej Nanjing Univ Informat Sci & Technol Coll Artificial Intelligence Nanjing 210044 Peoples R China Southeast Univ Sch Math Nanjing 210096 Peoples R China City Univ Hong Kong Sch Data Sci Dept Comp Sci Hong Kong Peoples R China Polish Acad Sci Syst Res Inst PL-02093 Warsaw Poland Tokyo Univ Agr & Technol Global Innovat Res Inst Tokyo 1838538 Japan RIKEN Adv Intelligence Project Tokyo 1030027 Japan

The last four decades have witnessed the birth and growth of neurodynamic optimization with numerous recurrent neural networks developed for solving various constrained optimization problems. Numerous results on neurodynamic optimization are reported in the literature,. In view of the diverse nature of the publications, this survey provides an updated overview of neurodynamic optimization to summarize the state-of-the-art results in terms of model structure, convergence property, and solvability scopes. It starts with an introduction and preliminaries, followed by categorizing many representative neural network models for constrained optimization, such as linear and quadratic programming, smooth and nonsmooth nonlinear programming, minimax optimization, distributed optimization, generalized-convex optimization, and global and mixed-integer optimization. In addition, it also delineates some perspective research topics for further investigations.

关键词： Optimization Neurodynamics Programming Vectors Recurrent neural networks Convergence Biological neural networks Constrained optimization recurrent neural network continuous and discontinuous state systems model complexity convergence criteria

来源：评论

学校读者我要写书评

暂无评论

Exploring the Learning Difficulty of Data: Theory and Measure

引用

ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA 2024年第4期18卷 1-37页

作者： Zhu, Weiyao Wu, Ou Su, Fengguang Deng, Yingjun Tianjin Univ Ctr Appl Math POB 92 Tianjin 300110 Peoples R China

"Easy/hard sample" is a popular parlance in machine learning. Learning difficulty of samples refers to how easy/hard a sample is during a learning procedure. An increasing need of measuring learning difficulty demonstrates its importance in machine learning (e.g., difficulty-based weighting learning strategies). Previous literature has proposed a number of learning difficulty measures. However, no comprehensive investigation for learning difficulty is available to date, resulting in that nearly all existing measures are heuristically defined without a rigorous theoretical foundation. This study attempts to conduct a pilot theoretical study for learning difficulty of samples. First, influential factors for learning difficulty are summarized. Under various situations conducted by summarized influential factors, correlations between learning difficulty and two vital criteria of machine learning, namely, generalization error and model complexity, are revealed. Second, a theoretical definition of learning difficulty is proposed on the basis of these two criteria. A practical measure of learning difficulty is proposed under the direction of the theoretical definition by importing the bias-variance trade-off theory. Subsequently, the rationality of theoretical definition and the practical measure are demonstrated, respectively, by analysis of several classical weighting methods and abundant experiments realized under all situations conducted by summarized influential factors. The mentioned weighting methods can be reasonably explained under the proposed theoretical definition and concerned propositions. The comparison in these experiments indicates that the proposed measure significantly outperforms the other measures throughout the experiments.

关键词： Learning difficulty generalization error bias-variance trade-off model complexity

来源：评论

学校读者我要写书评

暂无评论

complexity in modelling forest ecosystems: How much is enough?

引用

FOREST ECOLOGY AND MANAGEMENT 2008年第10期256卷 1646-1658页

作者： Kimmins, James P. (Hamish) Blanco, Juan A. Seely, Brad Welham, Clive Scoullar, Kim Univ British Columbia Fac Forestry Dept Forest Sci Vancouver BC V6T 1Z4 Canada Life Sci Programming Naramata BC V0H 1N0 Canada

The levels-of-integration concept in ecology suggests that while the population and community levels of biological organization provide understanding, they do not provide an adequate basis for predicting future states of populations, communities or ecosystems. Empirical models of populations and communities are implicitly ecosystem level because they are based on measures of the results of all past determinants that affected the populations and communities in question. They can be good predictors under unchanging conditions, but they are not able to predict for changing conditions, which requires the explicit representation of those ecological processes that are anticipated to change. The question addressed in this paper is: how does the performance of a "simple" stand-level model change as representations of these processes are added, and how much complexity is enough? Examples are given of change in performance of the ecosystem management model FORECAST run as a light competition model as nutrient regulation of growth and understory interactions are added singly or in combination. These examples are framed in comparisons of simulations for a white spruce stand in the in boreal mixedwood region in northern British Columbia, for a long-term thinning-fertilization trial in a Douglas-fir stand on southern Vancouver Island, and for yield decline in Chinese fir in south eastern China. It is concluded that the effect of adding complexity on the predictions of this model depends on the time scale of the simulation and the management or natural disturbance regime that is being simulated. The results supported Occam's razor: that models should be as simple as possible but as complex as necessary. (C) 2008 Elsevier B.V. All rights reserved.

关键词： Douglas-fir thinning trials Boreal mixedwoods Chinese fir yield decline Hybrid simulation FORECAST model model complexity

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：