Improving generalization of MLPs with multi-objective optimization

نویسندگان

  • Roselito de Albuquerque Teixeira
  • Antônio de Pádua Braga
  • Ricardo H. C. Takahashi
  • Rodney R. Saldanha
چکیده

This paper presents a new learning scheme for improving generalization of multilayer perceptrons. The algorithm uses a multi-objective optimization approach to balance between the error of the training data and the norm of network weight vectors to avoid over"tting. The results are compared with support vector machines and standard backpropagation. ( 2000 Elsevier Science B.V. All rights reserved.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Decisior Implementation in Neural Model Selection by Multi-objective Optimization

This work presents a new learning scheme for improving generalization of Multilayer Perceptrons (MLPs). The proposed Multi-objective algorithm (MOBJ) approach minimizes both the sum of squared error and the norm of network weight vectors to obtain the Pareto-optimal solutions [1]. Preliminar results are shown in [3]. Since the Pareto-optimal solutions are not unique, we need a decision phase in...

متن کامل

An Approach to Reducing Overfitting in FCM with Evolutionary Optimization

Fuzzy clustering methods are conveniently employed in constructing a fuzzy model of a system, but they need to tune some parameters. In this research, FCM is chosen for fuzzy clustering. Parameters such as the number of clusters and the value of fuzzifier significantly influence the extent of generalization of the fuzzy model. These two parameters require tuning to reduce the overfitting in the...

متن کامل

The Quasi-Normal Direction (QND) Method: An Efficient Method for Finding the Pareto Frontier in Multi-Objective Optimization Problems

In managerial and economic applications, there appear problems in which the goal is to simultaneously optimize several criteria functions (CFs). However, since the CFs are in conflict with each other in such cases, there is not a feasible point available at which all CFs could be optimized simultaneously. Thus, in such cases, a set of points, referred to as 'non-dominate' points (NDPs), will be...

متن کامل

Multi Objective Optimization Placement of DG Problem for Different Load Levels on Distribution Systems with Purpose Reduction Loss, Cost and Improving Voltage Profile Based on DAPSO Algorithm

Along with economic growth of countries which leads to their increased energy requirements,the problem of power quality and reliability of the networks have been more considered andin recent decades, we witnessed a noticeable growing trend of distributed generation sources(DG) in distribution networks. Occurrence of DG in distribution systems, in addition tochanging the utilization of these sys...

متن کامل

Urban Land-Use Allocation By A Cell-based Multi-Objective Optimization Algorithm

Allocating urban land-uses to land-units with regard to different criteria and constraints is considered as a spatial multi-objective problem. Generating various urban land-use layouts with respect to defined objectives for urban land-use allocation can support urban planners in confirming appropriate layouts. Hence, in this research, a multi-objective optimization algorithm based on grid is pr...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • Neurocomputing

دوره 35  شماره 

صفحات  -

تاریخ انتشار 2000