cross efficiency is a method with the idea of peer evaluation instead of self-evaluation, and is used for evaluation and ranking decision making units (dmus) in data envelopment analysis (dea). unlike most existing dea ranking models which can only rank a subset of dmus, for example non-efficient or extreme efficient dmus, cross efficiency can rank all dmus, even non-extreme ones. however, sinc...

We discuss how to build sparse one hidden layer MLP replacing the standard l2 weight decay penalty on all weights by an l1 penalty on the linear output weights. We will propose an iterative two step training procedure where the output weights are found using FISTA proximal optimization algorithm to solve a Lasso-like problem and the hidden weights are computed by unconstrained minimization. As ...

data envelopment analysis (dea) as a non-parametric method for efficiency measurement allows decision making units (dmus) to select the most advantageous weight factors in order to maximize their efficiency scores.  in most practical applications of dea presented in the literature, the presented models assume that all inputs are fully desirable. however, in many real situations undesirable inpu...

In this paper, by investigating the common weights concept and DEA models with nonlinear virtual inputs/outputs, we introduce a model for evaluating the decision making units with nonlinear virtual inputs and outputs based on the common weights.

Data envelopment analysis which is a nonparametric technique for evaluating relative efficiency of the decision making units with multiple inputs and outputs, has been a very popular method among researchers. While this nonparametric technique is popular, it has some drawbacks such as lack of discrimination in efficient units and weights dispersion .The present study, which is a model based on ...

A typical feedforward network has units arranged in a distinct layered topology. Units are connected to one another and each connection is associated with a real number, which is called the weight of the connection. During network training, the connection weights are adjusted in order to correctly classify the training data. The network weights are basically dependent on the training data set a...

The present study is an attempt toward evaluating the performance of portfolios and asset selectionusing cross-efficiency evaluation. Cross-efficiency evaluation is an effective way of ranking decisionmaking units (DMUs) in data envelopment analysis (DEA). Conventional DEA models assume nonnegativevalues for inputs and outputs. However, we know that unlike return and skewness, varianceis the on...

This paper explores approaches to the problem of spoken document retrieval (SDR), which is the task of automatically indexing and then retrieving relevant items from a large collection of recorded speech messages in response to a user specified natural language text query. We investigate the use of subword unit representations for SDR as an alternative to words generated by either keyword spott...