Life stressors and social resources affect posttreatment outcomes among depressed patients.

Examined pre- and posttreatment changes in life stressors and social resources and their association with treatment outcomes in a 12-mo follow-up of 424 Ss (aged 18–83 yrs) with unipolar depression. Assessment measures included the Family Environment Scale, Work Environment Scale, and Health and Daily Living Form. As expected, Ss reported significant, multidimensional improvements in their functioning at follow-up. There were modest increases in Ss' social resources but, suprisingly, no overall decrease in stressors. Life stress and resource factors were significantly related to Ss' functioning at follow-up, even after considering the severity of their dysfunction at treatment intake and their length of treatment. Implications for developing a general body of knowledge about the process of recovery and relapse in behavioral disorders are discussed. (38 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Feature subset selection and ranking for data dimensionality reduction

A new unsupervised forward orthogonal search (FOS) algorithm is introduced for feature selection and ranking. In the new algorithm, features are selected in a stepwise way, one at a time, by estimating the capability of each specified candidate feature subset to represent the overall features in the measurement space. A squared correlation function is employed as the criterion to measure the dependency between features and this makes the new algorithm easy to implement. The forward orthogonalization strategy, which combines good effectiveness with high efficiency, enables the new algorithm to produce efficient feature subsets with a clear physical interpretation     

Neighborhood detection and rule selection from cellular automatapatterns

Using genetic algorithms (GAs) to search for cellular automation (CA) rules from spatio-temporal patterns produced in CA evolution is usually complicated and time-consuming when both, the neighborhood structure and the local rule are searched simultaneously. The complexity of this problem motivates the development of a new search which separates the neighborhood detection from the GA search. In the paper, the neighborhood is determined by independently selecting terms from a large term set on the basis of the contribution each term makes to the next state of the cell to be updated. The GA search is then started with a considerably smaller set of candidate rules pre-defined by the detected neighhorhood. This approach is tested over a large set of one-dimensional (1-D) and two-dimensional (2-D) CA rules. Simulation results illustrate the efficiency of the new algorithm     

Extracting Boolean rules from CA patterns

A multiobjective genetic algorithm (GA) is introduced to identify both the neighborhood and the rule set in the form of a parsimonious Boolean expression for both one- and two-dimensional cellular automata (CA). Simulation results illustrate that the new algorithm performs well even when the patterns are corrupted by static and dynamic noise.     

GENERALISED TRANSFER FUNCTIONS OF NEURAL NETWORKS

When artificial neural networks are used to model non-linear dynamical systems, the system structure which can be extremely useful for analysis and design, is buried within the network architecture. In this paper, explicit expressions for the frequency response or generalised transfer functions of both feedforward and recurrent neural networks are derived in terms of the network weights. The derivation of the algorithm is established on the basis of the Taylor series expansion of the activation functions used in a particular neural network. This leads to a representation which is equivalent to the non-linear recursive polynomial model and enables the derivation of the transfer functions to be based on the harmonic expansion method. By mapping the neural network into the frequency domain information about the structure of the underlying non-linear system can be recovered. Numerical examples are included to demonstrate the application of the new algorithm. These examples show that the frequency response functions appear to be highly sensitive to the network topology and training, and that the time domain properties fail to reveal deficiencies in the trained network structure.     

Parametric Characteristic Analysis for Generalized Frequency Response Functions of Nonlinear Systems

In order to explicitly reveal the relationship between system frequency response functions and model parameters which define system nonlinearities, and consequently unveil a direct connection from model parameters to system frequency response characteristics, a parametric characteristic analysis approach is proposed for Volterra systems described by a nonlinear differential equation (NDE). The parametric characteristics of the generalized frequency response functions (GFRFs) for the NDE model are established, and some important properties are discussed, which can explicitly reveal what model parameters contribute and how these parameters affect the GFRFs. Based on the parametric characteristic analysis, it is demonstrated how the system frequency domain characteristics are related to the system time domain model parameters and how the output frequency response function can now be determined explicitly with a detailed polynomial structure. These new results provide a significant and novel insight into the analysis and design of nonlinear systems in the frequency domain. Several examples are included to illustrate the results.     

Nonlinear system identification using wavelet networks

A novel identification scheme using wavelet networks is presented for nonlinear dynamical systems. Based on fixed wavelet networks, parameter adaptation laws are developed using a Lyapunov synthesis approach. This guarantees the stability of the overall identification scheme and the convergence of both the parameters and the state errors, even in the presence of modelling errors. Using the decomposition and reconstruction techniques of multiresolution decompositions, variable wavelet networks are introduced to achieve a desired estimation accuracy and a suitable sized network, and to adapt to variations of the characteristics and operating points in nonlinear systems. B-spline wavelets are used to form the wavelet networks and the identification scheme is illustrated using a simulated example.     

A new direct approach of computing multi-step ahead predictions for non-linear models

A new direct approach of computing multi-step ahead predictions for non-linear time series is introduced. The covariance of the parameter estimates associated with, and the mean squared k -step ahead prediction errors of the new direct approach are smaller than those obtained using the conventional direct approach. Numerical examples are included to illustrate the application of the new direct approach.     

Non-linear output frequency response functions for multi-input non-linear Volterra systems

The concept of non-linear output frequency response functions (NOFRFs) is extended to the non-linear systems that can be described by a multi-input Volterra series model. A new algorithm is also developed to determine the output frequency range of non-linear systems from the frequency range of the inputs. These results allow the concept of NOFRFs to be applied to a wide range of engineering systems. The phenomenon of the energy transfer in a two degree of freedom non-linear system is studied using the new concepts to demonstrate the significance of the new results.