Neural Network Studies. 3. Variable Selection in the Cascade-Correlation Learning Architecture

Pruning methods for feed-forward artificial neural networks trained by the cascade-correlation learning algorithm are proposed. The cascade-correlation algorithm starts with a small network and dynamically adds new nodes until the analyzed problem has been solved. This feature of the algorithm removes the requirement to predefine the architecture of the neural network prior to network training. The developed pruning methods are used to estimate the importance of large sets of initial variables for quantitative structure−activity relationship studies and simulated data sets. The calculated results are compared with the performance of fixed-size back-propagation neural networks and multiple regression analysis and are carefully validated using different training/test set protocols, such as leave-one-out and full cross-validation procedures. The results suggest that the pruning methods can be successfully used to optimize the set of variables for the cascade-correlation learning algorithm neural networks. Th...

[1]  D. Maddalena,et al.  Prediction of receptor properties and binding affinity of ligands to benzodiazepine/GABAA receptors using artificial neural networks. , 1995, Journal of medicinal chemistry.

[2]  Igor V. Tetko,et al.  Application of a Pruning Algorithm To Optimize Artificial Neural Networks for Pharmaceutical Fingerprinting , 1998, J. Chem. Inf. Comput. Sci..

[3]  Robert J. Marks,et al.  Neurosmithing: improving neural network learning , 1998 .

[4]  Russell Reed,et al.  Pruning algorithms-a survey , 1993, IEEE Trans. Neural Networks.

[5]  Desmond J. Maddalena.  Review Biologicals & Immunologicals: Applications of artificial neural networks to quantitative structure-activity relationships , 1996 .

[6]  Igor V. Tetko,et al.  Efficient Partition of Learning Data Sets for Neural Network Training , 1997, Neural Networks.

[7]  M. Hosseini,et al.  Using Artificial Neural Networks To Classify the Activity of Capsaicin and Its Analogues , 1997, J. Chem. Inf. Comput. Sci..

[8]  Igor V. Tetko,et al.  Data modelling with neural networks: Advantages and limitations , 1997, J. Comput. Aided Mol. Des..

[9]  Igor V. Tetko,et al.  Neural network studies, 1. Comparison of overfitting and overtraining , 1995, J. Chem. Inf. Comput. Sci..

[10]  S. V. Antonenko,et al.  HIV-1 reverse transcriptase inhibitor design using artificial neural networks. , 1994, Journal of medicinal chemistry.

[11]  Yoshifusa Ito,et al.  Approximation Capability of Layered Neural Networks with Sigmoid Units on Two Layers , 1994, Neural Computation.

[12]  Igor V. Tetko,et al.  Neural Network Studies, 2. Variable Selection , 1996, J. Chem. Inf. Comput. Sci..

[13]  Carl E. Rasmussen,et al.  Pruning from Adaptive Regularization , 1994, Neural Computation.

[14]  Elie Bienenstock,et al.  Neural Networks and the Bias/Variance Dilemma , 1992, Neural Computation.

[15]  D. Livingstone,et al.  Structure-activity relationships of antifilarial antimycin analogues: a multivariate pattern recognition study. , 1990, Journal of medicinal chemistry.

[16]  P. Petersen,et al.  Synthesis and structure-activity relationships of new 7-[3-(fluoromethyl)piperazinyl]- and -(fluorohomopiperazinyl)quinolone antibacterials. , 1990, Journal of medicinal chemistry.