Title :
Exact representations from feed-forward networks
Author :
Melnik, Ofer ; Pollack, Jordan
Author_Institution :
Volen Center for Complex Syst., Brandeis Univ., Waltham, MA, USA
Abstract :
We present an algorithm to extract representations from multiple hidden layer, multiple output feedforward perceptron threshold networks. The representation is based on polytopic decision regions in the input space and is exact not an approximation like most other network analysis methods. Multiple examples show some of the knowledge that can be extracted from networks by using this algorithm, including the geometrical form of artifacts and bad generalization. We compare threshold and sigmoidal networks with respect to the expressiveness of their decision regions, and also prove lower bounds for any algorithm which extracts decision regions from arbitrary neural networks
Keywords :
computational complexity; feedforward neural nets; multilayer perceptrons; bad generalization; exact representations; expressiveness; multiple hidden layer multiple output feedforward perceptron threshold networks; network analysis; polytopic decision regions; sigmoidal networks; Algorithm design and analysis; Computer networks; Data mining; Feedforward neural networks; Feedforward systems; Multilayer perceptrons; Neural networks; Partitioning algorithms; Sensitivity analysis; Transfer functions;
Conference_Titel :
Neural Networks, 2000. IJCNN 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on
Conference_Location :
Como
Print_ISBN :
0-7695-0619-4
DOI :
10.1109/IJCNN.2000.861350