Fuzzy multi-layer perceptron, inferencing and rule generation

Mitra, Sushmita; Pal, Sankar K.

doi:10.1109/72.363450

Cited by 167 publications

(61 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Substituting (10) in (1), we have (11) where is the output of the th output node and is that of the hidden node, corresponding to the presented interval, connected to the th output node. From (11) we have (12) Similarly, considering the complement-interval of the feature , we can write (13) Therefore (14) The outputs and are calculated using (1) with appropriate input values.…”

Section: B An Examplementioning

confidence: 99%

See 1 more Smart Citation

Knowledge-based fuzzy MLP for classification and rule generation

Pal

1997

IEEE Trans. Neural Netw.

Self Cite

View full text Add to dashboard Cite

Abstract-A new scheme of knowledge-based classification and rule generation using a fuzzy multilayer perceptron (MLP) is proposed. Knowledge collected from a data set is initially encoded among the connection weights in terms of class a priori probabilities. This encoding also includes incorporation of hidden nodes corresponding to both the pattern classes and their complementary regions. The network architecture, in terms of both links and nodes, is then refined during training. Node growing and link pruning are also resorted to. Rules are generated from the trained network using the input, output, and connection weights in order to justify any decision(s) reached. Negative rules corresponding to a pattern not belonging to a class can also be obtained. These are useful for inferencing in ambiguous cases. Results on real life and synthetic data demonstrate that the speed of learning and classification performance of the proposed scheme are better than that obtained with the fuzzy and conventional versions of the MLP (involving no initial knowledge encoding). Both convex and concave decision regions are considered in the process.

show abstract

Section: B An Examplementioning

confidence: 99%

“…This is found to be useful for inferencing in ambiguous cases. Note that, the rule generation procedures described in this article are different from that reported in [11]. The model is capable of handling input in numerical, linguistic, and set forms, and can tackle uncertainty due to overlapping classes.…”

mentioning

confidence: 99%

Knowledge-based fuzzy MLP for classification and rule generation

Pal

1997

IEEE Trans. Neural Netw.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Summarizing, this module is realized using the secondorder RNN given by (39) and (40). It is straightforward to verify that if I 1 (τ ) = I 2 (τ ), for all τ = 1, 2, .…”

Section: −L(t)mentioning

confidence: 99%

“…Two examples of such a synergy are the adaptive network-based fuzzy inference system (ANFIS) [22], which is a feedforward network representation of the fuzzy reasoning process (see [34] for an extension to RNNs), and the fuzzy-MLP, which is a feedforward network with fuzzified inputs [40], [46]. Neuro-fuzzy systems have been intensively discussed in the literature (see the survey [39]), but rarely do they contain feedback connections [23].…”

Section: Introductionmentioning

confidence: 99%

A New Approach to Knowledge-Based Design of Recurrent Neural Networks

Kolman

Margaliot

2008

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Abstract-A major drawback of artificial neural networks (ANNs) is their black-box character. This is especially true for recurrent neural networks (RNNs) because of their intricate feedback connections. In particular, given a problem and some initial information concerning its solution, it is not at all clear how to design an RNN that is suitable for solving this problem.In this paper, we consider a fuzzy rule-base with a special structure, referred to as the fuzzy all-permutations rulebase (FARB). Inferring the FARB yields an input-output mapping that is mathematically equivalent to that of an RNN.We use this equivalence to develop two new knowledge-based design methods for RNNs. The first method, referred to as the direct approach, is based on stating the desired functioning of the RNN in terms of several sets of symbolic rules, each one corresponding to a sub-network. Each set is then transformed into a suitable FARB.The second method is based on first using the direct approach to design a library of simple modules, such as counters or comparators, and realize them using RNNs. Once designed, the correctness of each RNN can be verified. Then, the initial design problem is solved by using these basic modules as building blocks. This yields a modular and systematic approach for knowledge-based design of RNNs. We demonstrate the efficiency of these approaches by designing RNNs that recognize both regular and non-regular formal languages.

show abstract

“…This all leads to the necessity of having adaptive fuzzy systems [6,7,8]. The intent is to patch the defects in the definition of the membership functions, or sometimes even to generate the rules of the membership functions from scratch in the learning process.…”

Section: Introductionmentioning

confidence: 99%