A Perceptron with Optimized Backpropagation Learning Algorithm to preset a Temper mill machine: NEUROSKIN
نویسندگان
چکیده
Abstract: We present in this paper a two-years long experience of cooperation between industry and research for the resolution of a real world problem. After the presentation of the industrial context, we develop the theoretical and experimental studies and results in order to find the best architecture of neural networks to preset parameters of a temper mill machine in steel industry. We discuss on these results and compare with the results obtained by a the cascade-correlation learning architecture [2] results and the physical model results. Then we conclude on future work.
منابع مشابه
Optimization for Distributed Committee Machines in The Knowledge Discovery in Distributed Databases Process
The knowledge discovery in distributed databases represents the overall process of obtaining useful information from data that is stored in a replication topology on multiple computing systems. Distributed committee machines are formed by many neural networks that work in a distributed manner on multiple computing systems for resolving one data mining task. These architectures are the most suit...
متن کاملLoading Discriminative Feature Representations in Hidden Layer
This work explores the neural features that are trained by decreasing a discriminative energy. It directly resolves the unfaithful representation problem and the ambiguous internal representation problem in various backpropagation training algorithms for MLP. It also indirectly overcomes the premature saturation problem. Keywords—Multilayer perceptron; deep learning; Boltzmann machine; ambiguou...
متن کاملNeuron Selectivity as a Biologically Plausible Alternative to Backpropagation
In this paper, we aim to develop alternative methods to backpropagation that more closely resemble biological computation. While backpropagation has been an extremely valuable tool in machine learning applications, there is no evidence that neurons can back propagate errors. We propose two methods intended to model the intrinsic selectivity of biological neurons to certain features. Both method...
متن کاملBayesian-driven Multi-layer Perceptron Applied to Liver Fibrosis Stadialization
This paper proposes the application to the liver fibrosis stadialization of a novel training technique of feed-forward neural networks based on the Bayesian paradigm. Using the Pearson’s r correlation coefficient instead of the standard backpropagation algorithm to update the synaptic weights of a multi-layer perceptron, the proposed model is compared with traditional machine learning algorithm...
متن کاملAn Experimental Comparison of Symbolic and Connectionist Learning Algorithms
Despite the fact that many symbolic and connectionist (neural net) learning algorithms are addressing the same problem of learning from classified examples, very little Is known regarding their comparative strengths and weaknesses. This paper presents the results of experiments comparing the ID3 symbolic learning algorithm with the perceptron and back-propagation connectionist learning algorith...
متن کامل