Learning Algorithms for Neural-Net Decision Support
TL;DR: The performance of the classical back-propagation algorithm for training artificial neural networks can be improved by applying modified approaches which have better convergence properties, and are faster than the commonly used steepest gradient search method.
Abstract: Artificial Neural Networks present a new paradigm for decision support that is adaptive and capable of integrating knowledge acquisition, problem solving, and learning. In this paper, we show that the performance of the classical back-propagation algorithm for training artificial neural networks can be improved by applying modified approaches which have better convergence properties, and are faster than the commonly used steepest gradient search method. Simulation results on test problems show that the proposed algorithms perform better than the steepest gradient search algorithm in terms of faster convergence with learning accuracy at least as good as in classical back propagation. Inductive learning algorithms (ID3 and NEWQ) and Probit are also used to compare the relative performances of connectionist, inductive learning and statistical procedures. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.
Citations
More filters
TL;DR: This approach to drug discovery is illustrated by reviewing target sites for existing antibiotics and considering how this information might be applied for the discovery of new agents inhibiting peptidoglycan synthesis, tRNA synthesis, transcription and DNA replication.
Abstract: The introduction of antibiotics for the chemotherapy of bacterial infections has been one of the most important medical achievements of the past 50 years However, the emergence of bacterial resistance to antibiotics undermines the therapeutic utility of existing agents, creating a requirement for the discovery of new antibacterial drugs Several drug discovery strategies have emerged, including incremental improvements to existing antibiotics by chemical manipulation and the search for novel drug targets based on genomic approaches An alternative strategy seeks to exploit opportunities for drug discovery arising from an understanding of the mode of action of existing antibiotics Thus biochemical pathways or processes inhibited by antibiotics already in clinical use may nevertheless contain key functions that represent unexploited targets for further drug discovery A major benefit of employing pathways or processes that are already known to contain drug targets is that proof of principle for drug intervention is already established This approach to drug discovery is illustrated by reviewing target sites for existing antibiotics and considering how this information might be applied for the discovery of new agents inhibiting peptidoglycan synthesis, tRNA synthesis, transcription and DNA replication
125 citations
TL;DR: In this paper, the authors consider practical aspects of building and validating neural network models of manufacturing processes, and illustrate the recommended approaches with two diverse case studies with two case studies.
Abstract: Neural networks are beginning to be used for the modelling of complex manufacturing processes, usually for process and quality control. Often these models are used to identify optimal process settings. Since a neural network is an empirical model, it is highly dependent on the data used in construction and validation. Using data directly from production ensures availability and fidelity, however, the samples may not reflect the entire range of probable operation and, in particular, may not include the optimal process settings. Supplementing production data with observations gathered from designed experiments alleviates the problem of overly focused or incomplete production data sets. This paper considers practical aspects of building and validating neural network models of manufacturing processes, and illustrates the recommended approaches with two diverse case studies.
123 citations
Cites background from "Learning Algorithms for Neural-Net ..."
...There have also been many adjustments to the learning process of supervised neural networks to improve the models in certain circumstances, such as iterative training (Sun et al. 1995) or backpropagation speed-ups (Piramuthu et al. 1993)....
[...]
...This provided the motivation for the project to develop a predictive model of the slip casting process which would allow the manufacturer to optimize controllable process parameters without wasteful and time consuming test casts....
[...]
TL;DR: The Molekulmasse (Mr) and the Lipophilie (Lipschitz) as mentioned in this paper can be found in a Leitverbindungen geringer Wirksamkeit zu Wirkstoffen.
Abstract: Die Optimierung von Leitverbindungen geringer Wirksamkeit zu Wirkstoffen geht oft mit einer Zunahme der Molekulmasse (Mr) und der Lipophilie einher. Treffer mit Affinitaten auf μM-Niveau, entdeckt durch das Durchmustern von Leitstruktur-Bibliotheken, geben diesem Optimierungsprozes Spielraum, wie im Diagramm anhand der Verteilungen von Mr fur eine Leitstruktur-Bibliothek (1), bekannte oral wirksame Arzneien (2) und eine typische kombinatorische Bibliothek (3) gezeigt ist. y=Prozentanteil mit einer bestimmten Molekulmasse.
23 citations
Posted Content•
TL;DR: Optimization and Stochastic Processes Applied to Economy and Finance -- is the name of this book translated to English; It has been used at the IME-USP - The Institute of Mathematics and Statistics of the University of Sao Paulo, since 1993.
Abstract: Optimization and Stochastic Processes Applied to Economy and Finance -- is the name of this book translated to English; It has been used at the IME-USP - The Institute of Mathematics and Statistics of the University of Sao Paulo, since 1993.
Contents: Ch.1: Linear Programming; Ch.2: Non-Linear Programming; Ch.3: Quadratic Programming; Ch.4: Markowitz Model; Ch.5: Dynamic Programming; Ch.6: LQG Estimation and Control; Ch.7: Decision Trees; Ch.8: Pension Funds; Ch.9: Mixed Portfolios Including Derivative Contracts; Appendices: App.A: Matlab; App.B: Critical-Point Software; App.C: Computational Linear Algebra; App.D: Probability; App.E: Computer Codes.
This book is written in Portuguese language.
11 citations
TL;DR: This study applies neural network modelling, using as inputs established descriptors of physiologic and anatomic injury severity, and indicates that neural networks show significant promise for trauma patient outcome evaluation.
5 citations