265,850 Hits in 3.3 sec

Neural Computation in Medicine: Perspectives and Prospects [chapter]

Richard Dybowski
2000 Artificial Neural Networks in Medicine and Biology  
In 1998, over 400 papers on artificial neural networks (ANNs) were published in the context of medicine, but why is there this interest in ANNs?  ...  Finally, we briefly look at two directions in which ANNs are likely to develop, namely the use of Bayesian statistics and knowledgedata fusion.  ...  Bayesian ANNs have not yet found their way into general use, but, given their capabilities, we expect them to take a prominent role in medical neural computation.  ... 
doi:10.1007/978-1-4471-0513-8_4 dblp:conf/annimab/Dybowski00 fatcat:lrtzpnhjmzf4fequfmqd4sswqi

Confusions about Consciousness [chapter]

D. Papineau
1998 Perspectives in Neural Computing  
'How are conscious states "generated" by neural activity?'  ...  bringing in any extra forces operating only in living bodies.  ... 
doi:10.1007/978-1-4471-3427-5_10 fatcat:7ddxa4o53rerdatjoyuj5f7xia

Learning in Artificial Neural Networks: A Statistical Perspective

Halbert White
1989 Neural Computation  
Learning in Artificial Neural Networks 457 Theorem 1.  ...  Generalizing the PAC model for neural net and other learning applications. UCSC Computer Research Laboratory Tech. Rep. UCSC-CRL- 89-30. Hecht-Nielsen, R. 1989.  ... 
doi:10.1162/neco.1989.1.4.425 fatcat:3qwhkn2345bnvbofx44qpjj6b4

Entropy Estimation in Turing's Perspective

Zhiyi Zhang
2012 Neural Computation  
The fundamental switch in perspective brings about substantial gain in estimation accuracy for every distribution with finite entropy.  ...  In general, a uniform variance upper bound is established for the entire class of distributions with finite entropy that decays at a rate of O(ln(n)/n) compared to O([ln(n)] 2 /n) for the plug-in; in a  ...  different perspective in estimating entropy.  ... 
doi:10.1162/neco_a_00266 pmid:22295985 fatcat:bf2c4r3tgvh5flk6pqz7k2ti7m

Neural Learning and Weight Flow on Stiefel Manifold [chapter]

Simone Fiori, Aurelio Uncini, Francesco Piazza
1999 Perspectives in Neural Computing  
The aim of this paper is to present a new class of learning models for linear as well as non-linear neural layers called Orthonormal Strongly-Constrained SOC or Stiefel.  ...  In a ONP, the target of the adaptation rule for a neural network is to learn an orthonormal weight matrix related in some a way to the input signal.  ...  The Deco-Brauer's theory In 2 Deco and Brauer introduced the concept of volume-conserving neural networks.  ... 
doi:10.1007/978-1-4471-0811-5_36 fatcat:6i42rg3vorbnpdqfs3m5m6zreu

Introduction to the Financial Markets [chapter]

John G. Taylor
2002 Perspectives in Neural Computing  
Thus if we wish to employ economic data for which we have information, then only a monthly approach can be used. 6 Neural Networks and the Financial Markets 1985 1986 1987 1988 1989 1990 1991 1992 1993  ...  This changes demand and confidence in a stock or in the stock market as a whole, and thereby also causes changes in demand in bonds (usually regarded as a safe haven compared to riskier stocks).  ...  It will thus cast light across a number of areas, each of which has been developed to a level of expertise which appears appropriate to attempt to describe in a connected form.  ... 
doi:10.1007/978-1-4471-0151-2_1 fatcat:sms6tqsya5hnhhjdwbg4o4wadi

A Generic Neural Network Approach For Constraint Satisfaction Problems [chapter]

E. P. K. Tsang, C. J. Wang
1992 Perspectives in Neural Computing  
This paper describes a neural network approach for solving CSPs which aims at providing prompt responses.  ...  Since the neural network model lends itself to the VLSI implementation of parallel processing architectures, the limited number of cycles required by GENET to find the solutions for the tested problems  ...  Reynold and Kar-Lik Wong in this research. Jenny Emby has greatly improved the readability of this paper.  ... 
doi:10.1007/978-1-4471-2003-2_2 fatcat:ejh53jd3hjeq5ak36iyv3xuiwq

A Unified Approach to Sequential Constructive Methods [chapter]

Marco Muselli
1999 Perspectives in Neural Computing  
In most cases this leads to a large reduction of the computational cost.  ...  A general treatment of a particular class of learning techniques for neural networks, called sequential constructive methods, is proposed.  ...  the samples contained in a given training set S.  ... 
doi:10.1007/978-1-4471-0811-5_41 fatcat:zdsg4fwnx5fqzdoq5m7ffw2kre

A Comparative Evaluation of Sequential Constructive Methods [chapter]

Marco Muselli
1999 Perspectives in Neural Computing  
In particular, the neural networks trained show good generalization ability also when the learning time is very low.  ...  Four of them incrementally build the hidden layer of the resulting neural network by subsequently adding threshold units.  ...  Furthermore, the low computational cost of HC allows the reduction of the total CPU time in all the three tests.  ... 
doi:10.1007/978-1-4471-0811-5_40 fatcat:hgb5bs3irfdo7cpfqykf7n5lj4

Astronomical Object Recognition by means of Neural Networks [chapter]

R. Tagliaferri, G. Longo, S. Andreon, S. Zaggia, N. Capuano, G. Gargiulo
1999 Perspectives in Neural Computing  
Preliminary tests show that neural nets are more effective than traditional techniques.  ...  We discuss here the application of different types of neural nets to the detection and extraction of celestial objects.  ...  In this way we use single layer NNs in the computation on the test set. Experimental results The experiments were performed on a 1° by 1° degree extracted from one of the plates.  ... 
doi:10.1007/978-1-4471-0811-5_15 fatcat:t3ixll2t4bcbld5d7md55mzeuq

Training Data Selection for Optimal Generalization with Noise Variance Reduction in Neural Networks [chapter]

Sethu Vijayakumar, Masashi Sugiyama, Hidemitsu Ogawa
1999 Perspectives in Neural Computing  
We formalize the learning problem in neural networks as an inverse problem using a functional analytic framework and use the Averaged Projection criterion as our optimization criterion for learning.  ...  In this paper, we discuss the problem of active training data selection in the presence of noise.  ...  The generalization ability of a learning system in a noisy environment is a delicate balance on how well it can select data to enlarge the approximation space and at the same time, reduce noise variance  ... 
doi:10.1007/978-1-4471-0811-5_14 fatcat:phi2ypixkzd75awvtlqioxci5e

Neural networks in an artificial life perspective [chapter]

Stefano Nolfi, Domenico Parisi
1997 Lecture Notes in Computer Science  
Explicitly viewing Neural Networks in an Artificial Life perspective has a number of consequences that make research on what we will call Artificial Life Neural Networks (ALNNs) rather different from traditional  ...  In the last few years several researchers within the Artificial Life and Mobile Robotics community used Artificial Neural Networks.  ...  Much work on neural networks views them as abstract computational devices or information processing machines which, by adopting a brain style of computation, are capable of input/output mappings (behaviors  ... 
doi:10.1007/bfb0020241 fatcat:xkm4yqlvyrcjtfxujboxtspl54

A Connectionist Approach to Spatial Memory and Planning [chapter]

G. Bugmann
1998 Perspectives in Neural Computing  
The third one is the most promising theory of SM in humans and potentially allows for the most exible planning at the lowest computational and memory cost in robot implementations.  ...  The sparse neural network model described here is characterized by: i) the use of a resistive-grid paradigm for backward planning, transforming the view-graph into a dynamic value-map ii) the use of a  ...  Taylor, Chris Hindle and the Neural and Adaptive Systems Group have a l l o wed to set the scene for this investigation. Many thanks to Alan Simpson for his support.  ... 
doi:10.1007/978-1-4471-3427-5_5 fatcat:gc42ys6wa5bajnekd4oqgma3gu

Influence of extracellular oscillations on neural communication: a computational perspective

Zoran Tiganj, Sylvain Chevallier, Eric Monacelli
2014 Frontiers in Computational Neuroscience  
Neural communication generates oscillations of electric potential in the extracellular medium.  ...  It is unclear whether this influence should be considered only as noise or it has some functional role in neural communication.  ...  The models are implemented in neural simulator NEURON (Hines and Carnevale, 1997 ) that we run on a Linux-based cluster, consisting of 75 dual-core computers.  ... 
doi:10.3389/fncom.2014.00009 pmid:24570661 pmcid:PMC3916728 fatcat:m75hhrfw5fc6rnwfpndoama34m

Computation in Neuromorphic Analog VLSI Systems [chapter]

Giacomo Indiveri
2002 Perspectives in Neural Computing  
implement the physical processes that underlie neural computation.  ...  In recent times the term "neuromorphic" has also been used to describe mixed analog/digital VLSI systems that implement computational models of real neural systems.  ...  Conclusion During the last decade CMOS aVLSI has been used to construct a wide range of neural analogs, from single synapses to sensory arrays, and simple systems.  ... 
doi:10.1007/978-1-4471-0219-9_1 dblp:conf/wirn/Indiveri01 fatcat:b3rfkeuhwbeujirov45u5t243i
« Previous Showing results 1 — 15 out of 265,850 results