Perturbations, Accuracy and Robustness in Neural Networks

2005 ◽  
pp. 2282-2287
Author(s):  
Cesare Alippi ◽  
Giovanni Vanini
Keyword(s):  
Neural Networks ◽  
Robustness Analysis ◽  
Theoretical Aspect ◽  
Neural Model ◽  
Knowledge Space

A robustness analysis for neural networks, namely the evaluation of the effects induced by perturbations affecting the network weights, is a relevant theoretical aspect since weights characterise the “knowledge space” of the neural model and, hence, its inner nature.

Robustness in Neural Networks

2011 ◽  
pp. 3314-3321
Author(s):  
Cesare Alippi ◽  
Manuel Roveri ◽  
Giovanni Vanini
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Wireless Sensor Networks ◽  
Robustness Analysis ◽  
Neural Model ◽  
Point Of View ◽  
Wireless Sensor ◽  
Knowledge Space ◽  
Finite Precision ◽  
Free Parameters

The robustness analysis for neural networks aims at evaluating the influence on accuracy induced by perturbations affecting the computational flow; as such it allows the designer for estimating the resilience of the neural model w.r.t perturbations. In the literature, the robustness analysis of neural networks generally focuses on the effects of perturbations affecting biases and weights. The study of the network’s parameters is relevant both from the theoretical and the application point of view, since free parameters characterize the “knowledge space” of the neural model and, hence, its intrinsic functionality. A robustness analysis must also be taken into account when implementing a neural network (or the intelligent computational system into which a neural network is inserted) in a physical device or in intelligent wireless sensor networks. In these contexts, perturbations affecting the weights of a neural network abstract uncertainties such as finite precision representations, fluctuations of the parameters representing the weights in analog solutions (e.g., associated with the production process of a physical component), ageing effects or more complex, and subtle uncertainties in mixed implementations.

A Perturbation Size-Independent Analysis of Robustness in Neural Networks by Randomized Algorithms

2011 ◽  
pp. 22-40 ◽  
Author(s):  
C. Alippi
Keyword(s):  
Neural Network ◽  
Randomized Algorithms ◽  
Robustness Analysis ◽  
Neural Model ◽  
Neural Structure ◽  
Knowledge Space ◽  
Related Literature ◽  
Independent Analysis ◽  
Intractable Problem ◽  
Perturbation Size

This chapter presents a general methodology for evaluating the loss in performance of a generic neural network once its weights are affected by perturbations. Since weights represent the “knowledge space” of the neural model, the robustness analysis can be used to study the weights/performance relationship. The perturbation analysis, which is closely related to sensitivity issues, relaxes all assumptions made in the related literature, such as the small perturbation hypothesis, specific requirements on the distribution of perturbations and neural variables, the number of hidden units and a given neural structure. The methodology, based on Randomized Algorithms, allows reformulating the computationally intractable problem of robustness/sensitivity analysis in a probabilistic framework characterised by a polynomial time solution in the accuracy and confidence degrees.

The TP Chromatic Aberration Estimation by Using Neural Networks

2011 ◽  
Vol 121-126 ◽  
pp. 4239-4243 ◽  
Author(s):  
Du Jou Huang ◽  
Yu Ju Chen ◽  
Huang Chu Huang ◽  
Yu An Lin ◽  
Rey Chue Hwang
Keyword(s):  
Neural Networks ◽  
Learning Algorithm ◽  
Back Propagation ◽  
Physical Property ◽  
Chromatic Aberration ◽  
Neural Model ◽  
Artificial Intelligent ◽  
Error Back Propagation ◽  
The Neural Networks ◽  
Simulation Results

The chromatic aberration estimations of touch panel (TP) film by using neural networks are presented in this paper. The neural networks with error back-propagation (BP) learning algorithm were used to catch the complex relationship between the chromatic aberration, i.e., L.A.B. values, and the relative parameters of TP decoration film. An artificial intelligent (AI) estimator based on neural model for the estimation of physical property of TP film is expected to be developed. From the simulation results shown, the estimations of chromatic aberration of TP film are very accurate. In other words, such an AI estimator is quite promising and potential in commercial using.

scholarly journals Semiotic Aggregation in Deep Learning

Entropy ◽  
2020 ◽  
Vol 22 (12) ◽  
pp. 1365
Author(s):  
Bogdan Muşat ◽  
Răzvan Andonie
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Decision Model ◽  
Deep Neural Networks ◽  
Neural Model ◽  
Network Layers ◽  
Saliency Maps ◽  
Spatial Entropy ◽  
Insight Into

Convolutional neural networks utilize a hierarchy of neural network layers. The statistical aspects of information concentration in successive layers can bring an insight into the feature abstraction process. We analyze the saliency maps of these layers from the perspective of semiotics, also known as the study of signs and sign-using behavior. In computational semiotics, this aggregation operation (known as superization) is accompanied by a decrease of spatial entropy: signs are aggregated into supersign. Using spatial entropy, we compute the information content of the saliency maps and study the superization processes which take place between successive layers of the network. In our experiments, we visualize the superization process and show how the obtained knowledge can be used to explain the neural decision model. In addition, we attempt to optimize the architecture of the neural model employing a semiotic greedy technique. To the extent of our knowledge, this is the first application of computational semiotics in the analysis and interpretation of deep neural networks.

scholarly journals Prediction of Slot Shape and Slot Size for Improving the Performance of Microstrip Antennas Using Knowledge-Based Neural Networks

2014 ◽  
Vol 2014 ◽  
pp. 1-9 ◽  
Author(s):  
Taimoor Khan ◽  
Asok De
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Prior Knowledge ◽  
Neural Model ◽  
Microstrip Antennas ◽  
Performance Parameters ◽  
Dual Frequency ◽  
Knowledge Based ◽  
Slot Size ◽  
Good Agreement

In the last decade, artificial neural networks have become very popular techniques for computing different performance parameters of microstrip antennas. The proposed work illustrates a knowledge-based neural networks model for predicting the appropriate shape and accurate size of the slot introduced on the radiating patch for achieving desired level of resonance, gain, directivity, antenna efficiency, and radiation efficiency for dual-frequency operation. By incorporating prior knowledge in neural model, the number of required training patterns is drastically reduced. Further, the neural model incorporated with prior knowledge can be used for predicting response in extrapolation region beyond the training patterns region. For validation, a prototype is also fabricated and its performance parameters are measured. A very good agreement is attained between measured, simulated, and predicted results.

MREM, Discrete Recurrent Network for Optimization

2011 ◽  
pp. 1112-1120
Author(s):  
Enrique Mérida-Casermeiro ◽  
Domingo López-Rodríguez ◽  
Juan M. Ortiz-de-Lazcano-Lobato
Keyword(s):  
Neural Networks ◽  
Combinatorial Optimization ◽  
Image Recognition ◽  
Single Neuron ◽  
Optimization Problems ◽  
Neural Model ◽  
Recurrent Network ◽  
Neural Models ◽  
Seminal Work ◽  
Combinatorial Optimization Problems

Since McCulloch and Pitts’ seminal work (McCulloch & Pitts, 1943), several models of discrete neural networks have been proposed, many of them presenting the ability of assigning a discrete value (other than unipolar or bipolar) to the output of a single neuron. These models have focused on a wide variety of applications. One of the most important models was developed by J. Hopfield in (Hopfield, 1982), which has been successfully applied in fields such as pattern and image recognition and reconstruction (Sun et al., 1995), design of analogdigital circuits (Tank & Hopfield, 1986), and, above all, in combinatorial optimization (Hopfield & Tank, 1985) (Takefuji, 1992) (Takefuji & Wang, 1996), among others. The purpose of this work is to review some applications of multivalued neural models to combinatorial optimization problems, focusing specifically on the neural model MREM, since it includes many of the multivalued models in the specialized literature.

scholarly journals To contemplate quantitative and qualitative water features by neural networks method

2011 ◽  
Vol 48 (No. 7) ◽  
pp. 322-326 ◽  
Author(s):  
M. Neruda ◽  
R. Neruda
Keyword(s):  
Neural Networks ◽  
Fourier Series ◽  
Computer Science ◽  
Hybrid System ◽  
Back Propagation ◽  
Neural Model ◽  
Black Box ◽  
Layer Number ◽  
Accurate Forecast ◽  
Hidden Layer

An application deals with calibration of neural model and Fourier series model for Ploučnice catchment. This approach has an advantage, that the network choice is independent of other example’s parameters. Each networks, and their variants (different units and hidden layer number) can be connected in as a black box and tested independently. A Stuttgart neural simulator SNNS and a multiagent hybrid system Bang2 developed in Institute of Computer Science, AS CR have been used for testing. A perceptron network has been constructed, which was trained by back propagation method improved with a momentum term. The network is capable of an accurate forecast of the next day runoff based on the runoff and rainfall values from previous day.

Sign in / Sign up

Export Citation Format

Share Document