- Research Article
4
- 10.1162/artl_a_00364
Making Artificial Brains: Components, Topology, and Optimization.
- May 12, 2022
- Artificial life
- Christoph Adami
Making Artificial Brains: Components, Topology, and Optimization.
Two Artificial Neural Network (ANN) reconstruction methods for Electrical Impedance Tomography (EIT) have been presented in this paper. The problem under study concerns the reconstruction of the conductivity distribution inside the investigated area, using the information collected from the boundary. The first approach consists in ANN learning using electrical potential vectors, which were obtained from numerical solution of the forward problems. The second method using a standard feed-forward multilayered neural networks, applies the circuit representation for the finite clement discretization. Using the quadrilateral finite element, the neural network structure for EIT problem has been proposed. The advantages and disadvantages both methods with respect to classical approach are discussed in detail.
Making Artificial Brains: Components, Topology, and Optimization.
Making Artificial Brains: Components, Topology, and Optimization.
Artificial neural networks and instructional technology
AFTER MORE THAN TWO DECADES OF NEGLECT, artificial neural networks (ANNs) are emerging as a strong contender in artificial intelligence. However, researchers whose leanings are more toward expert systems or other types of symbolic formalisms for knowledge representations warn against overly enthusiastic claims depicting ANNs as replicating the process of human thinking. While some of these objections are justified, the technology warrants a closer look by designers of educational software. Though neural networks are usually considered as a part of artificial intelligence, they are very different from the AI formalisms to which we’re accustomed. Expert systems use if-then rules to arrive at an answer. These rules are formulated by a knowledge engineer who works closely with a content expert to derive a model of the problem space. Neural networks, however, are fed sample cases (called “training sets≓), learn how to recognize patterns in the sample data, and then use this experience in handling new cases. Because neural nets are trained rather than programmed, the only overhead — other than the initial design — is in collecting and formatting the data sets (Carlson, 1990b, p. 50). This paper first describes the essentials of how an ANN learns, using a classic application in pattern recognition to demonstrate an operating ANN. Of the several “cognitive≓ roles ANNs are suitable for, two seem especially appropriate for computer-mediated education: (1) intelligent filters and (2) spreading, associative memories. To illustrate each, I describe prototypes I’m working on at the Air Force Armstrong Laboratory.
Read moreA novel neural network model with distributed evolutionary approach for big data classification
The considerable improvement of technology produced for various applications has resulted in a growth in data sizes, such as healthcare data, which is renowned for having a large number of variables and data samples. Artificial neural networks (ANN) have demonstrated adaptability and effectiveness in classification, regression, and function approximation tasks. ANN is used extensively in function approximation, prediction, and classification. Irrespective of the task, ANN learns from the data by adjusting the edge weights to minimize the error between the actual and predicted values. Back Propagation is the most frequent learning technique that is used to learn the weights of ANN. However, this approach is prone to the problem of sluggish convergence, which is especially problematic in the case of Big Data. In this paper, we propose a Distributed Genetic Algorithm based ANN Learning Algorithm for addressing challenges associated with ANN learning for Big data. Genetic Algorithm is one of the well-utilized bio-inspired combinatorial optimization methods. Also, it is possible to parallelize it at multiple stages, and this may be done in an extremely effective manner for the distributed learning process. The proposed model is tested with various datasets to evaluate its realizability and efficiency. The results obtained from the experiments show that after a specific volume of data, the proposed learning method outperformed the traditional methods in terms of convergence time and accuracy. The proposed model outperformed the traditional model by almost 80% improvement in computational time.
Read moreHydrodynamic inline force prediction on vertical cylinders: a comparative study of neural network and its adaptive wavelets (wavenets)
The design of the neural network model and its adaptive wavelets (wavelet networks and wavenets) was used to estimate the wave-induced hydrodynamic inline force acting on a vertical cylinder. The data used to calibrate and validate the models were obtained from an experiment. In the brain, wavelet neural networks (WNNs) use wavelets to activate their hidden layers of neurons. In WNNs, both the position and dilation of the wavelets are optimized along with the weights. In one special approach to this kind of network construction, the position and dilation of the wavelets are fixed and only the weights of the network are optimized. In the present study, the neural network procedure and the above mentioned approach were employed to design a WNN, a so-called wavenet, using feed-forward neural network topology and its training method. Then, a comparison of these two methods was made. Numerical results demonstrate that both networks are capable of predicting hydrodynamic inline force. Furthermore, the combination of the neural network concept and the wavelet theory i.e. wavenet provides a more robust tool rather than standard feed-forward neural network, considering its more appropriate ability to predict any other data which the network had not experienced before. The results of this study can contribute to reducing the errors in future efforts to predict hydrodynamic inline force using WNNs, and thus improve the reliability of that prediction in comparison to the ANN and other methods. Therefore, this method can be applied to relevant engineering projects with satisfactory results.
Read morePrediction of Missing Hydro-Meteorological Data Series Using Artificial Neural Networks (ANN) for Upper Tana River Basin, Kenya
Accurate prediction of missing hydro-meteorological data is crucial in planning, design, development and management of water resources systems. In the present research, prediction of such data using Artificial Neural Networks (ANN) based on temporal and spatial auto-correlation has been conducted for upper Tana River basin in Kenya. Different ANN models were formulated using a combination of numerous data delays in the ANN input layer. The findings show that the best models comprise of a feed-forward neural network trained on Levenberg-Marquardt algorithm with single hidden layer. Additionally, the best ANN architecture model for predicting missing stream flow data was at gauge station 4CC03 with correlation coefficient and MSE of 0732 and 0.242 respectively during validation. Temporal auto-correlation of the observed and the predicted stream flow values were evaluated using a correlation coefficient R that resulted to highest value of 0.756 at gauge station 4AB05. The best ANN model for prediction of missing precipitation data was at station 9037112 with R value of 0.970. In both cases the best performance was at epochs 9 and 20 respectively. The spatial auto-correlation show that the best ANN architecture model for prediction of missing stream flow data was at gauge station 4CC03 with R value of 0.723, while the one for precipitation was at station 9037096 with R value of 0.712 during the validation. The results indicate that the spatial auto-correlation of hydro-meteorological data using ANN is better than the temporal auto-correlation in the data prediction in upper Tana River basin.
Read moreEvaluation of Methods for On-line Image Analysis of Electrical Impedance Tomography Images.
This paper describes the manner in which EIT (Electrical Impedance Tomography) was used to online estimate separation efficiency in solid/liquid separators. Although off-line image analysis can be used to determine the characteristics of the internal flow, it is necessary to develop a technique for on-line control of the separation process. Three different methods were evaluated to determine a suitable strategy for on-line image analysis of electrical impedance tomography data: (i) Use of back- propagation artificial neural networks (ANN), (ii) air core scaling and (iii) use of radial basis functions and detection of abrupt changes. One case study is presented: A 44 mm internal-body diameter hydrocyclone pilot-scale plant.
Read moreBidirectional Optical Neural Networks Based on Free-Space Optics Using Lens Arrays and Spatial Light Modulator.
This paper introduces a novel architecture-bidirectional optical neural network (BONN)-for providing backward connections alongside forward connections in artificial neural networks (ANNs). BONN incorporates laser diodes and photodiodes and exploits the properties of Köhler illumination to establish optical channels for backward directions. Thus, it has bidirectional functionality that is crucial for algorithms such as the backpropagation algorithm. BONN has a scaling limit of 96 × 96 for input and output arrays, and a throughput of 8.5 × 1015 MAC/s. While BONN's throughput may rise with additional layers for continuous input, limitations emerge in the backpropagation algorithm, as its throughput does not scale with layer count. The successful BONN-based implementation of the backpropagation algorithm requires the development of a fast spatial light modulator to accommodate frequent data flow changes. A two-mirror-like BONN and its cascaded extension are alternatives for multilayer emulation, and they help save hardware space and increase the parallel throughput for inference. An investigation into the application of the clustering technique to BONN revealed its potential to help overcome scaling limits and to provide full interconnections for backward directions between doubled input and output ports. BONN's bidirectional nature holds promise for enhancing supervised learning in ANNs and increasing hardware compactness.
Read morePENENTUAN PARAMETER LEARNING RATE SELAMA PEMBELAJARAN JARINGAN SYARAF TIRUAN BACKPROPAGATION MENGGUNAKAN ALGORITMA GENETIKA

 
 
 
 One of the weakness in backpropagation Artificial neural network(ANN) is being stuck in local minima. Learning rate parameter is an important parameter in order to determine how fast the ANN Learning. This research is conducted to determine a method of finding the value of learning rate parameter using a genetic algorithm when neural network learning stops and the error value is not reached the stopping criteria or has not reached the convergence. Genetic algorithm is used to determine the value of learning rate used is based on the calculation of the fitness function with the input of the ANN weights, gradient error, and bias. The calculation of the fitness function will produce an error value of each learning rate which represents each candidate solutions or individual genetic algorithms. Each individual is determined by sum of squared error value. One with the smallest SSE is the best individual. The value of learning rate has chosen will be used to continue learning so that it can lower the value of the error or speed up the learning towards convergence.
 
 
 
 
 The final result of this study is to provide a new solution to resolve the problem in the backpropagation learning that often have problems in determining the learning parameters. These results indicate that the method of genetic algorithms can provide a solution for backpropagation learning in order to decrease the value of SSE when learning of ANN has been static in large error conditions, or stuck in local minima
Read moreDevelopment of an Artificial Neural Aeroelastic Network (AN^2) for the Prediction of Multiple Flutter Crossing
This paper presents an investigation of an artificial neural network (ANN) to predict the flutter characteristics of a simple wing that under went multiple weight variation changes. The weight variations were used to simulate the addition of non-structural weight or structural repair weight. The test case wing had 24 designated repair areas where repair weights were added as lumped masses. Weight and location of the repairs were randomly generated to increase the sample size and help the generalization of the ANN. The ANN was evaluated for both the flutter speed and flutter frequency for multiple levels of structural damping. The number of test cases required to train the neural network was also evaluated for both the error between the input and output results as well as the generalization of the neural networks.
Read morePhysics based data driven method for the crashworthiness design of origami composite tubes
A novel method based on a physics informed data driven model is developed to design an origami composite crash tube. The structure consists of two axially stacked basic components, called modules. Each module presents lower and upper square sections with an octagonal section in the middle. The parameters of the octagonal cross-section and the height of each module are optimized to maximize the energy absorption of the tube when subjected to an axial impact. In contrast to standard surrogate modelling techniques, whose accuracy only depends on the amount of available data, a Physics-informed Neural Network (PINN) scheme is adopted to correlate the crushing response of the single modules to that of the whole origami tube, constraining the data driven method to physically consistent predictions. The PINN is first trained on the results obtained with an experimentally validated Finite Element model and then used to optimize the structure. Results show that the PINN can accurately predict the crushing response of the origami tube, while consistently reducing the computational effort required to explore the whole design domain. Also, the comparison with a standard Feed Forward Neural Network (FFNN) shows that the PINN scheme leads to more accurate results.
Read moreNon-parametric multiple inputs prediction model for magnetic field dependent complex modulus of magnetorheological elastomer
This study introduces a novel platform to predict complex modulus variables as a function of the applied magnetic field and other imperative variables using machine learning. The complex modulus prediction of magnetorheological (MR) elastomers is a challenging process, attributable to the material’s highly nonlinear nature. This problem becomes apparent when considering various possible fabrication parameters. Furthermore, traditional parametric modeling methods are limited when applied to solve larger-scale cases involving large databases. Consequently, the application of non-parametric modeling such as machine learning has gained increasing attraction in recent years. Therefore, this work proposes a data-driven approach for predicting multiple input-dependent complex moduli using feedforward neural networks. Besides excitation frequency and magnetic flux density as operating conditions, the inputs consider compositions and curing conditions represented by magnetic particle weight percentage and the curing magnetic field, respectively. Extreme learning machines and artificial neural networks were used to train the models. The simulation results obtained at various curing conditions and other inputs confirm that the predicted complex modulus has high accuracy with an R2 of about 0.997, as compared to the experimental results. Furthermore, the predicted complex modulus pattern and magnetorheological effect agree with the experimental data using both the learned and unlearned data.
Read more<title>New image reconstruction method in dynamic electrical impedance tomography</title>
Image reconstruction in electrical impedance tomography (EIT) is a non-linear inverse problem. The linear model is always used in most of the reconstruction algorithm in dynamic EIT, which causes large errors of image reconstruction. In this paper, we proposed a new image reconstruction method in dynamic EIT. In the method the artificial neural network based on the error back propagation algorithm (BP ANN) is used to express the non-linear relation between the impedance change position inside the measured object and the voltage change value measured on the surface of the object. Thus, the location ofthe impedance change can be decided by the measured voltage change on the surface, and then the impedance change image will be reconstructed with linear approximated method. The reconstructed error will be decreased largely, because the impedance change position can be detected precisely by our proposed method. The experimental results indicate that the precision of the reconstructed image with our method is greatly higher than that with the back projection method.
Read moreDeep Learning Techniques for Predicting Changes in the Ecosystem
In the modern age, we depend on technology for our daily needs, from groceries to booking tickets for rides. The technology supports us by understanding our requirements. This is done by using Machine Learning. Machine Learning deals with understanding human behavior and providing suggestions for our requirements. The changes in the ecosystem affect the living creatures who depend on the ecosystem. One of the subsets of Machine Learning that play a vital role in saving the lives of living creatures is Deep Learning. Deep Learning is a representational learning of artificial neural networks. Deep Learning keeps on improving so that it can imitate human intelligence more accurately. Artificial Neural Networks is another subset of Machine Learning and helps in the growth of Deep Learning. There are different classes of artificial neural networks, two of the important classes are the Convolutional Neural Network (CNN) and the Recurrent Neural Network (RNN). The patterns in images are recognized by CNN. So, CNN majorly deals with image recognition and processing. RNN helps recognize sequential data and uses this pattern of sequential data to predict the likely scenarios in the ecosystem. The model, which uses the algorithm of RNN and CNN, should be trained and tested with the data for better efficiency.
Read moreThe structure dilemma in biological and artificial neural networks
Brain research up to date has revealed that structure and function are highly related. Thus, for example, studies have repeatedly shown that the brains of patients suffering from schizophrenia or other diseases have a different connectome compared to healthy people. Apart from stochastic processes, however, an inherent logic describing how neurons connect to each other has not yet been identified. We revisited this structural dilemma by comparing and analyzing artificial and biological-based neural networks. Namely, we used feed-forward and recurrent artificial neural networks as well as networks based on the structure of the micro-connectome of C. elegans and of the human macro-connectome. We trained these diverse networks, which markedly differ in their architecture, initialization and pruning technique, and we found remarkable parallels between biological-based and artificial neural networks, as we were additionally able to show that the dilemma is also present in artificial neural networks. Our findings show that structure contains all the information, but that this structure is not exclusive. Indeed, the same structure was able to solve completely different problems with only minimal adjustments. We particularly put interest on the influence of weights and the neuron offset value, as they show a different adaption behaviour. Our findings open up new questions in the fields of artificial and biological information processing research.
Read moreVLSI Bio-Inspired Microsystem for Robust Microarrray Image Analysis and Recognition
A VLSI bio-inspired microsystem for real-time, on-site, robust image analysis and recognition has been developed. A corresponding novel artificial neural network (ANN) learning algorithm using new sigmoid-logarithmic transfer function is invented. Our results show the trained new ANN can recognize low fluorescence patterns better than the conventional sigmoidal ANN does. A differential logarithmic imaging chip is designed for calculating logarithm of relative intensities of fluorescence signals. The single-rail logarithmic circuit and a prototype ANN chip are designed, fabricated and characterized.
Read more