Downloads & Free Reading Options - Results

Talk by Zoran Tiganj of Indiana University.  Given to the Redwood Center for Theoretical Neuroscience at UC Berkeley. Abstract: Building artificial agents that can mimic human learning and reasoning has been a longstanding objective in artificial intelligence. I will discuss some of the empirical data and computational models from neuroscience and cognitive science that could help us advance towards this goal. Specifically, I will talk about the importance of structured representations of knowledge, particularly about mental or cognitive maps for time, space, and concepts. I will present data from recent behavioral and neural studies, which suggest that the brain maintains a scale-invariant mental timeline of the past and uses it to construct a compressed mental timeline of the future. From the computational perspective, these findings illustrate how associative learning can play a role in building structured representations of knowledge. Finally, I will discuss possible strategies to incorporate these findings into building artificial agents, especially in memory-augmented and attention-based neural networks.

This paper presented neural network based maximum power point tracking on the design of photovoltaic power input to a DC DC boot converter to the load. Simulink model of photovoltaic array tested the neural network with different temperature and irradiance for maximum power point of a photovoltaic system. DC DC boot converter is used in load when an average output voltage is stable required which can be lower than the input voltage. At the end, the different temperature and irradiance of the data collected from the photovoltaic array system is used to train the neutral network and output efficiency of the designed DC DC boot converter with MPPT control strategy is accepted the maximum power amount to show the result voltage, current and power output for each different have been presented. And also demonstrated that the neural network based MPPT tracking require less time and more accurate results than the other algorithm based MPPT. By Myint Thuzar | Cho Hnin Moh Moh Aung "Artificial Neural Network for Solar Photovoltaic System Modeling and Simulation" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-5 , August 2019, URL: https://www.ijtsrd.com/papers/ijtsrd27867.pdf Paper URL https://www.ijtsrd.com/engineering/electrical-engineering/27867/artificial-neural-network-for-solar-photovoltaic-system-modeling-and-simulation/myint-thuzar

A hardware fingerprinting primitive known as physical unclonable function (PUF) has a huge potential for secret-key cryptography and identification/authentication applications. The hardware fingerprint is manifested by the random and unique binary strings extracted from the integrated circuit (IC) which exist due to inherent process variations during its fabrication. PUF technology has a huge potential to be used for device identification and authentication in resource-constrained internet of things (IoT) applications such as wireless sensor networks (WSN). A secret computational model of PUF is suggested to be stored in the verifier’s database as an alternative to challenge and response pairs (CRPs) to reduce area consumption. Therefore, in this paper, the design steps to build a PUF model in NodeMCU ESP8266 using an artificial neural network (ANN) are presented. Arbiter-PUF is used in our study and NodeMCU ESP8266 is chosen because it is suitable to be used as a sensor node or sink in WSN applications. ANN with a resilient back-propagation training algorithm is used as it can model the non-linearity with high accuracy. The results show that ANN can model the arbiter-PUF with approximately 99.5% prediction accuracy and the PUF model only consumes 309,889 bytes of memory space.

Natural Language Processing (NLP) systems commonly leverage bag-of-words co-occurrence techniques to capture semantic and syntactic word relationships. The resulting word-level distributed representations often ignore morphological information, though character-level embeddings have proven valuable to NLP tasks. We propose a new neural language model incorporating both word order and character order in its embedding. The model produces several vector spaces with meaningful substructure, as evidenced by its performance of 85.8% on a recent word-analogy task, exceeding best published syntactic word-analogy scores by a 58% error margin. Furthermore, the model includes several parallel training methods, most notably allowing a skip-gram network with 160 billion parameters to be trained overnight on 3 multi-core CPUs, 14x larger than the previous largest neural network.

In this paper, we propose convolutional neural networks for learning an optimal representation of question and answer sentences. Their main aspect is the use of relational information given by the matches between words from the two members of the pair. The matches are encoded as embeddings with additional parameters (dimensions), which are tuned by the network. These allows for better capturing interactions between questions and answers, resulting in a significant boost in accuracy. We test our models on two widely used answer sentence selection benchmarks. The results clearly show the effectiveness of our relational information, which allows our relatively simple network to approach the state of the art.

In this paper we compare different types of recurrent units in recurrent neural networks (RNNs). Especially, we focus on more sophisticated units that implement a gating mechanism, such as a long short-term memory (LSTM) unit and a recently proposed gated recurrent unit (GRU). We evaluate these recurrent units on the tasks of polyphonic music modeling and speech signal modeling. Our experiments revealed that these advanced recurrent units are indeed better than more traditional recurrent units such as tanh units. Also, we found GRU to be comparable to LSTM.

Recurrent neural network (RNN) based character-level language models (CLMs) are extremely useful for modeling out-of-vocabulary words by nature. However, their performance is generally much worse than the word-level language models (WLMs), since CLMs need to consider longer history of tokens to properly predict the next one. We address this problem by proposing hierarchical RNN architectures, which consist of multiple modules with different timescales. Despite the multi-timescale structures, the input and output layers operate with the character-level clock, which allows the existing RNN CLM training approaches to be directly applicable without any modifications. Our CLM models show better perplexity than Kneser-Ney (KN) 5-gram WLMs on the One Billion Word Benchmark with only 2% of parameters. Also, we present real-time character-level end-to-end speech recognition examples on the Wall Street Journal (WSJ) corpus, where replacing traditional mono-clock RNN CLMs with the proposed models results in better recognition accuracies even though the number of parameters are reduced to 30%.

A speaker cluster-based speaker adaptive training (SAT) method under deep neural network-hidden Markov model (DNN-HMM) framework is presented in this paper. During training, speakers that are acoustically adjacent to each other are hierarchically clustered using an i-vector based distance metric. DNNs with speaker dependent layers are then adaptively trained for each cluster of speakers. Before decoding starts, an unseen speaker in test set is matched to the closest speaker cluster through comparing i-vector based distances. The previously trained DNN of the matched speaker cluster is used for decoding utterances of the test speaker. The performance of the proposed method on a large vocabulary spontaneous speech recognition task is evaluated on a training set of with 1500 hours of speech, and a test set of 24 speakers with 1774 utterances. Comparing to a speaker independent DNN with a baseline word error rate of 11.6%, a relative 6.8% reduction in word error rate is observed from the proposed method.

Electricity consumption is increasing gradually and this trend will continue in the future. In addition, rapid network control systems using the resources offered by power electronics and control microelectronics have been recently studied and developed, and are currently in normal application for some, for others, in pilot applications or as prototypes. This paper attempts to show that these systems are referred to by the general acronym flexible alternative current transmission systems (FACTS) similarly dethroned the traditional systems while offering better solutions and solving the energy quality problem such as the hybrid system (unified power flow controller (UPFC), or universal phase shifter regulator (UPSR)) which opens up new perspectives for more efficient operation of networks by continuous and rapid action on the various parameters of the network (voltage, phase shift, and impedance); thus, the power transits will be better controlled and the voltages better held, which will make it possible to increase the stability margins or tend towards the thermal limits of the lines. In this work, we used a classic control (PI-decoupled) and others while offering more flexibility of control thanks to the development of strategies identification/control based on generalized predictive control (GPC) with neural network to ensure robust control with advanced algorithms.

[1] Introduction: Dried fruits are one of the most important non-oil exports and the efforts should be made to grow the economy of the country by increasing their exports to world markets. Meanwhile, quince juice contains various minerals including iron, phosphorus, calcium, potassium and rich in vitamins such as vitamins A, C and B vitamins. Drying of food is one of the ways to keep its quality and increase its shelflife. During this process, the removal of moisture through the simultaneous transfer of heat and mass occurs. By transferring heat from the environment to the foodstuff, the heat energy evaporates the surface moisture. The drying process has a great impact on the product. In recent years, new and innovative techniques have been considered that increase the drying rate and maintain the quality of the product and infrared drying is one of these novel techniques.. Infrared systems are emitting electromagnetic waves with a wavelength of 700 nm to 1 mm. The advantage of using infrared is to minimize waste and prevent product quality loss due to reduced drying time can be mentioned. The need to predict product quality in each process makes it necesary to model and discover the relationship between factors that can affect the final quality of the product. Artificial neural networks have been considered as a meta-innovative algorithm for modeling and prediction, which can be favored by the ability of these networks to model and predict processes The complexity and discovery of non-random fluctuations in data and the ability to discover the interactions between variables, economical savings in the use and disconnection of classical model abusive constraints (Togrul et al., 2004), the ability to reduce The effect of non-effective variables on the model by setting internal parameters is the ability to predict the desired parameter variations with minimum parameters (Bowers et al., 2000).   Materials and methods: In this research, quince fruit (Variety of Isfahan) was purchased as the premium product of Isfahan Gardens and was kept at 0 ° C in the cold room prior to further experiments. The fruits were removed from the refrigerator one hour before processing and exposed to ambient temperature. After washing, surface moisture was removed by moisture absorbent paper and turned into slices with a constant thickness of 4 mm. The specimens were subjected to pre-treatment with an osmotic solution (vacuum for 70 minutes at a temperature of 40 ° C for 5 hours). For drying the samples, an infrared convective dryer with three voltages (800.400 and 1200 watts) and a constant speed of 0.5 m / s was used. In this way, the samples were placed under infrared lamps on a plate made from a grid and the weight of the samples was measured in a scale of 10 minutes by means of a scale and recorded on the computer. In order to achieve stable conditions in the system, the dryer was switched on 30 minutes before the process. The distance between the samples and the infrared lamp was fixed in all treatments at 16 cm. The drying process continued to reach a moisture content of 0.22 basis. To perform a puncture tests, quince slices were used in a Brookfield-based American LFRA-4500 tissue analysis device. In order to model these parameters in the drying process, the results of examining the quality of the samples, including the firmness of the tissue as well as the drying time, were used as network outputs. The power, concentration and pressure parameters were considered as network inputs. In this research, a multilayer perceptron network (MLP) was used. Due to its simplicity and high precision, this model has a great application in modeling the drying of agricultural products. Many functions in transmitting numbers from the previous layer to the next layer may be used (Tripathy et al., 2008).   Result & discussion: The results indicated that the stiffness of the tissue is reduced in vacuum conditions with increased power. So, the least amount of stiffness was related to osmotic sample dried at 1200 watts. By increasing the infrared power, the stiffness of the tissue decreases, the reason for this is probably the volume increase phenomenon that occurs during the rapid evaporation of moisture through infrared rays from inside the tissue. The results showed that at the start of the drying process, due to the high moisture content of the product, the moisture loss rate is high. Gradually, with the advent of time and reduced initial moisture content, the rate of moisture reduction naturally decreases. At lower power, the drying time is longer and with increasing power, the drying time decreases due to the increase of the thermal gradient inside the product and consequently the increase in the rate of evaporation of the moisture content of the product. The results of this study showed that the neural artificial network, as a powerful tool, can estimate the stiffness parameters of the tissue and the drying time with high precision. The most suitable neural network structure to predict these parameters with a 3-7-2 topology along with logarithmic activation functions with a total explanation coefficient above 0.9923 represent the best results. Also, by increasing the drying capacity and using osmotic dehydration, the drying time and the stiffness of the tissue samples is decreased.

Neural networks provide an alternative method of building models of human performance. They can learn behavior from examples, reducing the need for many identical repetitions and intensive analysis. A properly trained net can be very robust in its response to a novel stimulus. This opens the door to modeling performance in the presence of an interactive stimulus. Neural networks provide the possibility of robust models that can operate interactively in real time, depending on the size and architecture of the net and the application. A neural network architecture derived from recurrent back propagation is presented which learn to mimic human behavior and performance in a sample task. It shows operating characteristics similar to those of human subjects, and even makes the same kinds of mistakes. Possible application are discussed. (js)

Neural networks provide an alternative method of building models of human performance. They can learn behavior from examples, reducing the need for many identical repetitions and intensive analysis. A properly trained net can be very robust in its response to a novel stimulus. This opens the door to modeling performance in the presence of an interactive stimulus. Neural networks provide the possibility of robust models that can operate interactively in real time, depending on the size and architecture of the net and the application. A neural network architecture derived from recurrent back propagation is presented which learn to mimic human behavior and performance in a sample task. It shows operating characteristics similar to those of human subjects, and even makes the same kinds of mistakes. Possible application are discussed. (js)

This article is from Journal of Visualized Experiments : JoVE . Abstract Migraine and its transformation to chronic migraine are healthcare burdens in need of improved treatment options. We seek to define how neural immune signaling modulates the susceptibility to migraine, modeled in vitro using spreading depression (SD), as a means to develop novel therapeutic targets for episodic and chronic migraine. SD is the likely cause of migraine aura and migraine pain. It is a paroxysmal loss of neuronal function triggered by initially increased neuronal activity, which slowly propagates within susceptible brain regions. Normal brain function is exquisitely sensitive to, and relies on, coincident low-level immune signaling. Thus, neural immune signaling likely affects electrical activity of SD, and therefore migraine. Pain perception studies of SD in whole animals are fraught with difficulties, but whole animals are well suited to examine systems biology aspects of migraine since SD activates trigeminal nociceptive pathways. However, whole animal studies alone cannot be used to decipher the cellular and neural circuit mechanisms of SD. Instead, in vitro preparations where environmental conditions can be controlled are necessary. Here, it is important to recognize limitations of acute slices and distinct advantages of hippocampal slice cultures. Acute brain slices cannot reveal subtle changes in immune signaling since preparing the slices alone triggers: pro-inflammatory changes that last days, epileptiform behavior due to high levels of oxygen tension needed to vitalize the slices, and irreversible cell injury at anoxic slice centers. In contrast, we examine immune signaling in mature hippocampal slice cultures since the cultures closely parallel their in vivo counterpart with mature trisynaptic function; show quiescent astrocytes, microglia, and cytokine levels; and SD is easily induced in an unanesthetized preparation. Furthermore, the slices are long-lived and SD can be induced on consecutive days without injury, making this preparation the sole means to-date capable of modeling the neuroimmune consequences of chronic SD, and thus perhaps chronic migraine. We use electrophysiological techniques and non-invasive imaging to measure neuronal cell and circuit functions coincident with SD. Neural immune gene expression variables are measured with qPCR screening, qPCR arrays, and, importantly, use of cDNA preamplification for detection of ultra-low level targets such as interferon-gamma using whole, regional, or specific cell enhanced (via laser dissection microscopy) sampling. Cytokine cascade signaling is further assessed with multiplexed phosphoprotein related targets with gene expression and phosphoprotein changes confirmed via cell-specific immunostaining. Pharmacological and siRNA strategies are used to mimic and modulate SD immune signaling.

This article is from Journal of Visualized Experiments : JoVE . Abstract Migraine and its transformation to chronic migraine are healthcare burdens in need of improved treatment options. We seek to define how neural immune signaling modulates the susceptibility to migraine, modeled in vitro using spreading depression (SD), as a means to develop novel therapeutic targets for episodic and chronic migraine. SD is the likely cause of migraine aura and migraine pain. It is a paroxysmal loss of neuronal function triggered by initially increased neuronal activity, which slowly propagates within susceptible brain regions. Normal brain function is exquisitely sensitive to, and relies on, coincident low-level immune signaling. Thus, neural immune signaling likely affects electrical activity of SD, and therefore migraine. Pain perception studies of SD in whole animals are fraught with difficulties, but whole animals are well suited to examine systems biology aspects of migraine since SD activates trigeminal nociceptive pathways. However, whole animal studies alone cannot be used to decipher the cellular and neural circuit mechanisms of SD. Instead, in vitro preparations where environmental conditions can be controlled are necessary. Here, it is important to recognize limitations of acute slices and distinct advantages of hippocampal slice cultures. Acute brain slices cannot reveal subtle changes in immune signaling since preparing the slices alone triggers: pro-inflammatory changes that last days, epileptiform behavior due to high levels of oxygen tension needed to vitalize the slices, and irreversible cell injury at anoxic slice centers. In contrast, we examine immune signaling in mature hippocampal slice cultures since the cultures closely parallel their in vivo counterpart with mature trisynaptic function; show quiescent astrocytes, microglia, and cytokine levels; and SD is easily induced in an unanesthetized preparation. Furthermore, the slices are long-lived and SD can be induced on consecutive days without injury, making this preparation the sole means to-date capable of modeling the neuroimmune consequences of chronic SD, and thus perhaps chronic migraine. We use electrophysiological techniques and non-invasive imaging to measure neuronal cell and circuit functions coincident with SD. Neural immune gene expression variables are measured with qPCR screening, qPCR arrays, and, importantly, use of cDNA preamplification for detection of ultra-low level targets such as interferon-gamma using whole, regional, or specific cell enhanced (via laser dissection microscopy) sampling. Cytokine cascade signaling is further assessed with multiplexed phosphoprotein related targets with gene expression and phosphoprotein changes confirmed via cell-specific immunostaining. Pharmacological and siRNA strategies are used to mimic and modulate SD immune signaling.

Ionospheric models have been developed to interpret Relocatable Over-the-Horizon Radar data. This thesis examines the applicability of neural networks to ionospheric modeling in support of Relocatable Over-the-Horizon Radar. Two neural networks were used for this investigation. The flrst network was trained and tested on experimental ionospheric sounding data. Results showed neural networks are excellent at modeling ionospheric data for a given day. The second network was trained on ionospheric models and tested on experimental data. Results showed neural networks are able to learn many ionospheric models and the modeling network generally agreed with the experimental data.

Feedforward Neural Network (FNN)-based language models estimate the probability of the next word based on the history of the last N words, whereas Recurrent Neural Networks (RNN) perform the same task based only on the last word and some context information that cycles in the network. This paper presents a novel approach, which bridges the gap between these two categories of networks. In particular, we propose an architecture which takes advantage of the explicit, sequential enumeration of the word history in FNN structure while enhancing each word representation at the projection layer through recurrent context information that evolves in the network. The context integration is performed using an additional word-dependent weight matrix that is also learned during the training. Extensive experiments conducted on the Penn Treebank (PTB) and the Large Text Compression Benchmark (LTCB) corpus showed a significant reduction of the perplexity when compared to state-of-the-art feedforward as well as recurrent neural network architectures.

A dataset of 854 small unmanned aerial system (SUAS) flight experiments from 2005-2009 is analyzed to determine significant factors that contribute to mishaps. The data from 29 airframes of different designs and technology readiness levels were aggregated. 20 measured parameters from each flight experiment are investigated, including wind speed, pilot experience, number of prior flights, pilot currency, etc. Outcomes of failures (loss of flight data) and damage (injury to airframe) are classified by logistic regression modeling and artificial neural network analysis. From the analysis, it can be concluded that SUAS damage is a random event that cannot be predicted with greater accuracy than guessing. Failures can be predicted with greater accuracy (38.5% occurrence, model hit rate 69.6%). Five significant factors were identified by both the neural networks and logistic regression. SUAS prototypes risk failures at six times the odds of their commercially manufactured counterparts. Likewise, manually controlled SUAS have twice the odds of experiencing a failure as those autonomously controlled. Wind speeds, pilot experience, and pilot currency were not found to be statistically significant to flight outcomes. The implications of these results for decision makers, range safety officers and test engineers are discussed.

The purposes of this research were: (1) validating Kim's (2007) simulation method by applying analytic methods and (2) comparing the two different Robust Parameter Design methods with three measures of performance (label accuracy for enemy, friendly, and clutter). Considering the features of CID, input variables were defined as two controllable (threshold combination of detector and classifier) and three uncontrollable (map size, number of enemies and friendly). The first set of experiments considers Kim's method using analytical methods. In order to create response variables, Kim's method uses Monte Carlo simulation. The output results showed no difference between simulation and the analytic method. The second set of experiments compared the measures of performance between a standard RPD used by Kim and a new method using Artificial Neural Networks (ANNs). To find optimal combinations of detection and classification thresholds, Kim's model uses regression with a combined array design, whereas the ANNs method uses ANN with a crossed array design. In the case of label accuracy for enemy, Kim's solution showed the higher expected value, however it also showed a higher variance. Additionally, the model's residuals were higher for Kim's model.

The paper presents the results of neuromodeling of the volume distribution of axon spikes in myelinated axons. The nature of this is shown to be much more complex than is currently believed distribution, and the significance of the effect of features related to volume is substantiated dissemination. Based on the analysis of electrochemical transmission mechanisms, it is assumed that the substrate long-term memory can be myelin sheaths of axons, which is important in understanding the development of relevant pathologies, in particular, Parkinson's and Alzheimer's diseases.

Large amount of small Unmanned Aerial Vehicles (sUAVs) are projected to operate in the near future. Potential sUAV applications include, but not limited to, search and rescue, inspection and surveillance, aerial photography and video, precision agriculture, and parcel delivery. sUAVs are expected to operate in the uncontrolled Class G airspace, which is at or below 500 feet above ground level (AGL), where many static and dynamic constraints exist, such as ground properties and terrains, restricted areas, various winds, manned helicopters, and conflict avoidance among sUAVs. How to enable safe, efficient, and massive sUAV operations at the low altitude airspace remains a great challenge. NASA's Unmanned aircraft system Traffic Management (UTM) research initiative works on establishing infrastructure and developing policies, requirement, and rules to enable safe and efficient sUAVs' operations. To achieve this goal, it is important to gain insights of future UTM traffic operations through simulations, where the accurate trajectory model plays an extremely important role. On the other hand, like what happens in current aviation development, trajectory modeling should also serve as the foundation for any advanced concepts and tools in UTM. Accurate models of sUAV dynamics and control systems are very important considering the requirement of the meter level precision in UTM operations. The vehicle dynamics are relatively easy to derive and model, however, vehicle control systems remain unknown as they are usually kept by manufactures as a part of intellectual properties. That brings challenges to trajectory modeling for sUAVs. How to model the vehicle's trajectories with unknown control system? This work proposes to use a neural network to model a vehicle's trajectory. The neural network is first trained to learn the vehicle's responses at numerous conditions. Once being fully trained, given current vehicle states, winds, and desired future trajectory, the neural network should be able to predict the vehicle's future states at next time step. A complete 4-D trajectory are then generated step by step using the trained neural network. Experiments in this work show that the neural network can approximate the sUAV's model and predict the trajectory accurately.

During the past decade, machine learning has become extremely popular and can be found in many aspects of our every day life. Nowayadays with explosion of data while rapid growth of computation capacity, Distributed Deep Neural Networks (DDNNs) which can improve their performance linearly with more computation resources, have become hot and trending. However, there has not been an in depth study of the performance of these systems, and how well they scale. In this paper we analyze CNTK, one of the most commonly used DDNNs, by first building a performance model and then evaluating the system two settings: a small cluster with all nodes in a single rack connected to a top of rack switch, and in large scale using Blue Waters with arbitary placement of nodes. Our main focus was the scalability of the system with respect to adding more nodes. Based on our results, this system has an excessive initialization overhead because of poor I/O utilization which dominates the whole execution time. Because of this, the system does not scale beyond a few nodes (4 in Blue Waters). Additionally, due to a single server-multiple worker design the server becomes a bottleneck after 16 nodes limiting the scalability of the CNTK.

Speaker intent detection and semantic slot filling are two critical tasks in spoken language understanding (SLU) for dialogue systems. In this paper, we describe a recurrent neural network (RNN) model that jointly performs intent detection, slot filling, and language modeling. The neural network model keeps updating the intent estimation as word in the transcribed utterance arrives and uses it as contextual features in the joint model. Evaluation of the language model and online SLU model is made on the ATIS benchmarking data set. On language modeling task, our joint model achieves 11.8% relative reduction on perplexity comparing to the independent training language model. On SLU tasks, our joint model outperforms the independent task training model by 22.3% on intent detection error rate, with slight degradation on slot filling F1 score. The joint model also shows advantageous performance in the realistic ASR settings with noisy speech input.

When humans read text, they fixate some words and skip others. However, there have been few attempts to explain skipping behavior with computational models, as most existing work has focused on predicting reading times (e.g.,~using surprisal). In this paper, we propose a novel approach that models both skipping and reading, using an unsupervised architecture that combines a neural attention with autoencoding, trained on raw text using reinforcement learning. Our model explains human reading behavior as a tradeoff between precision of language understanding (encoding the input accurately) and economy of attention (fixating as few words as possible). We evaluate the model on the Dundee eye-tracking corpus, showing that it accurately predicts skipping behavior and reading times, is competitive with surprisal, and captures known qualitative features of human reading.

The convolutional neural networks (CNNs) have proven to be a powerful tool for discriminative learning. Recently researchers have also started to show interest in the generative aspects of CNNs in order to gain a deeper understanding of what they have learned and how to further improve them. This paper investigates generative modeling of CNNs. The main contributions include: (1) We construct a generative model for the CNN in the form of exponential tilting of a reference distribution. (2) We propose a generative gradient for pre-training CNNs by a non-parametric importance sampling scheme, which is fundamentally different from the commonly used discriminative gradient, and yet has the same computational architecture and cost as the latter. (3) We propose a generative visualization method for the CNNs by sampling from an explicit parametric image distribution. The proposed visualization method can directly draw synthetic samples for any given node in a trained CNN by the Hamiltonian Monte Carlo (HMC) algorithm, without resorting to any extra hold-out images. Experiments on the challenging ImageNet benchmark show that the proposed generative gradient pre-training consistently helps improve the performances of CNNs, and the proposed generative visualization method generates meaningful and varied samples of synthetic images from a large-scale deep CNN.

Click here to view the University of Florida catalog record

Here we demonstrate that the activity of neural ensembles can be quantitatively modeled. We first show that an ensemble dynamical model (EDM) accurately approximates the distribution of voltages and average firing rate per neuron of a population of simulated integrate-and-fire neurons. EDMs are high-dimensional nonlinear dynamical models. To faciliate the estimation of their parameters we present a dimensionality reduction method and study its performance with simulated data. We then introduce and evaluate a maximum-likelihood method to estimate connectivity parameters in networks of EDMS. Finally, we show that this model an methods accurately approximate the high-gamma power evoked by pure tones in the auditory cortex of rodents. Overall, this article demonstrates that quantitatively modeling brain activity at the ensemble level is indeed possible, and opens the way to understanding the computations performed by neural ensembles, which could revolutionarize our understanding of brain function.

Here we demonstrate that the activity of neural ensembles can be quantitatively modeled. We first show that an ensemble dynamical model (EDM) accurately approximates the distribution of voltages and average firing rate per neuron of a population of simulated integrate-and-fire neurons. EDMs are high-dimensional nonlinear dynamical models. To faciliate the estimation of their parameters we present a dimensionality reduction method and study its performance with simulated data. We then introduce and evaluate a maximum-likelihood method to estimate connectivity parameters in networks of EDMS. Finally, we show that this model an methods accurately approximate the high-gamma power evoked by pure tones in the auditory cortex of rodents. Overall, this article demonstrates that quantitatively modeling brain activity at the ensemble level is indeed possible, and opens the way to understanding the computations performed by neural ensembles, which could revolutionarize our understanding of brain function.

While processing polymetallic ores at the non-ferrous metallurgy problems arises connecting with the excellence of production and the efficient applying the technological devices-firing furnace and crusher machine. In earlier time, similar questions were solved due to the big practice experiences and using a mathematical modeling method. The mathematical model for optimizing those operating mode is a very complex and hard to calculation. Performing computations is needed too much time and many resources. Because the control of the agglomeration furnaces and other machines are including complex multiparameter processes. The method of the math modeling for optimization the operating mode to the firing furnace is replaced with modeling based on the neural network that is here a new method. The results obtained have shown that proposed methods based on the neural network modeling of metallurgical processes allow determining more accurate and adequate results of calculations than mathematical modeling by the traditional program. The use of new approaches for modeling the technological processes at the non-ferrous metallurgy gives opportunity to enhance an effectiveness of production excellence and to enhance an efficient applying the heat-energy equipment while a firing the sulfide polymetallic ores of non-ferrous metallurgy.

This article is from Frontiers in Computational Neuroscience , volume 7 . Abstract Drosophila larvae crawl by peristaltic waves of muscle contractions, which propagate along the animal body and involve the simultaneous contraction of the left and right side of each segment. Coordinated propagation of contraction does not require sensory input, suggesting that movement is generated by a central pattern generator (CPG). We characterized crawling behavior of newly hatched Drosophila larvae by quantifying timing and duration of segmental boundary contractions. We developed a CPG network model that recapitulates these patterns based on segmentally repeated units of excitatory and inhibitory (EI) neuronal populations coupled with immediate neighboring segments. A single network with symmetric coupling between neighboring segments succeeded in generating both forward and backward propagation of activity. The CPG network was robust to changes in amplitude and variability of connectivity strength. Introducing sensory feedback via “stretch-sensitive” neurons improved wave propagation properties such as speed of propagation and segmental contraction duration as observed experimentally. Sensory feedback also restored propagating activity patterns when an inappropriately tuned CPG network failed to generate waves. Finally, in a two-sided CPG model we demonstrated that two types of connectivity could synchronize the activity of two independent networks: connections from excitatory neurons on one side to excitatory contralateral neurons (E to E), and connections from inhibitory neurons on one side to excitatory contralateral neurons (I to E). To our knowledge, such I to E connectivity has not yet been found in any experimental system; however, it provides the most robust mechanism to synchronize activity between contralateral CPGs in our model. Our model provides a general framework for studying the conditions under which a single locally coupled network generates bilaterally synchronized and longitudinally propagating waves in either direction.

Models to predict tree survival and mortality can help to understand vegetation dynamics and to predict effects of climate change on native forests. The objective of the present study was to use Artificial Neural Networks, based on the competition index and climatic and categorical variables, to predict tree survival and mortality in Semideciduous Seasonal Forests in the Atlantic Forest biome. Numerical and categorical trees variables, in permanent plots, were used. The Agricultural Reference Index for Drought (ARID) and the distance-dependent competition index were the variables used. The overall efficiency of classification by ANNs was higher than 92% and 93% in the training and test, respectively. The accuracy for classification and number of surviving trees was above 99% in the test and in training for all ANNs. The classification accuracy of the number of dead trees was low. The mortality accuracy rate (10.96% for training and 13.76% for the test) was higher with the ANN 4, which considers the climatic variable and the competition index. The individual tree-level model integrates dendrometric and meteorological variables, representing a new step for modeling tree survival in the Atlantic Forest biome.

Research works are ongoing in mixing the biologically synthesized oil with the diesel for reducing the effect of global warming and climate change. From the review study, it is noted that the blended biodiesels require more assert about their practical viability. So, the non-edible Tamanu oil is synthesized and it is blended with diesel and its emission characteristics, engine performance and combustion characteristics are studied in our previous work. This paper attempts to model the diesel engine fueled with tamanu oil biodiesel blend. The proposed model exploits the context of neural network and the firefly algorithm is used to train it. After analyzing the various characteristics of the diesel engine, the acquired data is subjected to a proposed FF-NM method. The simulated results are statistically evaluated and the proposed modeling method is proved to be better than the other NM.

Neural network technology has recently demonstrated capabilities in areas important to personnel research such as statistical analysis, decision modeling, control, and forecasting. The present investigation indicates that three different neural network architectures are particularly suited to modeling many aspects of the Air Force personnel system: back propagation, learning vector quantization, and probabilistic neural networks. The primary advantage of neutral networks is their ability to derive nonlinear and interacting relationships among model variables. Two areas investigated in order to evaluate this capability were airmen reenlistment decisions and airman inventory modeling.

Lecture: Neural Modeling And Imaging Of Stuttering

Introduction Microwave drying compared to conventional hot air drying has many benefits to apply in food drying processes such as volumetric heating, high thermal efficiency, shorter drying time and improved product quality. In conventional microwave drying method, a fixed microwave power was used during the drying process. However, the water of the product evaporated and mass of product decreased over the time that resulted in microwave power density (MPD) increasing during the drying process. Increasing the power density, especially at the end of the process, sharply increased the product temperature. High temperature of products led to the deterioration of the product quality. Most research used variable microwave power program for preventing the risk of overheating and charring of product. The evaporation of the water causes the shrinkage of product. Therefore, many studies have used machine vision for measuring the shrinkage and this technology has been used in modeling and predicting the MC. Materials and Methods The fresh potato samples ( Solanum tuberosum cv. Santana) with 83% (w.b.) of initial MC were sliced into the chips of 5mm thickness. The developed drying systems consisted of microwave oven, lighting unit and imaging unit, temperature sensor, microwave power adjusting unit and a data acquisition unit (DAQ). A LabVIEW (V17.6, 2017) program was developed to integrate all measurements and adjusting the microwave power during the drying process. In this study, two sets of experiment with different aims have done. The first set of experiments was used for calculating the shrinkage by developed image processing algorithm and MC by offline mass measurement and then data sets were used to investigate the artificial neural networks (ANNs). The second set was used for evaluating the reliability of investigating models. The experiments, in the first set, were done with 8, 4 and 2.67 W g -1 . In the variable mode, the power varied in two/three steps with respect to the MC of samples during the drying process. Second set of experiments was done in two variable and constant power modes with 5 and 3 W g -1 . An image processing algorithm was developed to measure the shrinkage of potato slice during the drying process. In this study the feed forward ANN with back propagation algorithm was used. Two structures of ANN were used for modeling of MC. In the first model time and power density and the second model shrinkage and power density were used as input. Also moisture ratio was used as an output parameter in two models. Results and Discussion The obtained results indicated that for the first model the ANN with 2-3-1 structure had better results than others structures. This structure had 0.0713, 0.0337 and 0.0640 of RMSE and 0.9764, 0.9973 and 0.9800 of R for train, validation and test, respectively. For the second model, the 2-2-2-1 structure of ANN with 0.0780, 0.0816 and 0.0908 of RMSE and 0.9598, 0.9799 and 0.9746 of R for train, validation and test, respectively had better results than other structures. The evaluation of these models with a second data set showed that the second model with shrinkage and power density as input with 0.067 of RMSE and 0.994 of R had better results than the first model with 0.173 of RMSE and 0.961 of R. These consequences expressed that the second model had higher reliability for prediction of MC based on shrinkage and power density during drying process. Conclusion In this study, a microwave dryer was developed with a real-time image recording system and a microwave power level program during the drying process. Two ANN models were used for modeling of drying kinetics of the potato slices. Also image processing algorithm was investigated by measuring the shrinkage of potato slice during the drying process. The outcomes revealed that shrinkage as input in the ANN had great effect on MC prediction during the drying process.

Econophysics has developed as a research field that applies the formalism of Statistical Mechanics and Quantum Mechanics to address Economics and Finance problems. The branch of Econophysics that applies of Quantum Theory to Economics and Finance is called Quantum Econophysics. In Finance, Quantum Econophysics' contributions have ranged from option pricing to market dynamics modeling, behavioral finance and applications of Game Theory, integrating the empirical finding, from human decision analysis, that shows that nonlinear update rules in probabilities, leading to non-additive decision weights, can be computationally approached from quantum computation, with resulting quantum interference terms explaining the non-additive probabilities. The current work draws on these results to introduce new tools from Quantum Artificial Intelligence, namely Quantum Artificial Neural Networks as a way to build and simulate financial market models with adaptive selection of trading rules, leading to turbulence and excess kurtosis in the returns distributions for a wide range of parameters.

Fixed-vocabulary language models fail to account for one of the most characteristic statistical facts of natural language: the frequent creation and reuse of new word types. Although character-level language models offer a partial solution in that they can create word types not attested in the training corpus, they do not capture the "bursty" distribution of such words. In this paper, we augment a hierarchical LSTM language model that generates sequences of word tokens character by character with a caching mechanism that learns to reuse previously generated words. To validate our model we construct a new open-vocabulary language modeling corpus (the Multilingual Wikipedia Corpus, MWC) from comparable Wikipedia articles in 7 typologically diverse languages and demonstrate the effectiveness of our model across this range of languages.

This paper proposes a new data-driven approach for modeling detailed splashes for liquid simulations with neural networks. Our model learns to generate small-scale splash detail for fluid-implicit-particle methods using training data acquired from physically accurate, high-resolution simulations. We use neural networks to model the regression of splash formation using a classifier together with a velocity modification term. More specifically, we employ a heteroscedastic model for the velocity updates. Our simulation results demonstrate that our model significantly improves visual fidelity with a large amount of realistic droplet formation and yields splash detail much more efficiently than finer discretizations. We show this for two different spatial scales and simulation setups.

This study investigates how victim attitude responses (neutral vs. negative) influence third-party punishment decisions, using EEG and computational modeling. The goal is to understand the cognitive and neural mechanisms that underlie third-party punishment when victim feedback is incorporated. The EEG experiment aims to clarify the neural indicators of different victim attitudes (neutral vs. negative) under fair and unfair conditions, while the behavioral replication experiment seeks to replicate the behavioral results observed in the EEG study. Importantly, the study combines utility models to explore how parameters change in different attitude contexts, providing insights into the underlying psychological mechanisms.

We introduce a nonlinear modification of the classical Hawkes process, which allows inhibitory couplings between units without restrictions. The resulting system of interacting point processes provides a useful mathematical model for recurrent networks of spiking neurons with exponential transfer functions. The expected rates of all neurons in the network are approximated by a first-order differential system. We study the stability of the solutions of this equation, and use the new formalism to implement a winner-takes-all network that operates robustly for a wide range of parameters. Finally, we discuss relations with the generalised linear model that is widely used for the analysis of spike trains.

We introduce a nonlinear modification of the classical Hawkes process, which allows inhibitory couplings between units without restrictions. The resulting system of interacting point processes provides a useful mathematical model for recurrent networks of spiking neurons with exponential transfer functions. The expected rates of all neurons in the network are approximated by a first-order differential system. We study the stability of the solutions of this equation, and use the new formalism to implement a winner-takes-all network that operates robustly for a wide range of parameters. Finally, we discuss relations with the generalised linear model that is widely used for the analysis of spike trains.

No Abstract Available

In this study, the Artificial Neural Network (ANN) application is implemented for predicting the required concrete volume and amount of the steel reinforcement within the inversed-T-shaped and stem-stepped reinforced concrete (RC) walls. For this aim, seven-different RC wall designs were approached differentiated within the wall heights and various internal friction angles of backfill materials. Each RC wall is proportionally designed and subjected to active lateral earth pressure defined with the Mononobe-Okabe approach foreseen by Turkish Specification for Building to be Built in Seismic Zones (TSC-2007). Following the stability analysis of the RC retaining walls, the structural and reinforced concrete analyses are performed according to the Turkish Standard on Requirements for Design and Construction in Reinforced Concrete Structures (TS500-2000). Input parameters such as concrete volumes, weights of the steel bars, soil and wall material properties are subjected to the ANN modeling. The prediction of the concrete volume and amount of the steel bars are achieved with the implementation of the ANN model trained with the Artificial Bee Colony (ABC) algorithm. As a result of this study, it is revealed that ANN models are useful for verifying the existing RC retaining wall designs or performing preliminary designs for the L-shaped and stem-stepped cantilever retaining walls.

This article is from BMC Neuroscience , volume 15 . Abstract None

Statistical Analysis Plan (SAP) As part of the larger TIP project, 61 SAD patients and 41 healthy controls underwent an emotion regulation task with negative and neutral faces during fMRI scanning. We will use dynamic causal modeling (DCM) to shed light on potential disturbances in the effective connectivity of emotion regulation networks in social anxiety disorder (SAD).

A simple model that replicates the dynamics of spiking and spiking-bursting activity of real biological neurons is proposed. The model is a two-dimensional map which contains one fast and one slow variable. The mechanisms behind generation of spikes, bursts of spikes, and restructuring of the map behavior are explained using phase portrait analysis. The dynamics of two coupled maps which model the behavior of two electrically coupled neurons is discussed. Synchronization regimes for spiking and bursting activity of these maps are studied as a function of coupling strength. It is demonstrated that the results of this model are in agreement with the synchronization of chaotic spiking-bursting behavior experimentally found in real biological neurons.

http://uf.catalog.fcla.edu/uf.jsp?st=UF002056070&ix=pm&I=0&V=D&pm=1

We present a new convolutional neural network-based time-series model. Typical convolutional neural network (CNN) architectures rely on the use of max-pooling operators in between layers, which leads to reduced resolution at the top layers. Instead, in this work we consider a fully convolutional network (FCN) architecture that uses causal filtering operations, and allows for the rate of the output signal to be the same as that of the input signal. We furthermore propose an undecimated version of the FCN, which we refer to as the undecimated fully convolutional neural network (UFCNN), and is motivated by the undecimated wavelet transform. Our experimental results verify that using the undecimated version of the FCN is necessary in order to allow for effective time-series modeling. The UFCNN has several advantages compared to other time-series models such as the recurrent neural network (RNN) and long short-term memory (LSTM), since it does not suffer from either the vanishing or exploding gradients problems, and is therefore easier to train. Convolution operations can also be implemented more efficiently compared to the recursion that is involved in RNN-based models. We evaluate the performance of our model in a synthetic target tracking task using bearing only measurements generated from a state-space model, a probabilistic modeling of polyphonic music sequences problem, and a high frequency trading task using a time-series of ask/bid quotes and their corresponding volumes. Our experimental results using synthetic and real datasets verify the significant advantages of the UFCNN compared to the RNN and LSTM baselines.

We are now witnessing the increasing availability of event stream data, i.e., a sequence of events with each event typically being denoted by the time it occurs and its mark information (e.g., event type). A fundamental problem is to model and predict such kind of marked temporal dynamics, i.e., when the next event will take place and what its mark will be. Existing methods either predict only the mark or the time of the next event, or predict both of them, yet separately. Indeed, in marked temporal dynamics, the time and the mark of the next event are highly dependent on each other, requiring a method that could simultaneously predict both of them. To tackle this problem, in this paper, we propose to model marked temporal dynamics by using a mark-specific intensity function to explicitly capture the dependency between the mark and the time of the next event. Extensive experiments on two datasets demonstrate that the proposed method outperforms state-of-the-art methods at predicting marked temporal dynamics.

Microelectromechanical systems (MEMS) are a broad and rapidly expanding field that is currently receiving a great deal of attention because of the potential to significantly improve the ability to sense, analyze, and control a variety of processes, such as heating and ventilation systems, automobiles, medicine, aeronautical flight, military surveillance, weather forecasting, and space exploration. MEMS are very small and are a blend of electrical and mechanical components, with electrical and mechanical systems on one chip. This research establishes reliability estimation and prediction for MEMS devices at the conceptual design phase using neural networks. At the conceptual design phase, before devices are built and tested, traditional methods of quantifying reliability are inadequate because the device is not in existence and cannot be tested to establish the reliability distributions. A novel approach using neural networks is created to predict the overall reliability of a MEMS device based on its components and each component's attributes. The methodology begins with collecting attribute data (fabrication process, physical specifications, operating environment, property characteristics, packaging, etc.) and reliability data for many types of microengines. The data are partitioned into training data (the majority) and validation data (the remainder). A neural network is applied to the training data (both attribute and reliability); the attributes become the system inputs and reliability data (cycles to failure), the system output. After the neural network is trained with sufficient data. the validation data are used to verify the neural networks provided accurate reliability estimates. Now, the reliability of a new proposed MEMS device can be estimated by using the appropriate trained neural networks developed in this work.

The main way to produce cast iron is in the blast furnace. In the production of hot metal, the control of silicon is important. Alumina and silica react chemically with limestone and dolomite to form blast furnace slag. In this work, 12 artificial neural networks (ANNs) were modeled with different numbers of neurons in each hidden layer. The number of neurons varied between 10 and 200 neurons. ANNs were used to predict the silicon content of hot metal produced. The ANN with 30 neurons showed the best performance. In the test phase, the mathematical correlation was 97.5% and the mean square error (MSE) was 0.0006, and in the cross-validation phase, the mathematical correlation was 95.5% while the MSE was 0.00035.