Directions for future research in neural networks

Graph Neural Networks (GNNs), a generalization of deep neural networks on graph data have been widely used in various domains, ranging from drug discovery to recommender systems. However, GNNs on such applications are limited when there are few available samples. Meta-learning has been an important framework to address the lack of samples in machine learning, and in recent years, researchers have started to apply meta-learning to GNNs. In this work, we provide a comprehensive survey of different metalearning approaches involving GNNs on various graph problems showing the power of using these two approaches together. We categorize the literature based on proposed architectures, shared representations, and applications. Finally, we discuss several exciting future research directions and open problems.

Download Full-text

Artificial Intelligence Review

Advanced Methodologies and Technologies in Artificial Intelligence, Computer Simulation, and Human-Computer Interaction - Advances in Computer and Electrical Engineering ◽

10.4018/978-1-5225-7368-5.ch003 ◽

2019 ◽

pp. 23-39

Author(s):

Amal Kilani ◽

Ahmed Ben Hamida ◽

Habib Hamam

Keyword(s):

Artificial Intelligence ◽

Neural Networks ◽

Fuzzy Logic ◽

Intelligent System ◽

Route Planning ◽

Turing Test ◽

Future Research ◽

Interaction Data ◽

Research Directions ◽

Future Research Directions

In this chapter, the authors present a profound literature review of artificial intelligence (AI). After defining it, they briefly cover its history and enumerate its principal fields of application. They name, for example, information system, commerce, image processing, human-computer interaction, data compression, robotics, route planning, etc. Moreover, the test that defines an artificially intelligent system, called the Turing test, is also defined and detailed. Afterwards, the authors describe some AI tools such as fuzzy logic, genetic algorithms, and swarm intelligence. Special attention will be given to neural networks and fuzzy logic. The authors also present the future research directions and ethics.

Download Full-text

Forecasting Stock Returns with Artificial Neural Networks

Neural Networks in Business Forecasting ◽

10.4018/978-1-59140-176-6.ch003 ◽

2011 ◽

pp. 47-79 ◽

Cited By ~ 5

Author(s):

Suraphan Thawornwong ◽

David Enke

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Stock Returns ◽

Future Research ◽

Network Architectures ◽

Network Applications ◽

Artificial Neural ◽

Neural Network Applications

During the last few years there has been growing literature on applications of artificial neural networks to business and financial domains. In fact, a great deal of attention has been placed in the area of stock return forecasting. This is due to the fact that once artificial neural network applications are successful, monetary rewards will be substantial. Many studies have reported promising results in successfully applying various types of artificial neural network architectures for predicting stock returns. This chapter reviews and discusses various neural network research methodologies used in 45 journal articles that attempted to forecast stock returns. Modeling techniques and suggestions from the literature are also compiled and addressed. The results show that artificial neural networks are an emerging and promising computational technology that will continue to be a challenging tool for future research.

Download Full-text

Detection of Carolina Geranium (Geranium carolinianum) Growing in Competition with Strawberry Using Convolutional Neural Networks

Weed Science ◽

10.1017/wsc.2018.66 ◽

2018 ◽

Vol 67 (2) ◽

pp. 239-245 ◽

Cited By ~ 4

Author(s):

Shaun M. Sharpe ◽

Arnold W. Schumann ◽

Nathan S. Boyd

Keyword(s):

Neural Networks ◽

Image Classification ◽

Convolutional Neural Networks ◽

Future Research ◽

Study Objective ◽

Weed Interference ◽

Detection Technology ◽

Optimal Approach ◽

In Situ Detection

AbstractWeed interference during crop establishment is a serious concern for Florida strawberry [Fragaria×ananassa(Weston) Duchesne ex Rozier (pro sp.) [chiloensis×virginiana]] producers. In situ remote detection for precision herbicide application reduces both the risk of crop injury and herbicide inputs. Carolina geranium (Geranium carolinianumL.) is a widespread broadleaf weed within Florida strawberry production with sensitivity to clopyralid, the only available POST broadleaf herbicide.Geranium carolinianumleaf structure is distinct from that of the strawberry plant, which makes it an ideal candidate for pattern recognition in digital images via convolutional neural networks (CNNs). The study objective was to assess the precision of three CNNs in detectingG. carolinianum. Images ofG. carolinianumgrowing in competition with strawberry were gathered at four sites in Hillsborough County, FL. Three CNNs were compared, including object detection–based DetectNet, image classification–based VGGNet, and GoogLeNet. Two DetectNet networks were trained to detect either leaves or canopies ofG. carolinianum. Image classification using GoogLeNet and VGGNet was largely unsuccessful during validation with whole images (Fscore<0.02). CNN training using cropped images increasedG. carolinianumdetection during validation for VGGNet (Fscore=0.77) and GoogLeNet (Fscore=0.62). TheG. carolinianumleaf–trained DetectNet achieved the highestFscore(0.94) for plant detection during validation. Leaf-based detection led to more consistent detection ofG. carolinianumwithin the strawberry canopy and reduced recall-related errors encountered in canopy-based training. The smaller target of leaf-based DetectNet did increase false positives, but such errors can be overcome with additional training images for network desensitization training. DetectNet was the most viable CNN tested for image-based remote sensing ofG. carolinianumin competition with strawberry. Future research will identify the optimal approach for in situ detection and integrate the detection technology with a precision sprayer.

Download Full-text

Untangling Neural Nets

American Political Science Review ◽

10.1017/s0003055404001200 ◽

2004 ◽

Vol 98 (2) ◽

pp. 371-378 ◽

Cited By ~ 39

Author(s):

SCOTT DE MARCHI ◽

CHRISTOPHER GELPI ◽

JEFFREY D. GRYNAVISKI

Keyword(s):

Neural Network ◽

Neural Networks ◽

Logit Model ◽

Model Building ◽

Neural Nets ◽

Security Studies ◽

Future Research ◽

Linear Discriminant ◽

Nonparametric Approach ◽

Traditional Approaches

Beck, King, and Zeng (2000) offer both a sweeping critique of the quantitative security studies field and a bold new direction for future research. Despite important strengths in their work, we take issue with three aspects of their research: (1) the substance of the logit model they compare to their neural network, (2) the standards they use for assessing forecasts, and (3) the theoretical and model-building implications of the nonparametric approach represented by neural networks. We replicate and extend their analysis by estimating a more complete logit model and comparing it both to a neural network and to a linear discriminant analysis. Our work reveals that neural networks do not perform substantially better than either the logit or the linear discriminant estimators. Given this result, we argue that more traditional approaches should be relied upon due to their enhanced ability to test hypotheses.

Download Full-text

Computational Principles of Supervised Learning in the Cerebellum

Annual Review of Neuroscience ◽

10.1146/annurev-neuro-080317-061948 ◽

2018 ◽

Vol 41 (1) ◽

pp. 233-253 ◽

Cited By ~ 59

Author(s):

Jennifer L. Raymond ◽

Javier F. Medina

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Supervised Learning ◽

Brain Area ◽

Future Research ◽

Brain Areas ◽

Adaptive Mechanisms ◽

Networks Analysis ◽

Artificial Neural ◽

Organizational Principles

Supervised learning plays a key role in the operation of many biological and artificial neural networks. Analysis of the computations underlying supervised learning is facilitated by the relatively simple and uniform architecture of the cerebellum, a brain area that supports numerous motor, sensory, and cognitive functions. We highlight recent discoveries indicating that the cerebellum implements supervised learning using the following organizational principles: ( a) extensive preprocessing of input representations (i.e., feature engineering), ( b) massively recurrent circuit architecture, ( c) linear input–output computations, ( d) sophisticated instructive signals that can be regulated and are predictive, ( e) adaptive mechanisms of plasticity with multiple timescales, and ( f) task-specific hardware specializations. The principles emerging from studies of the cerebellum have striking parallels with those in other brain areas and in artificial neural networks, as well as some notable differences, which can inform future research on supervised learning and inspire next-generation machine-based algorithms.

Download Full-text

A STATE-OF-THE-ART REVIEW OF NEURAL NETWORKS FOR PERMEABILITY PREDICTION

The APPEA Journal ◽

10.1071/aj99019 ◽

2000 ◽

Vol 40 (1) ◽

pp. 341 ◽

Cited By ~ 9

Author(s):

A.G. Bruce ◽

P.M. Wong ◽

Y. Zhang ◽

H.A. Salisch ◽

C.C. Fung ◽

...

Keyword(s):

Neural Networks ◽

Oil And Gas ◽

State Of The Art ◽

Future Research ◽

Good Prediction ◽

New Era ◽

Multiple Networks ◽

Permeability Prediction ◽

Future Research Directions ◽

Oil And Gas Companies

This paper reviews the state-of-the-art of neural networks for permeability prediction from well logs. Good prediction of permeability is necessary for reservoir characterisation and is important for improving the reliability of the asset value of oil and gas companies. Two particular models, known as backpropagation and radial basis function networks, have been applied. From previous work, six innovative aspects are identified:choice of inputs;outlier detection and removal;data splitting;scaling;multiple networks; andprediction confidence.We have also provided a list of future research directions in the area, reflecting the current deficiencies of the use of neural networks. The topics are:the quality and quantity of core data;the maximum use of the logs;the compatibility of scales;the use of soft computing; andthe management of prediction confidence.The current applications are certainly the beginning of a new era. It is important for petrophysicists to take advantage of the advanced technologies.

Download Full-text

Research of Intelligent Control of Traffic Signal

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.644-650.107 ◽

2014 ◽

Vol 644-650 ◽

pp. 107-111

Author(s):

Xiang Li ◽

Jin Song Du ◽

Jing Tao Hu ◽

Xin Bi

Keyword(s):

Genetic Algorithm ◽

Neural Networks ◽

Particle Swarm Optimization ◽

Artificial Neural Networks ◽

Intelligent Control ◽

Traffic Signal ◽

Future Research ◽

Intelligent Algorithm ◽

Swarm Optimization ◽

Multi Agent

At present, in the field of intelligent control of traffic signal, most of scholars at home and abroad use fuzzy control and intelligent algorithm, such as genetic algorithm, ant colony optimization, particle swarm optimization, multi-agent, artificial neural networks, fuzzy method etc. This paper summarizes and analyzes these algorithms, points out the problems and shortcomings in the present research, puts forward the direction and trend in the future research. These works have certain directive significance to the research and development of intelligent control of traffic signal.

Download Full-text

FORECASTING FOREIGN EXCHANGE RATES WITH ARTIFICIAL NEURAL NETWORKS: A REVIEW

International Journal of Information Technology & Decision Making ◽

10.1142/s0219622004000969 ◽

2004 ◽

Vol 03 (01) ◽

pp. 145-165 ◽

Cited By ~ 76

Author(s):

WEI HUANG ◽

K. K. LAI ◽

Y. NAKAMORI ◽

SHOUYANG WANG

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Exchange Rates ◽

Network Architecture ◽

Future Research ◽

Promising Alternative ◽

Practical Applications ◽

Future Research Directions ◽

Artificial Neural ◽

Input Variables

Forecasting exchange rates is an important financial problem that is receiving increasing attention especially because of its difficulty and practical applications. Artificial neural networks (ANNs) have been widely used as a promising alternative approach for a forecasting task because of several distinguished features. Research efforts on ANNs for forecasting exchange rates are considerable. In this paper, we attempt to provide a survey of research in this area. Several design factors significantly impact the accuracy of neural network forecasts. These factors include the selection of input variables, preparing data, and network architecture. There is no consensus about the factors. In different cases, various decisions have their own effectiveness. We also describe the integration of ANNs with other methods and report the comparison between performances of ANNs and those of other forecasting methods, and finding mixed results. Finally, the future research directions in this area are discussed.

Download Full-text