scispace - formally typeset
Search or ask a question

What is learning in artificial neural network? 

Artificial neural network
Deep learning
Time delay neural network
Types of artificial neural networks
Converters

Answers from top 10 papers

More filters
Papers (10)Insight
Journal Article•DOI
Artificial neural networks to support petrographic classification of carbonate-siliciclastic rocks using well logs and textural information
Adrielle Silva, Irineu Lima Neto, Roseane M. Misságia, Marco Ceia, Abel G. Carrasquilla, Nathaly L. Archilha 
01 Jun 2015-Journal of Applied Geophysics
45 Citations
The artificial neural network (ANN) is a computational method based on human brain function and is efficient in recognizing previously trained patterns.
Open access•Journal Article•DOI
An Overview of Neural Network
Mohaiminul Islam, Guorong Chen, Shangzhu Jin 
29 Jun 2019
26 Citations
The artificial neural network learns by updating the network architecture and connection weights so that the network can efficiently perform a task.
Journal Article•DOI
Knowledge Incorporation into Neural Networks From Fuzzy Rules
Yaochu Jin, Bernhard Sendhoff 
01 Dec 1999-Neural Processing Letters
26 Citations
The incorporation of prior knowledge into neural networks can improve neural network learning in several respects, for example, a faster learning speed and better generalization ability.
Open access•Journal Article•DOI
Software Development Effort and Cost Estimation: Neuro-Fuzzy Model
Divya Kashyap, Ashish Tripathi, A. K. Misra 
01 Jan 2012-IOSR Journal of Computer Engineering
10 Citations
Artificial Neural Network has the ability to learn from previous data.
Open access•Journal Article•DOI
Application of artificial neural networks to predict the deflections of reinforced concrete beams
Mateusz Kaczmarek, Agnieszka Szymanska 
01 Jun 2016-Studia Geotechnica et Mechanica
16 Citations
Artificial neural network is a versatile modelling tool capable of making predictions of values that are difficult to obtain in numerical analysis.
Open access•Posted Content
Continual learning: A comparative study on how to defy forgetting in classification tasks.
Matthias De Lange, Rahaf Aljundi, Marc Masana, Sarah Parisot, Xu Jia, Ales Leonardis, Gregory G. Slabaugh, Tinne Tuytelaars 
18 Sep 2019
187 Citations
Artificial neural networks thrive in solving the classification problem for a particular rigid task, where the network resembles a static entity of knowledge, acquired through generalized learning behaviour from a distinct training phase.
Proceedings Article•DOI
Characterization of artificial neural network algorithms
T. Baker, Dan Hammerstrom 
08 May 1989
26 Citations
One popular artificial neural network model, the back-propagation algorithm, promises to be a powerful and flexible learning model.
Open access•Journal Article
A virtual laboratory for neural network controlled DC motors based on a DC-DC buck converter
Okan Bingöl, Serdar Paçaci 
01 Jan 2012-International Journal of Engineering Education
38 Citations
This training set can also guarantee effective learning andcomprehension of Artificial Neural Networks (ANN).
Journal Article•DOI
An educational tool for artificial neural networks
Omer Deperlioglu, Utku Kose 
01 May 2011-Computers & Electrical Engineering
32 Citations
By using these tools, users can also understand and learn working mechanism of a typical artificial neural network, using features of different models and related learning algorithms.
Open access•Posted Content
A differentiable measure of pointwise shared information.
Abdullah Makkeh, Aaron J. Gutknecht, Michael Wibral 
09 Feb 2020
8 Citations
Thus, it lends itself for example to local learning in artificial neural networks.

Related Questions

What is Artificial Neural Network?5 answersAn Artificial Neural Network (ANN) is an information processing paradigm inspired by the way biological nervous systems, such as the brain, process data. It consists of interconnected processing elements called neurons that work together to solve problems. ANNs learn from examples and can be trained for specific applications like pattern recognition or data classification. They mimic the behavior of the brain by changing the synaptic connections between neurons. ANNs are a branch of Artificial Intelligence (AI) and are considered a simplified model of the human brain. They aim to learn tasks by imitating the brain's behavior, where neurons communicate with each other through synapses. ANNs have various advantages and applications, and their training is often done using the backpropagation algorithm.
What is artificial neural network?4 answersAn artificial neural network (ANN) is an information processing paradigm based on how biological nervous systems process information. It consists of interconnected processing elements called neurons that work together to solve specific problems. ANNs learn from examples through a learning process, similar to humans. The neuronal synaptic connections in ANNs need to be changed for learning, just like in biological systems. ANNs have advantages such as pattern recognition and data classification. They have been successfully applied in various fields including computer vision, robotics, speech recognition, and natural language processing. ANNs can solve prediction and classification problems using algorithms like back-propagation, which adjusts the connection weights between neurons. They can also be used for data analysis and improving management in domains like fitness clubs.
What is Learning?5 answersLearning is a process of gaining knowledge, improving skills, developing attitudes, and producing personality. It involves the acquisition of knowledge through memory and repetition, as well as the application of that knowledge in real-life situations. Learning is a complex concept that encompasses domain-specific learning, individual differences, and cultural dimensions. It is a process that results in a relatively stable modification in stimulus-response relations through functional environmental interaction. Learning can be facilitated through interactive and fun activities, such as playing and using technology, which engage students and make the learning experience enjoyable. Overall, learning is an essential activity that individuals undertake to acquire knowledge, develop skills, and adapt to various environments in order to achieve independence and succeed in life.
What is learning in psychology?3 answersLearning in psychology refers to the process by which behavior is modified through practice and training. It is influenced by various factors such as environment, motivation, health, emotions, maturation, and aging. Learning involves obtaining, presenting, and transforming information into knowledge, which then guides human attitudes and behavior. It is a process of active interaction with the environment that leads to relatively permanent changes in knowledge, understanding, skills, and attitudes. Learning is a cognitive-psychological process that occurs through the interaction of an animal with its environment, resulting in a modification of subsequent behavior. The ability to learn is considered one of the most important features of intelligence, and it is a process through which knowledge and abilities are acquired through experience. Learning is not limited to academic subjects but also includes skills, interests, perceptions, and habits, which are influenced by internal and external factors.
Whats artificial neural networks?5 answersArtificial neural networks (ANNs) are computational methods that aim to mimic the behavior of the human brain. ANNs consist of interconnected neurons that communicate with each other through synapses. They learn by receiving stimuli and emitting signals, or "firing". ANNs have evolved since their inception in 1986 and have become a powerful tool in various fields such as computer vision, robotics, speech recognition, and natural language processing. They can approximate any computable function and are trained using labeled training data through the backpropagation algorithm, which is a form of gradient descent. In the context of lung cancer research, ANNs are used as efficient models for nonlinear regression or classification, providing valuable support for diagnosis and treatment decisions. ANNs have also been successful in applications such as system identification and control, face recognition, and solving nonlinear phenomena.
What is training in artificial neural network?10 answers

See what other people are reading

What commercial games are actually using reinforcement learning?
4 answers
Commercial games like Ubisoft's Roller Championsand multiplayer deathmatch first-person shooters utilizing platforms such as ViZDoom, DeepMind Lab, and Unity’s ML-Agentshave incorporated reinforcement learning (RL) for Artificial Intelligence (AI) design. In Roller Champions, RL systems are employed to train AI models quickly to adapt to gameplay changes, develop sophisticated strategies, and assist in game balance. On the other hand, in multiplayer deathmatch games, RL agents are created using various platforms and training architectures to compete against humans, potentially enhancing gameplay compared to traditional AI. These examples showcase the practical application of RL in modern video game development, demonstrating its effectiveness in enhancing AI capabilities and overall gaming experience.How can information geometry be used in machine learning?
5 answers
Information geometry, a field focusing on the geometric structures of probability distributions, plays a crucial role in machine learning applications. By treating parametric families of distributions as Riemannian manifolds equipped with metrics like the Fisher metric tensor, information geometry provides a framework for studying statistical models and their applications in various information sciences. Machine learning leverages information geometry concepts to analyze and process data efficiently. For instance, machine learning algorithms have been successfully applied to study mathematical structures, formulate conjectures, and aid in problem-solving by utilizing artificial intelligence techniques like neural networks and support vector machines. Additionally, information geometry can offer insights into skill discovery algorithms in reinforcement learning, showcasing how geometric perspectives can enhance learning processes in AI systems.What is the most popular paper in computer vision in 2024?
4 answers
The most popular paper in computer vision in 2024 is likely to focus on the advancements in artificial intelligence (AI) and its applications in various fields. Papers discussing the intersection of AI, deep learning, and neural networks with computer vision are gaining significant attention. Specifically, research on interpretable and robust machine learning methods, such as causal inference approaches, is becoming increasingly important for the practical deployment of AI algorithms. Additionally, papers exploring the potential of software-defined video analytics systems with cross-camera collaboration to overcome current limitations in video analytics are also gaining traction. Furthermore, studies on object tracking using convolutional neural networks are at the forefront, addressing challenges like camera motion, occlusion, and illumination effects. These areas are likely to be prominent in the most popular computer vision paper of 2024.How accurate are economic forecasts generated using BI technologies compared to traditional methods?
5 answers
Economic forecasts generated using Business Intelligence (BI) technologies exhibit high accuracy levels compared to traditional methods. BI technologies leverage big data, artificial intelligence, and machine learning to enhance forecasting precision. Traditional methods, reliant on limited sample data and human consciousness, often result in lower prediction accuracy. BI technologies, such as neural networks with different weights forecasting, offer improved accuracy by considering the varying importance of different factors in economic forecasting. Additionally, the integration of deep learning and big data analytics in economic forecasting models based on artificial intelligence methods has shown accuracy rates exceeding 90%. This shift towards BI technologies in economic forecasting signifies a significant advancement in prediction accuracy and overall forecasting reliability.Is there any machine learning study (not deep learning) using ECG and genetic for cardiovascular events prediction?
5 answers
Machine learning studies have been conducted for cardiovascular event prediction using ECG data. One study utilized machine learning algorithms like decision trees, support vector machines, k-nearest neighbors, Random Forest, and XGBoost to predict heart disease based on patient attributes and medical history, achieving an accuracy of 99.83%. Another research focused on predicting coronary artery disease using machine learning techniques such as random forest, decision tree, and gradient-boosted tree, with the gradient-boosted tree algorithm outperforming other methods. Additionally, a study employed Extreme Gradient Boosting, Random Forest, CART, Extra Tree Classifier, and Gradient Boosting Machine to predict cardiovascular disease, with the Extreme Gradient Boosting model achieving the highest accuracy of 91.9%. These studies showcase the effectiveness of machine learning in predicting cardiovascular events using ECG data and patient attributes.How bangladesh can overcome from global warming?
5 answers
Bangladesh, being severely impacted by global warming, can overcome this challenge through various strategies. Firstly, transitioning to renewable energy sources like solar and wind power can significantly reduce carbon emissions. Implementing climate-smart agricultural practices such as organic farming, crop rotation, and integrated pest management can enhance food production resilience to climate change. Additionally, raising awareness about climate change and its impacts is crucial for effective mitigation and adaptation efforts. Developing long-term adaptation plans, focusing on sustainable practices and biodiversity conservation, can help Bangladesh combat the adverse effects of global warming and build a more resilient future. By integrating these approaches and policies, Bangladesh can work towards mitigating the impacts of global warming and fostering a sustainable environment for future generations.In research on AI prediction of student performance, how to collection of data sets?
5 answers
In the realm of AI-driven prediction of student performance, the collection of datasets is a critical step that involves gathering a wide range of data types from various sources. The methodologies for data collection as highlighted in the research contexts include: 1. Collecting demographic and clickstream data from students interacting with Virtual Learning Environments (VLEs) to analyze and predict performance, emphasizing the importance of combining different data types for a comprehensive analysis. 2. Generating datasets from learning management systems like Blackboard, focusing on factors such as study mode (face-to-face or virtual), attendance at live lectures, midterm exam scores, and solved assessments to predict performance within a blended learning environment. 3. Utilizing data from secondary schools, including student background, habits, absenteeism, and social activities, augmented by merging datasets from different schools to increase sample size and diversity for a more accurate prediction. 4. Employing educational data mining techniques on data generated through e-learning platforms, e-admission systems, and automated result management systems, highlighting the need for automated techniques to process this data for performance prediction. 5. Predicting performance based on pre-admission academic achievements, demographics, and first-semester performance, addressing the challenge of imbalanced data through synthetic minority oversampling techniques. 6. Using neural networks to predict performance based on data collected specifically for the purpose of addressing student orientation and performance issues. 7. Automatically extracting features from multi-source heterogeneous behavior data of students to predict academic performance, showcasing the shift towards using deep learning models for feature extraction. 8. Generating synthetic data samples using improved conditional generative adversarial networks (CGAN) to address the issue of small dataset sizes in educational datasets. 9. Collecting data through online surveys and historical records of students’ grades and retention, utilizing a fuzzy inference system (FIS) for prediction, which underscores the use of self-reported learning strategies and historical performance data. These methodologies demonstrate a diverse approach to data collection, ranging from leveraging existing educational platforms and systems to creating synthetic data and utilizing self-reported information, all aimed at enhancing the accuracy and effectiveness of AI in predicting student performance.How correct maintenance prediction should be happened for elevator industry?
5 answers
Correct maintenance prediction in the elevator industry should leverage the advancements in machine learning and deep learning technologies to ensure safety, efficiency, and cost-effectiveness. Predictive maintenance, as opposed to traditional preventive maintenance, utilizes data-driven insights to predict and address potential failures before they occur. This approach is grounded in the collection and analysis of elevator sensor data, which, when processed through sophisticated algorithms, can forecast maintenance needs with high accuracy. The implementation of random forest classifiers and deep learning algorithms, such as LSTM (Long Short-Term Memory networks), has shown promising results in predicting elevator faults by analyzing sensor data recorded from maintenance actions. These methods outperform traditional feature engineering techniques by automatically extracting relevant features from the data, thus offering a more scalable and adaptable solution across different elevator systems. Moreover, the integration of IoT (Internet of Things) technologies enables real-time monitoring and analysis of elevator conditions. By employing edge computing devices that collect and process data directly on the elevator, maintenance predictions can be made more rapidly and accurately. This approach is complemented by the use of neural networks and big data platforms that build predictive models based on the elevator's operational data, further enhancing the prediction accuracy. The application of RBF (Radial Basis Function) neural networks and the construction of knowledge graphs also contribute to the refinement of maintenance prediction methods. These techniques allow for the identification of complex patterns and relationships within the data, facilitating a more nuanced understanding of elevator behavior and potential failure modes. In summary, correct maintenance prediction in the elevator industry should incorporate a combination of machine learning algorithms, deep learning techniques, IoT technologies, and data analytics platforms. This integrated approach enables the timely prediction of elevator faults, thereby improving safety, reducing downtime, and optimizing maintenance resources.What are the negative impact of Advanced airport infrastructure and ground handling technologies in terms of Accessibility for Passengers?
4 answers
Advanced airport infrastructure and ground handling technologies can have negative impacts on accessibility for passengers with special assistance needs. The complexity of airport ground handling operations, modeled as a multiple-fleet vehicle routing problem, can lead to challenges in providing efficient and high-quality solutions for passenger access. Additionally, the rapid growth in air passenger traffic puts pressure on ground access networks and airports, impacting accessibility. Passenger queue congestion in airport public transport systems can further exacerbate accessibility issues, leading to poor travel experiences and unexpected time costs. These factors collectively highlight the need for innovative strategies to address accessibility challenges posed by advanced airport infrastructure and ground handling technologies.How does the use of multi-objective technology impact the optimization process in electrical machines?
4 answers
The utilization of multi-objective technology in the optimization process of electrical machines significantly enhances predictive performance and computational efficiency. By leveraging correlated sub-objectives through methods like chained regression or multi-output artificial neural networks (ANNs), higher accuracy in predicting targets is achieved compared to separate predictions. Additionally, the application of a variational auto-encoder (VAE) allows for concurrent parametric multi-objective technology optimization for different machine technologies simultaneously, leading to efficient design optimization in a high-dimensional space. Furthermore, employing a hybrid data-and physics-driven model with deep neural networks (DNNs) as meta-models enables the prediction of key performance indicators (KPIs) while maintaining or even improving the quality of results compared to conventional finite element (FE) simulation-based optimization, all at a reduced computational cost.How do deep learning models impact the efficiency of decision-making in maintenance-critical environments?
5 answers
Deep learning models significantly enhance decision-making efficiency in maintenance-critical environments by enabling accurate predictions and proactive maintenance strategies. These models, such as deep recurrent neural networks (RNNs), CNN-GRU networks, and VGGNet16, leverage historical data and sensor information to forecast equipment performance, remaining useful life, and identify maintenance needs. By simulating maintenance scenarios and optimizing resource allocation through deep reinforcement learning (DRL) frameworks, these models outperform human decisions, leading to reduced downtime, lower costs, and improved system reliability. The application of deep learning in maintenance management not only streamlines operations but also empowers stakeholders to make informed decisions promptly, ultimately enhancing collaboration and overall maintenance effectiveness.