- Neural networks mimic the behavior of the human brain so computers can recognize patterns and solve problems.
- Types include feedforward networks, multilayer perceptrons, convolutional and recurrent networks, LSTMs and sequence-to-sequence models.
- Neural networks model nonlinear, complex relationships between inputs and outputs, including highly volatile data such as financial time series.
- A neural network with multiple hidden layers is called a deep neural network, the basis of deep learning.
- Uses include fraud detection, medical diagnosis, logistics optimization and energy demand forecasting.
How does Neural Network work?
In the fields of AI, machine learning, and deep learning, neural networks mimic the behavior of the human brain, allowing computers to recognize patterns and solve problems.
What are the types of Neural Networks?
Below are the nine different types of neural networks:
- Feed Forward Neural Network
- Multilayer Perceptron
- Convolutional Neural Network
- Radial Basis Functional Neural Network
- Recurrent Neural Network
- LSTM (long short-term memory)
- Sequence to Sequence Models
- Modular Neural Network
- Perceptron
Why are Neural Networks important?
In real-life situations, neural networks can help people solve complex problems. As a result, they can learn and model the relationships between inputs and outputs that are nonlinear and complex. They can also make generalizations and inferences, and uncover hidden relationships, patterns, and predictions. Neural Networks also model highly volatile data (such as financial time series data) and variances needed to predict rare events (such as fraud detection). Neural networks can improve decision-making in areas such as:
- Credit card and Medicare fraud detection
- Optimization of logistics for transportation networks
- Medical and disease diagnosis
- Financial predictions for stock prices, currency, options, futures, bankruptcy, and bond ratings
- Electrical load and energy demand forecasting
- Chemical compound identification
Neural Network vs. Other Technologies & Methodologies
Neural network vs. machine learning
Machine Learning uses algorithms to analyze data, learn from the data, and discover patterns of interest. At the same time, an Artificial Neural Network is a set of algorithms used in machine learning for modeling data using graphs of neurons.
Neural networks vs. deep learning
A neural network with multiple hidden layers and multiple nodes in each layer is called a deep learning system or a deep neural network. Deep learning involves the development of algorithms that can be used to train and predict outcomes from complex data.
Neural networks vs. random forest
Random forest is an ensemble of decision trees, while neural networks learn through layers of connected nodes. Both are machine learning techniques; deep learning uses neural networks with many layers.
Neural network vs. decision tree
A neural network is an assembly of nodes that resembles the human brain, while a decision tree is a top-down approach to looking at data that is easy to follow.
Neural Network in H2O AI
Deep learning in open-source H2O-3 is a multilayer feedforward neural network trained with stochastic gradient descent and backpropagation. H2O-3 deep learning documentation.