Neural networks requires to apply several state-of-the-art techniques such as choice of activation function, or network design to push their … More

# Tag: neural networks

# How Vectorization Saves Life in Neural Networks

Developers tend to handle problems with conditional statements and loops. This is the number one topic of developers and data … More

# Convolutional Autoencoder: Clustering Images with Neural Networks

Previously, we’ve applied conventional autoencoder to handwritten digit database (MNIST). That approach was pretty. We can apply same model to … More

# Autoencoder: Neural Networks For Unsupervised Learning

Neural networks are like swiss army knifes. They can solve both classification and regression problems. Surprisingly, they can also contribute … More

# Handling Overfitting with Dropout in Neural Networks

Overfitting is trouble maker for neural networks. Designing too complex neural networks structure could cause overfitting. So, dropout is introduced to … More

# Facial Expression Recognition with Keras

Kaggle announced facial expression recognition challenge in 2013. Researchers are expected to create models to detect 7 different emotions from human … More

# Logarithm of Sigmoid As a Neural Networks Activation Function

Previously, we’ve reviewed sigmoid function as activation function for neural networks. Logarithm of sigmoid states it modified version. Unlike to … More

# Softsign as a Neural Networks Activation Function

Activation functions play pivotal role in neural networks. As an alternative to hyperbolic tangent, softsign is an activation function for … More

# Softmax as a Neural Networks Activation Function

In fact, convolutional neural networks popularize softmax so much as an activation function. However, softmax is not a traditional activation function. … More

# A Gentle Introduction to Convolutional Neural Networks

Convolutional neural networks (aka CNN and ConvNet) are modified version of traditional neural networks. These networks have wide and deep … More