How to Determine the Optimal Learning Rate of Your Neural Network
One of the biggest challenges in building a deep learning model is choosing the right hyper-parameters. If the hyper-parameters aren’t ideal, the network may not be able to produce optimal results or development could be far more challenging. Perhaps the most difficult parameter to determine is... Read more
Guidelines for Choosing an Optimizer and Loss Functions When Training Neural Networks
There’s no one right way to train a neural network. These models serve various functions with multiple data sets, so what produces a high-performing model in one instance may not in another. As a result, effective training relies on a series of tools and strategies. Two... Read more
Neural Network Models Can Hide Malware, Research Shows
Neural networks are one of the most exciting developments in data science. These advanced AI models could revolutionize forecasting and fraud detection processes, and likely will before long. That makes the recent discovery that malware can hide within them all the more concerning. A group of... Read more
Variational Auto-Encoders for Customer Insight
Neural networks are sometimes perceived as super complicated. They’re not. The most attractive application, in my opinion, of neural networks for small and medium-sized businesses, is in customer segmentation, and in my upcoming workshop at ODSC East 2020, “Variational Auto-Encoders for Customer Insight,” I will show... Read more
Build a First Neural Network
Neural networks are weirdly good at translating languages and identifying dogs by breed, but they can be intimidating to get started with. In an effort to smooth this on-ramp, I created a neural network framework specifically for teaching and experimentation. It’s called Cottonwood and this notebook... Read more
Building a Custom Convolutional Neural Network in Keras
In this article, we’ll walk through building a custom convolutional neural network (CNN) to classify images without relying on pre-trained models. There are a number of popular pre-trained models (e.g. Inception, VGG16, ResNet50) out there that are helpful for overcoming sampling deficiencies; they have already been... Read more
Building Neural Networks with Perceptron, One Year Later — Part I
Around one year ago now, I started writing for Open Data Science after presenting neural networks with Perceptron at the ODSC conference. Since then, a lot has changed. People have found fascinating uses for the software, and also help contribute to it. In this series I’ll... Read more
Save a Neural Net, Use a Linear Model
Anyone that got their computer science degree in the past five years is probably familiar with this presentation: “We trained a neural network on this dataset over a two-day span to separate our data. Unfortunately, the results weren’t very good; we got around 60 percent accuracy.... Read more
Using Neural Networks, C-RNN-GAN to Generate Music
C-RNN-GAN is a continuous recurrent neural network with adversarial training that contains LSTM cells, therefore it works very well with continuous time series data. For example, music files! Structure of Discriminator and Generator in C-RNN-GAN (picture from original paper) How Does C-RNN-GAN work? C-RNN_GAN is a recurrent... Read more