Machine Learning

A few months ago, I wrote a tutorial on how to classify images using Convolutional Neural Networks (specifically, VGG16) pre-trained on the ImageNet dataset with Python and the Keras deep learning library. The pre-trained networks inside of Keras are capable…

Last Updated on July 1, 2021 Today’s blog post is inspired from an email I received from Jason, a student at the University of Rochester. Jason is interested in building a custom object detector using the HOG + Linear SVM…

In the previous section, we discussed gradient descent, a first-order optimization algorithm that can be used to learn a set of classifier weights for parameterized learning. However, this “vanilla” implementation of gradient descent can be prohibitively slow to run on…

Nearly all of deep learning is powered by one very important algorithm: Stochastic Gradient Descent (SGD). — Goodfellow, Bengio, and Courville (2016) At this point, we have a strong understanding of the concept of parameterized learning. We previously discussed the…

If you’ve been following along with this series of blog posts, then you already know what a huge fan I am of Keras. Keras is a super powerful, easy to use Python library for building neural networks and deep learning…

Many strategies used in machine learning are explicitly designed to reduce the test error, possibly at the expense of increased training error. These strategies are collectively known as regularization. —Goodfellow, Bengio, and Courville (2016) In previous lessons, we discussed two…

Last week, we discussed Multi-class SVM loss; specifically, the hinge loss and squared hinge loss functions. A loss function, in the context of Machine Learning and Deep Learning, allows us to quantify how “good” or “bad” a given classification function…

A couple weeks ago,we discussed the concepts of both linear classification and parameterized learning. This type of learning allows us to take a set of input data and class labels, and actually learn a function that maps the input to…

In last week’s post, I introduced the k-NN machine learning algorithm which we then applied to the task of image classification. Using the k-NN algorithm, we obtained 57.58% classification accuracy on the Kaggle Dogs vs. Cats dataset challenge: The question…