Register to attend this workshop at the bottom of this page. Zoom link will be emailed after filling out the registration form.

Machine learning (ML) methods are widely adopted by the academia and industry for applications in science, engineering, healthcare, and humanities. One of the greatest advantages of ML is that they are pretty general and applicable for predictive analytics in many fields. For example, if someone learned how to apply ML methods to classify animal images, they can apply the same set of protocols in classifying birds or cars or motorboats. To apply ML for a specific problem, the practitioners don’t have to go through all the complex mathematics or a lot of statistics but they need to learn the best practices to train and validate the models.

In this workshop, after a brief overview, we will focus on doing hands-on training in applying ML models on various data types including image, text, and time series. We will work through the use cases of classification and regression problems and discuss where to apply supervised or unsupervised methods.

Objectives of the workshop

• Understand supervised and unsupervised methods
• Define metrics for classification vs regression
• Find out which features are important in a given dataset
• Learn to apply ML models such as Decision Trees, Random Forest, and Support Vector Machines
• Perform clustering and dimensionality reductions (PCA, t-sne, K-means, etc.)
• Search the parameter space – hyperparameter optimization

If you have questions or need help, please email Bala Desinghu.

Amarel account: Apply here as soon as possible. You must have an Amarel account set up before the workshop.

VPN setup: You have to be connected on Rutgers’ network or be on VPN to access Amarel resources.

SSH setup: Windows users must install an SSH client like PuTTY or MobaXterm. Alternatively, Windows 10 users can install the complete Windows Subsystem for Linux.

Register to attend the workshop: