Machine Learning Basics

Join us for an hour Zoom as Darren Sargent shares the basics of Machine Learning. Darren has a rich background of experience as a Software Developer.

We will walk through a basic machine learning example that uses the "MNIST" dataset, containing images of many thousands of handwritten digits and associated labels. Using Python 3, Keras, Tensorflow and Numpy, we will build a simple neural network with hidden layers, which we will then train on the MNIST images dataset, and measure its accuracy. We will try having the network make a few predictions manually to get a feel for how it works. We'll look at how to save the model so it can be reloaded at some later date. We will then try tweaking some of the hyperparameters of the neural network and retraining it to see what the effect has on the way the network learns and its accuracy. Following the demonstration we will have time for questions and discussion.
Optional, if we have time: brief discussion of evolutionary algorithms. How might we tackle this same problem with an evolutionary algorithm?

We'll have a 30-minute demo followed by 20 minutes of Q&A. We'll wrap the last 10 minutes and ask for any follow-ups or if anyone else would like to share their projects.

Hope you can join us!
Knowledge Mavens