Artificial Neural Networks are a fascinating computational approach modeled to react similar to a biological brain to solve problems. Neural networks are a very powerful tool that enable machines to teach and evolve themselves. You can utilize this power in almost any application. Don’t be intimidated by the equations and Greek symbols of this cutting-edge technology, I will guide you on how to slay this technology and make it bend to your will! In this session, we are going to demystify the presumed complexity of Neural Networks. No need for a PhD or mathematical background, after this overview, you will be discussing these concepts around the water cooler. There will be very little math and lots of coding. The goal is for everyone to become acquainted with Neural Network from a pragmatic standpoint.
In this session, we will take a closer look at artificial neural networks inspired by biological neurons. We will study how these neurons can be modeled in a digital counterpart. After a short introduction of the actor model, the communalities between neurons and how Actors work by building an asynchronous and reactive neural network will be demonstrated.
Just when you think it couldn’t get any better, I will show you how to employ the functional paradigm to leverage multicore machines and GPUs to make your neural network predictions infinitely faster through parallelism.
By the end of this talk, you will learn the basic concepts of Neural Network and how to apply functional concurrency to estimate future stock prices at smoking fast speeds…and perhaps get rich while practicing!