Speakers and Workshops
M "Em" Noelle
Master of Ceremony
Words can't describe Em. Artist, aerialist, and maker. Still somehow finds time to write code and guide product.
Designing Your Neural Networks: A Step By Step Walkthrough
Neural networks are powerful beasts that give you a lot of levers to tweak! The sheer size of customizations that they offer can be overwhelming to even seasoned practitioners. In this talk I’ll give you a framework for making smart decisions about your neural network architecture!
We’ll explore lots of different facets of neural networks in this talk, including how to setup a basic neural network (including choosing the number of hidden layers, hidden neurons, batch sizes etc.) We’ll learn about the role momentum and learning rates play in influencing model performance. And finally we’ll explore the problem of vanishing gradients and how to tackle it using non-saturating activation functions, BatchNorm, better weight initialization techniques and early stopping.
I hope this guide will serve as a good starting point in your adventures.
Easy-to-Use Composable Deep Learning Framework in Elixir
In this talk, we will discuss the current state of the Annex framework, including philosophy, capabilities, extensions, and intentions. We will discuss how to extend the capabilities of Annex by implementing the most basic unit of transformation in Annex, the Annex.Layer behavior. Using structs and modules that implement the Annex.Layer behavior we will compose a deep neural network capable of performing state-of-the-art machine learning classification tasks. We will discuss how Annex’s performance can be tuned by parallelizing work through Elixir processes and native code integration.
All workshops are hands on and you will almost always need a laptop!
TinyML on Arduino
This workshop will introduce you to the process of deploying machine learning (ML) models to microcontrollers (MCUs). An Arduino Nano 33 BLE Sense board will be used to record real-time sensor data. This data will then be used to develop and train an ML model "in the cloud" using TensorFlow. Finally, the ML model will be converted so that it can be run on the Arduino with real-time performance.
Prerequisites: Bring your computer. Arduino will supply the hardware kit. If your laptop only has USB C, be sure to bring an adapter as only USB A to micro USB cables will be provided.
Introduction to Rust and Image Processing
Rust, as you've heard by now, is a systems language pursuing the trifecta: safety, concurrency, and speed. This makes it well suited to machine learning and data science tasks. Rust experience will not be required, only existing programming experience.We'll spend half the session on a with an introduction to Rust and the rest using functional programming style to develop a machine learning primitive called Convolution which can be used for many tasks including edge detection, sharpening, blur and more.
Intro to NLP (Natural Language Processing)
In this workshop we will walk you though NLP using Python. Having Python 3 installed ahead of time will be super helpful but not necessary.
How to draw a ghost and other ways crayons can be fun
The marketing materials of the esp32 say it can do anything . . . but can it draw? By itself? We will see in this 2 hour hands on hackathon where we all learn together. Alternative title: "The one were where everyone gets a board"
Roy van de Water
EdgeAI: Training and classifying imagery on a Raspberry Pi camera
Using a TensorFlow Convolutional Neural Network (CNN) combined with a Raspberry Pi 4 and camera, participants will train a custom network to detect images in the camera and turn them into actions for a video game. Everyone will compete in a multiplayer capture the flag game, by reducing heavy imagery rapidly on-device into simple game commands sent over a network.
James W. Clohessy
Study Jam: Machine Learning and Data
Big data, machine learning, and artificial intelligence are today’s hot computing topics, but these fields are quite specialized and introductory material is hard to come by. Fortunately, GCP provides user-friendly services in these areas and Qwiklabs has you covered with this introductory-level quest, so you can take your first steps with tools like Big Query, Cloud Speech API, and Cloud ML Engine.
Edge ML: Speaking on the Edge
In this workshop we will deploy a machine learning model to the SparkFun Edge, a microcontroller designed by Google and SparkFun to help developers experiment with ML on tiny devices. The model can recognize basic commands from speech while only being 20kb in size. After successfully deploying the basic model, we will tinker with the code to change its behavior and respond to different commands.