Event space sponsored by: Stack Exchange
Refreshments sponsored by: Bloomberg
Special Note:A link to the tutorial materials will be provided 1 - 2 weeks before the event
Event Description
This is a hands-on introduction to neural networks using the Python library, Keras. No prior knowledge of neural networks is required, but basic knowledge of Python is required.
Topics covered:
Introduction to feedforward neural networks (nodes, layers, activation functions, how a neural network produces output, how a neural network learns)
Introduction to Keras
Overview of different activation functions
Regularization - L1 / L2, dropout, early stopping
Hands on application using the MNIST & CIFAR datasets
Introduction to convolutional neural networks
Improving the way networks learn (Optimizers,Weight initialization,Tips and tricks for building and training networks)
Optimizers (Why Stochastic Gradient Descent isnt perfect,Momentum,Brief overview of other options)
Installation
Please come with Keras + either Theano or Tensorflow installed and Python version 2.7 - 3.5. Installation instructions are here: https://keras.io/#installation
Event agenda
9:30 AM - 10 AM: ** arrive during this time if you have any installation questions **
10 AM - 1 PM : Part 1 - Intro to Feedforward Neural Networks, Intro to Keras
1 PM - 2 PM: lunch
2 PM - 5 PM: Part 2 - regularization, convolutional networks, improving how networks learn
Reading Preparation (optional)
No preparation is required but if you are keen to start learning then I suggest reading the Keras documentation and/or starting to read the Introduction to Neural Networks tutorial on my blog.
https://keras.io/
https://learningmachinelearning.org/2016/07/26/introduction-to-neural-networks/
About the Speaker
Laura Graesser is studying for an MS in computer science at NYU, focusing on machine learning. Laura is particularly interested in neural networks and their application to computer vision problems, cross-fertilization between computer vision and NLP, the representations perspective (machine learning as data transformation and representation), and the manifold hypothesis.
In her spare time, Laura enjoys dancing, listening to jazz, going to art exhibitions, and writing about machine learning.