Deep Learning ventures into territory associated with Artificial Intelligence. This course will demonstrate how neural networks can improve practice in various disciplines, with examples drawn primarily from financial engineering. Students will gain an understanding of deep learning techniques, including how alternate data sources such as images and text can advance practice within finance.
Deep Learning and Neural Networks for Financial Engineering
New York University (NYU) via edX Professional Education
Overview
Syllabus
Week 0: Classical Machine Learning: Overview
-
Guided entry for students who have not taken the first course in the series
-
Notational conventions
-
Basic ideas: linear regression, classification
-
Recipe for Machine Learning
Week 1: Introduction to Neural Networks and Deep Learning
-
Neural Networks Overview
-
Coding Neural Networks: Tensorflow, Keras
-
Practical Colab
Week 2 : Convolutional Neural Networks
-
A neural network is a Universal Function Approximator
-
Convolutional Neural Networks (CNN): Introduction
-
CNN: Multiple input/output features
-
CNN: Space and time
Week 3: Recurrent Neural Networks
-
Recurrent Neural Networks (RNN): Introduction
-
RNN Overview
-
Generating text with an RNN
Week 4: Training Neural Networks
-
Back propagation
-
Vanishing and exploding gradients
-
Initializing and maintaining weights
-
Improving trainability
-
How big should my Neural Network be ?
Week 5: Interpretation and Transfer Learning
-
Interpretation: Preview
-
Transfer Learning
-
Tensors, Matrix Gradients
Week 6: Advanced Recurrent Architectures
-
Gradients of an RNN
-
RNN Gradients that vanish and explode
-
Residual connections
-
Neural Programming
-
LSTM
-
Attention: introduction
Week 7: Advanced topics
-
Neural Language Processing (NLP)
-
Interpretation: what is going on inside a Neural Network
-
Attention
-
Adversarial examples
-
Final words
Taught by
Ken Perry
