FA691 Deep Learning for Finance


Course Catalog Description

Introduction

The application of Machine Learning (ML) and Artificial Intelligence (AI) to finance does not just focus on the knowledge of algorithms. While the understanding of the algorithms used is fundamental to the discipline, it is also necessary to understand the tradeoffs of each algorithm, how they scale when used in production, and how to explain the problem, solution, and field with people who are not technically proficient.
Prerequisite: Students must have taken FA590 or comparable introduction to machine learning methods


Campus Fall Spring Summer
On Campus X
Online X

Instructors

Professor Email Office
Zachary Feinstein
zfeinste@stevens.edu Babbio Center 628

More Information

Course Objectives

In this course, students will (generally):
• Be able to apply neural networks to financial problems
• Be able to evaluate the performance of different methods to determine the best method/hyperparameters
• Learn how to interpret results of machine learning in a financial context


Course Outcomes

  1. Create and apply neural network models for regression and classification tasks in finance.
  2. Evaluate performance of trained machine learning models
  3. Understand applicability of diverse techniques in deep learning

Course Resources

Textbooks

  1. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press.
  2. Dixon, M. F., Halperin, I., Bilokon, P. (2020). Machine Learning in Finance: From Theory to Practice. Germany: Springer International Publishing.
  3. Selected research papers to be made available on Canvas

Grading

Grading Policies

Weights
Attendance and Short Assignments 20%
Homeworks 35%
Project 30%
Presentation 15%
Total grade 100%

Lecture Outline

Week Topics Readings HW
1 Review of Supervised Machine Learning Dixon et al.: Chapter 1
Goodfellow et al.: Chapter 5		 Homework 1: Review of Python (assigned)
2 Feed-forward neural networks with applications in finance Dixon et al.: Chapter 4
Goodfellow et al.: Chapter 6		 Homework 1 (due)
Homework 2: Application of feed-forward neural networks (assigned)
Project (assigned)
3 Convolutional neural networks with applications in finance Dixon et al.: Chapter 8.5
Goodfellow et al.: Chapter 9		 Homework 2 (due)
Homework 3: Application of neural networks for financial data (assigned)
Project proposal due
4
Applications to Finance		 Research papers; Examples:
Clark et al. (2020): A machine learning efficient frontier
Golbayani et al. (2020): Application of deep neural networks to assess corporate credit rating
5 Recurrent neural networks & long short-term memory (LSTM) networks with applications to financial time series Dixon et al.: Chapter 8.2-8.4
Goodfellow et al.: Chapter 10		 Homework 3 (due)
Homework 4: Application of neural networks for time series (assigned)
6 Special topics (ex: generative adversarial networks (GANs) or reinforcement learning for deep hedging) Dependent on topic area chosen Homework 4 (due)
7 Presentations Project (due)