95661 - ADVANCED MACHINE LEARNING

Learning outcomes

The Advanced ML course is geared towards state of the art application of neural network to pricing and market risk problem. The studend will acquire a sound knowledge of the principles underlying Neural Networks and will be guided in a tour of the relevant literature concerning the exploitation of machine learning for pricing of highly exotic products and applications to market risk managment. Altough the approach demands very large scale computing facilities, impossible to be provided to the students, nonetheless students will learn how to design solutions to this type of problem and will gain hands on experience of the methodology on simpler and smaller toy models.

Course contents

Basic Text Analysis with Python
What is Text Mining
The NLTK (Natural Language Tool-Kit) package
Text Preprocessing
Lemmatization and Stemming
Measuring Word Frequency
Applied Text Analysis with Python
Text Vectorization
Frequency Vectors
One-Hot Encoding
Term Frequency-Inverse Document Frequency
Classification for Text Analysis
Clustering for Text Similarity
Clustering by Document Similarity
From Words to Vector
Computing Document Similarity

Deep Learning Pricing
Black & Scholes and Heston Model
Portfolios Sensitivities

Introduction to Genetic Algorithms

Readings/Bibliography

• John C. Hull, Machine Learning in Business, An Introduction to the World of Data Science, Amazon (2019)
  
• Paul Wilmott, Machine Learning, An Applied Mathematics Introduction, Panda Ohana Publishing (2019)
• Sebastian Raschka and Vahid Mirjalili, Python Machine Learning, Packt (2019)
• Francois Chollet, Deep Learning Python, Manning (2018)

Teaching methods

Lessons are based on slides and Jupyter Notebook, delivered online in advance on "Virtuale".

To take the course, it is mandatory to have attended the introductory machine learning course and the computational finance course (Prof. Pietro Rossi).

The modeling aspects relating to the pricing of derivatives will be deemed to be known to the students. 

Assessment methods

The final exam consists of a small project chosen by the student from a series of proposals that can include both aspects of Natural Language processing (eg Sentiment Analysis) or related to the pricing of derivatives and other topics covered during the course.

The project must be chosen in advance and described in a document (preferably a jupyter notebook) that will be discussed with the teacher during the examination.

Teaching tools

• Slides (power point/pdf)
• Selected literature
• Jupyter Notebook
• Python Code Snippet

Office hours

See the website of Giovanni Della Lunga