Παρουσίαση/Προβολή

(LOXR528) -  ΓΕΩΡΓΙΟΣ ΧΑΛΑΜΑΝΔΑΡΗΣ

Περιγραφή Μαθήματος

The course is an introduction to machine learning techniques. Upon completion of the course, students can understand machine learning techniques and implement applications to solve financial problems using the appropriate Python libraries.

Content:

• Classification techniques
  • K-means,
  • Support Vector Machine,
  • naive Bayes classifier,
  • random forests
• Methods of generalized regression
  • Ridge
  • LASSO
  • LARS).
• Types of neural networks
  • Multilayer Perceptrons,
  • Convolutional NNs,
  • Recurrent NNs,
  • Self-Organized Maps,
  • Kernel Networks
• Characteristic applications
  • credit-scoring
  • algorithmic trading
  • portfolio management
  • fraud detection
• The distinction between supervised, unsupervised, and reinforced learning Python libraries (Tensorflow, Keras).

The course evaluation is conducted via a written examination (70%) and a compulsory project (30%). The project is a case study on machine-learning applications on financial data. The project is designed to test students on their programming and problem-solving abilities and their reporting/submitting written work backed up by computations/estimations.

The written examination is a combination of multiple-choice questions, open-ended questions, and problems to be solved.

Ημερομηνία δημιουργίας

Κυριακή, 4 Δεκεμβρίου 2022