Machine Learning, 3GIFA (First-Year Financial Engineering)

Teaching Assistant for the Machine Learning course in the Financial Engineering track (3GIFA, first year of the engineering cycle), February 2024 to April 2024.

Overview

Supported the delivery of a core Machine Learning course for first-year engineering students in the Financial Engineering track, bridging rigorous mathematical foundations with applied modeling for financial decision-making. The course emphasized end-to-end model development, from data preparation to evaluation and interpretation, with a focus on use cases relevant to banking, credit risk, and quantitative finance.

Responsibilities

• Led weekly recitation and hands-on laboratory sessions reinforcing lecture material on supervised and unsupervised learning.
• Designed and graded problem sets, programming assignments, and the final project, providing detailed written feedback to students.
• Mentored student teams on applied projects, including credit scoring, fraud detection, and time-series forecasting of financial indicators.
• Held regular office hours to support students with Python implementation, model debugging, and conceptual questions.

Topics Covered

• Supervised learning: linear and logistic regression, decision trees, ensemble methods (random forests, gradient boosting).
• Unsupervised learning: clustering and dimensionality reduction (PCA).
• Model evaluation, cross-validation, the bias-variance trade-off, and regularization.
• Practical workflow in Python using NumPy, pandas, and scikit-learn.

Outcomes

Helped students translate theoretical concepts into working models, strengthening their ability to apply machine learning to real financial engineering problems and to communicate results rigorously.