Advanced Machine Learning Methods

Course details

This course dives deep into the advanced techniques and methodologies in Machine Learning (ML). Building on foundational knowledge of ML, the course explores state-of-the-art algorithms, ensemble methods, and optimization techniques. It prepares students for research and development roles in AI and ML, enabling them to tackle complex real-world problems.

Start Date

Spring 2026 Semester

Duration

14-15 weeks

Prerequisites

Basic understanding of AI, ML, and Statistics. "Introduction to Machine Learning" or equivalent course recommended.

Learning Objectives

Gain mastery over advanced machine learning algorithms.
Understand ensemble techniques and their applications.
Delve deep into optimization methods in ML.
Engage with current research in the field of ML.
Apply advanced ML methods to complex real-world challenges.

Key Features

Hands-on assignments and projects on real datasets.
Guest lectures from AI & ML industry experts.
Emphasis on practical implementation and optimization.
Discussion and analysis of current ML research.

Learning path

Unit 1: Advanced Regression Techniques (3)
Unit 2: Ensemble Methods (5)
Unit 3: Advanced Classification Techniques (4)
Unit 4: Optimization in ML (4)
Unit 5: Time Series and Sequence Analysis (4)
Unit 6: Unsupervised Deep Learning (4)
Unit 7: Advanced Topics in ML (4)

Topics covered

Ridge and Lasso Regression
Polynomial Regression
ElasticNet Regression

Introduction to Ensemble Techniques
Bagging and Boosting
Random Forests
AdaBoost and Gradient Boosting Machines
Stacking

Support Vector Machines
Kernel SVM
Neural Network Classifiers
Imbalanced Classes and Remedies

Gradient Descent Variants
Convex Optimization
Regularization Techniques
Hyperparameter Tuning

ARIMA and Prophet Models
LSTM and GRU
Attention Mechanisms
Time Series Forecasting with ML

Self Organizing Maps (SOM)
Autoencoders
Generative Adversarial Networks (GAN)
Variational Autoencoders

Reinforcement Learning Overview
Transfer Learning
Multi-task Learning
ML in Modern NLP and Vision Tasks

Grading:

Class Participation: 10%
Weekly Assignments and Problem Sets: 30%
Group Projects and Case Analyses: 20%
Mid-Term Examination: 20%
Final Examination: 20%

Reading Material: A blend of core textbooks, seminal research papers, and contemporary articles and case studies.

Assignments: Practical assignments on contemporary datasets, ensuring students get hands-on experience with the methods taught.

Faculty: A leading academic or industry expert in advanced ML methodologies.

Teaching Instructors: TBD