Course Introduction
Course overview and expectations.
- Workshop
- Python
- Presenter
- Dr. Babak Hossein Khalaj
Lectures
Ordered session timeline with workshops and presenters.
Course Introduction
Data-science in production and buisness contexts.
- Workshop
- NumPy and Pandas
- Presenter
- Dr. Babak Hossein Khalaj
Data Models
Foundations of data representation and modeling.
- Workshop
- Matplotlib and Seaborn
- Presenter
- Dr. Babak Hossein Khalaj
Databases
Relational databases, querying, and data acquisition.
- Workshop
- SQL
- Presenter
- Dr. Babak Hossein Khalaj
Data Visualization
Principles of exploratory and explanatory visualization.
- Presenter
- Dr. Babak Hossein Khalaj
ML Dataflow – Part 1: Data Processing
Data collection, cleaning, preprocessing, and feature preparation.
- Workshop
- Web Scraping
- Presenter
- Dr. Babak Hossein Khalaj
Probability and Statistics
Review on probability and statistics; based primarily on Tibshirani et al.
- Presenter
- Dr. Amir Hossein Saberi
Probability and Statistics
Probability, estimation, uncertainty, and elementary inference for statistical learning.
- Presenter
- Dr. Amir Hossein Saberi
Regression
Regression taught from a statistical-learning perspective, based primarily on Tibshirani et al.
- Workshop
- Scikit-learn
- Presenter
- Dr. Amir Hossein Saberi
Supervised Machine Learning
Classification, regularization, model selection, and related topics; based primarily on Tibshirani et al.
- Presenter
- Dr. Amir Hossein Saberi
Unsupervised and Semi-Supervised Learning
Clustering, dimensionality reduction, and unsupervised learning; semi-supervised learning included if time permits. Based primarily on Tibshirani et al.
- Presenter
- Dr. Amir Hossein Saberi
Causality
Causal concepts and related material from the original syllabus.
- Workshop
- NetworkX and Plotly
- Presenter
- Dr. Babak Hossein Khalaj
ML Dataflow – Part 2: Model Evaluation
Evaluation protocols, validation, metrics, and error analysis.
- Presenter
- Dr. Babak Hossein Khalaj
Introduction to Deep Learning: Neural Networks
Deep-learning foundations based on Sergey Levine’s course materials.
- Workshop
- PyTorch and TensorFlow
- Presenter
- Dr. Amir Hossein Saberi
Deep Learning: MLPs, Feed-Forward and Recurrent Networks
Core neural architectures and optimization, based on Sergey Levine’s course materials.
- Presenter
- Dr. Amir Hossein Saberi
Transformers: Foundations and Architectures
Replaces the original Classic NLP, Modern NLP Architecture, and LLM survey sequence; covers attention and transformer fundamentals.
- Presenter
- Dr. Amir Hossein Saberi
Building a Small Language Model
Hands-on construction of a small LLM, using Andrej Karpathy’s video lectures as the main practical reference.
- Presenter
- Dr. Amir Hossein Saberi
Generative Models
Foundations of generative modeling, based on the principal/original research papers.
- Presenter
- Dr. Amir Hossein Saberi
Diffusion Models
Diffusion and score-based generative modeling, based on the principal/original research papers.
- Presenter
- Dr. Amir Hossein Saberi
Modern Time-Series Modeling with Transformers
Moved after transformers and generative models to use the required neural-sequence foundations; based on recent transformer-based time-series papers.
- Presenter
- Dr. Amir Hossein Saberi
ML Dataflow – Part 3: Model Implementation
Implementation, reproducibility, and integration of trained models.
- Presenter
- Dr. Babak Hossein Khalaj
ML Dataflow – Part 4: Pipeline
End-to-end machine-learning pipelines.
- Presenter
- Dr. Babak Hossein Khalaj
ML Dataflow – Part 5: Monitoring
Deployment monitoring, drift, reliability, and lifecycle management.
- Presenter
- Dr. Babak Hossein Khalaj