Welcome to the "General Machine Learning" course! 🎉 Whether you're a researcher, student, or data enthusiast, this course will equip you with essential machine learning concepts, from foundational principles to advanced evaluation techniques.
- Dates: May 21, 2025 – Juli 04, 2025
- Time: Wednesdays, 15:00 – 17:00
- Location: Virtual (Zoom link provided)
- Prerequisites: Basic Python & statistics knowledge recommended
This course covers core ML concepts and practical techniques to build robust models. By the end, you'll:
- 📊 Understand ML fundamentals – features vs. targets, linear/non-linear models, and hyperparameter tuning
- ⚖️ Master model evaluation – train-test splits, cross-validation, and metrics for performance assessment
- 🔍 Reduce dimensionality – tackle the curse of dimensionality with PCA and feature selection
- 🤖 Interpret models – analyze biases, SHAP/LIME explanations, and model's weights
- 🚫 Prevent data leakage – identify and mitigate leakage types for reliable models
⚠️ Handle imbalanced data – address class imbalance, missing values, and confounds
- Core principles, real-world applications, and feature/target relationships
- Linear vs. non-linear models and hyperparameter optimization
- Train-test splits, K-fold CV, nested CV, and stratification
- Evaluation metrics for different contexts
- Curse of dimensionality, feature selection (reverse/forward), and PCA
- Data normalization for scalable performance
- Analyzing SVM weights, detecting biases, and SHAP/LIME explanations
- Types of leakage (test-to-train, feature-to-target) and mitigation strategies
- Handling imbalanced data, missing values, and confounds
🔗 Full Program: Google Doc
📂 Slides & Materials: Sciebo
👨💻👩💻 See you in class! Happy learning! 👨💻👩💻