Seeking Input: Ideal “Intro to ML” Syllabus
Hello everyone,
I’m in the process of designing an “Introduction to Machine Learning” syllabus and would appreciate your input. What would your ideal syllabus include?
Considerations:
- The syllabus should cover both supervised and unsupervised learning.
- Assume students have a reasonable understanding of linear algebra and statistics.
- Focus should be on ML algorithms and practical programming.
- The course duration is 20 hours.
I’m interested in hearing your thoughts on essential topics, structure, and any specific resources or methodologies you’d recommend. The aim is to create a comprehensive and effective learning path for beginners in machine learning.
Thank you in advance for your contributions!
You must log in or register to comment.
Hi, some suggestions based on course developed by us (removed some stuff to fit in 20 hours):
- Concept of learning: supervised, unsupervised, and semi-supervised.
- Regression: simple and multiple linear regression, polynomial regression, regularization, SVR, decision trees, random forest regression.
- Classification: Logistic regression, K-NN, SVM, kernel SVM, naive Bayes, decision trees, random forests.
- Dimensionality Reduction: PCA, UMAP.
- Clustering: k-means, hierarchical clustering.
- Feature selection and extraction.
- Evaluation and generalization: training set, test and validation, cross-validation, parameter tuning, grid search, XGBoost.
- Python notebooks and libraries for ML: numpy, pandas, matplotlib, seaborn, scikit-learn, scipy and others.
Hope it’s useful!