Machine Learning
Linear models, neural networks, clustering, and evaluation.
0/10 complete
- Gradient DescentThe optimization workhorse of machine learning — follow the slope downhill to minimize a loss.easy
- Linear & Logistic RegressionThe foundation of supervised learning — fitting lines and curves to make predictions.easy
- Supervised LearningLearn a mapping from labeled examples — features, training, and the overfitting trap.easy
- Neural NetworksLayers of artificial neurons that learn complex patterns from data — the engine of modern AI.medium
- Model EvaluationBeyond accuracy — precision, recall, F1-score, and how to tell if your model is actually good.easy
- Clustering & k-MeansUnsupervised learning — group unlabeled points by similarity with the k-means assign-and-move loop.easy
- Decision TreesCarve the feature space into regions with a flow-chart of yes/no splits chosen to purify the data.medium
- RegularizationTame overfitting by penalizing complexity — how L2 regularization smooths an over-eager fit.medium
- Dimensionality Reduction & PCACompress many features into a few by projecting onto the directions of greatest variance.medium
- Convolution & CNNsThe building block of computer vision — slide a small learnable kernel over an image to extract features.medium