LogicMojo Machine Learning

Visual, step-by-step lessons covering the foundations of classical machine learning, neural networks, forecasting, and interview preparation.

← All teaching

InstructorNitin Nilesh
CourseLogicMojo 2026
LevelFoundations to intermediate
FormatInteractive webpages

This course is organized as a sequence of visual teaching packs. The explanations emphasize intuition first, then connect that intuition to equations, algorithms, evaluation choices, and practical modeling decisions.

The current release contains browser-based course pages only. Notebooks, datasets, assignments, recordings, and downloadable course archives are intentionally excluded and may be added separately later.

Course materials

Explore the topics

Open a topic to enter its interactive lesson pack. Use the overview and numbered pages inside each pack to move through the material.

  1. 1 Linear Regression Build intuition for lines, residuals, loss functions, OLS, gradient descent, evaluation, and multiple regression.
  2. 2 Polynomial Regression Move beyond straight lines while understanding generalization, bias–variance tradeoffs, validation, and regularization.
  3. 3 Logistic Regression Connect the sigmoid, decision boundaries, log loss, odds, regularization, metrics, and class imbalance.
  4. 4 K-Nearest Neighbors Learn instance-based prediction, distance measures, choosing k, scaling, and KNN-based imputation.
  5. 5 Principal Component Analysis See covariance, eigenvectors, projection, explained variance, and practical dimensionality reduction geometrically.
  6. 6 Naive Bayes Develop Bayesian intuition and apply it to text classification, smoothing, log-space computation, and model variants.
  7. 7 Support Vector Machines Understand maximum-margin classification, soft margins, kernels, practical tuning, and support-vector regression.
  8. 8 Decision Trees Follow a tree from impurity and information gain through recursion, pruning, evaluation, and regression.
  9. 9 Random Forests Connect bootstrap sampling, bagging, feature randomness, out-of-bag evaluation, tuning, and interpretation.
  10. 10 Gradient Boosting Build boosting from weak learners through residual fitting, gradient intuition, and classification.
  11. 11 Model Evaluation Choose useful classification and regression metrics, thresholds, ROC and PR curves, and evaluation workflows.
  12. 12 Time-Series Forecasting Work from forecasting contracts and baselines through lags, stationarity, smoothing, ARIMA, validation, and ML workflows.
  13. 13 Neural-Network Foundations Trace neurons, activations, forward passes, losses, backpropagation, and parameter updates from first principles.
  14. 14 ML Interview Preparation Prepare across ML revision, question answering, system design, resumes, job search, outreach, mocks, and coding.