Free coursebeginner Registration open

Introduction to Machine Learning

A practical foundation in machine-learning problem framing, data preparation, modelling, validation, evaluation, interpretation and responsible deployment.

Provider
AISynapseBiz
Instructor
Self-paced · no live professor
Duration
6 weeks · 18+ hours video · self-paced
Certificate
Course Certificate

Choose an offering

Delivery
Recorded / self-paced
Schedule
Self-paced · 6 weeks · finish whenever you are ready
Timezone
UTC
Dates
Rolling
Course overview

About this course

A professional beginner course that teaches the full machine-learning workflow rather than isolated algorithms. Learners begin with problem framing and data splits, then study regression, classification, feature engineering, regularization, trees and ensembles, unsupervised learning, validation, metrics, explainability, leakage, drift and deployment. Original AISynapse lessons are paired with checkpoint questions and AI-graded applied work so learners practise evidence-based modelling decisions.

Learning outcomes

Frame supervised and unsupervised learning problems correctly; separate training, validation and test data; build and interpret regression and classification baselines; select appropriate metrics; recognize overfitting, leakage and class imbalance; understand regularization, trees and ensemble models; create defensible feature pipelines; use cross-validation appropriately; interpret feature importance and model explanations; describe monitoring and drift; present a complete, responsible ML workflow.

Prerequisites

  • ▸Basic Python familiarity is recommended.
  • ▸Learners should understand variables, functions and simple tabular data operations; no prior machine-learning knowledge is required.

Weekly outline

  • ▸Week 1: ML Workflow and Regression Foundations
  • ▸Week 2: Classification and Data Preparation
  • ▸Week 3: Generalisation, Regularisation and Tree Models
  • ▸Week 4: Unsupervised Learning, Validation and Model Selection
  • ▸Week 5: Imbalanced Decisions, Interpretability and Explainability
  • ▸Week 6: Deployment, Monitoring and Professional ML Capstone
Structured weekly sequence

Curriculum

Open a week to inspect its lessons and applied focus. The current foundation pathway contains only the six active curriculum weeks; historical versions remain preserved privately for existing learner progress.

01

Week 1

ML Workflow and Regression Foundations

13 lessons3 checkpoints1 applied assignment

Frame a machine-learning problem correctly and build regression intuition from features, targets, data splits, baselines, residuals and metrics.

Applied work for this week

Complete the learning path and submit the applied task to AI SYNAPSE REVIEW.

  1. 01What machine learning is / Prediction versus explanation
  2. 01Week 1: The Machine Learning Workflow
  3. 02Problem framing / Features and targets
  4. 03Unit of observation / Training data
  5. 04Validation data / Test data
  6. 05Baselines / Loss and metrics
  7. 06Reproducibility / End-to-end workflow
  8. 07Continuous targets / Linear regression intuition
  9. 08Coefficients and predictions / Residuals
  10. 09Mean squared error / Mean absolute error
  11. 10R-squared intuition / Scaling considerations
  12. 11Nonlinearity / Outliers and sensitivity
  13. 12Regression diagnostics / Building a baseline regression
02

Week 2

Classification and Data Preparation

13 lessons3 checkpoints1 applied assignment

Build classification intuition and prepare data safely with encoding, scaling, transformations, leakage control and reproducible pipelines.

Applied work for this week

Complete the learning path and submit the applied task to AI SYNAPSE REVIEW.

  1. 01Week 2: Regression
  2. 01Classes and labels / Probabilities versus classes
  3. 02Logistic regression intuition / Decision thresholds
  4. 03Confusion matrix / Accuracy limitations
  5. 04Precision / Recall
  6. 05F1 score / ROC intuition
  7. 06PR curve intuition / Building a baseline classifier
  8. 07Missing-data strategy / Categorical variables
  9. 08Encoding / Scaling
  10. 09Transformations / Interactions
  11. 10Date features / Text features intuition
  12. 11Feature leakage / Train-fit transformations
  13. 12Pipelines / Documenting features
03

Week 3

Generalisation, Regularisation and Tree Models

13 lessons3 checkpoints1 applied assignment

Control overfitting and model complexity with validation, regularisation, trees, random forests and gradient boosting intuition.

Applied work for this week

Complete the learning path and submit the applied task to AI SYNAPSE REVIEW.

  1. 01Training versus generalisation / Overfitting symptoms
  2. 01Week 3: Classification
  3. 02Underfitting / Model complexity
  4. 03Bias variance trade-off / L1 regularization
  5. 04L2 regularization / Elastic net intuition
  6. 05Hyperparameters / Learning curves intuition
  7. 06Validation decisions / Controlling complexity
  8. 07Decision trees / Splits and impurity intuition
  9. 08Tree depth / Random forests
  10. 09Bagging / Gradient boosting
  11. 10XGBoost intuition / Nonlinear interactions
  12. 11Feature importance / Overfitting in trees
  13. 12Tuning priorities / Choosing tree models
04

Week 4

Unsupervised Learning, Validation and Model Selection

13 lessons3 checkpoints1 applied assignment

Explore clustering, PCA and anomaly detection, then compare models honestly with cross-validation and time/group-aware validation.

Applied work for this week

Complete the learning path and submit the applied task to AI SYNAPSE REVIEW.

  1. 01Why unsupervised learning / Clustering goals
  2. 01Week 4: Features and Data Preparation
  3. 02K-means intuition / Distance and scaling
  4. 03Choosing K / Hierarchical clustering intuition
  5. 04Dimensionality reduction / PCA intuition
  6. 05Anomaly detection / Cluster interpretation
  7. 06Validation challenges / Responsible use of segments
  8. 07Holdout validation / Cross-validation
  9. 08Stratification / Time-aware validation
  10. 09Grouped validation / Hyperparameter tuning
  11. 10Nested evaluation concept / Model comparison
  12. 11Uncertainty in estimates / Multiple testing intuition
  13. 12Selection bias / Locking the final test
05

Week 5

Imbalanced Decisions, Interpretability and Explainability

13 lessons3 checkpoints1 applied assignment

Choose metrics and thresholds for imbalanced problems and explain model behaviour responsibly with global/local interpretation tools.

Applied work for this week

Complete the learning path and submit the applied task to AI SYNAPSE REVIEW.

  1. 01What imbalance means / Why accuracy misleads
  2. 01Week 5: Validation and Model Selection
  3. 02Precision recall trade-off / PR-AUC intuition
  4. 03Class weights / Resampling concepts
  5. 04Threshold tuning / Cost-sensitive decisions
  6. 05Calibration / Top-k decisions
  7. 06Operational capacity / Metric selection by business cost
  8. 07Global versus local explanation / Linear coefficients
  9. 08Tree importance / Permutation importance
  10. 09SHAP intuition / Partial dependence intuition
  11. 10Correlated features / Causal interpretation warning
  12. 11Communicating uncertainty / Explaining to stakeholders
  13. 12Model cards / Responsible explanation
06

Week 6

Deployment, Monitoring and Professional ML Capstone

13 lessons3 checkpoints1 applied assignment

Move from notebook to governed deployment, monitor performance and drift, and deliver a complete reproducible ML capstone.

Applied work for this week

Complete the learning path and submit the applied task to AI SYNAPSE REVIEW.

  1. 01From notebook to service / Batch versus real-time scoring
  2. 01Week 6: Professional ML Project
  3. 02Model serialization concepts / Input validation
  4. 03Feature consistency / Monitoring predictions
  5. 04Performance monitoring / Data drift
  6. 05Concept drift / Retraining triggers
  7. 06Versioning and rollback / Governance after launch
  8. 07Frame the problem / Define target and observation
  9. 08Design data split / Build baseline
  10. 09Prepare features / Train candidate models
  11. 10Choose metrics / Validate honestly
  12. 11Interpret results / Assess risks and limitations
  13. 12Plan deployment monitoring / Present the final model
Programme evidence

What the learner studies and is assessed on

This summary shows the complete learning and assessment trail used for course completion. Certificate eligibility is created only after the required weekly work, knowledge checks and applied assessments are completed. For this free course, the verified certificate is an optional paid add-on available at the end.

What the learner will be able to do

  • ✓Frame supervised and unsupervised learning problems correctly; separate training, validation and test data; build and interpret regression and classification baselines; select appropriate metrics; recognize overfitting, leakage and class imbalance; understand regularization, trees and ensemble models; create defensible feature pipelines; use cross-validation appropriately; interpret feature importance and model explanations; describe monitoring and drift; present a complete, responsible ML workflow.

Material examined

  1. Week 1: ML Workflow and Regression FoundationsWhat machine learning is / Prediction versus explanation, Week 1: The Machine Learning Workflow, Problem framing / Features and targets, Unit of observation / Training data, Validation data / Test data, Baselines / Loss and metrics, Reproducibility / End-to-end workflow, Continuous targets / Linear regression intuition, Coefficients and predictions / Residuals, Mean squared error / Mean absolute error, R-squared intuition / Scaling considerations, Nonlinearity / Outliers and sensitivity, Regression diagnostics / Building a baseline regressionLesson material · practical work · knowledge quiz
  2. Week 2: Classification and Data PreparationWeek 2: Regression, Classes and labels / Probabilities versus classes, Logistic regression intuition / Decision thresholds, Confusion matrix / Accuracy limitations, Precision / Recall, F1 score / ROC intuition, PR curve intuition / Building a baseline classifier, Missing-data strategy / Categorical variables, Encoding / Scaling, Transformations / Interactions, Date features / Text features intuition, Feature leakage / Train-fit transformations, Pipelines / Documenting featuresLesson material · practical work · knowledge quiz
  3. Week 3: Generalisation, Regularisation and Tree ModelsTraining versus generalisation / Overfitting symptoms, Week 3: Classification, Underfitting / Model complexity, Bias variance trade-off / L1 regularization, L2 regularization / Elastic net intuition, Hyperparameters / Learning curves intuition, Validation decisions / Controlling complexity, Decision trees / Splits and impurity intuition, Tree depth / Random forests, Bagging / Gradient boosting, XGBoost intuition / Nonlinear interactions, Feature importance / Overfitting in trees, Tuning priorities / Choosing tree modelsLesson material · practical work · knowledge quiz
  4. Week 4: Unsupervised Learning, Validation and Model SelectionWhy unsupervised learning / Clustering goals, Week 4: Features and Data Preparation, K-means intuition / Distance and scaling, Choosing K / Hierarchical clustering intuition, Dimensionality reduction / PCA intuition, Anomaly detection / Cluster interpretation, Validation challenges / Responsible use of segments, Holdout validation / Cross-validation, Stratification / Time-aware validation, Grouped validation / Hyperparameter tuning, Nested evaluation concept / Model comparison, Uncertainty in estimates / Multiple testing intuition, Selection bias / Locking the final testLesson material · practical work · knowledge quiz
  5. Week 5: Imbalanced Decisions, Interpretability and ExplainabilityWhat imbalance means / Why accuracy misleads, Week 5: Validation and Model Selection, Precision recall trade-off / PR-AUC intuition, Class weights / Resampling concepts, Threshold tuning / Cost-sensitive decisions, Calibration / Top-k decisions, Operational capacity / Metric selection by business cost, Global versus local explanation / Linear coefficients, Tree importance / Permutation importance, SHAP intuition / Partial dependence intuition, Correlated features / Causal interpretation warning, Communicating uncertainty / Explaining to stakeholders, Model cards / Responsible explanationLesson material · practical work · knowledge quiz
  6. Week 6: Deployment, Monitoring and Professional ML CapstoneFrom notebook to service / Batch versus real-time scoring, Week 6: Professional ML Project, Model serialization concepts / Input validation, Feature consistency / Monitoring predictions, Performance monitoring / Data drift, Concept drift / Retraining triggers, Versioning and rollback / Governance after launch, Frame the problem / Define target and observation, Design data split / Build baseline, Prepare features / Train candidate models, Choose metrics / Validate honestly, Interpret results / Assess risks and limitations, Plan deployment monitoring / Present the final modelLesson material · practical work · knowledge quiz · final project