61.501 Production-Ready Machine Learning

Course description

This course provides a comprehensive foundation in machine learning, covering both fundamental concepts and algorithms of machine learning (ML) and practical implementation. Students will delve into supervised learning methods such as regression, classification, decision trees, support vector machines, and ensemble techniques including random forests and gradient boosting. Unsupervised learning approaches such as K-means clustering, hierarchical clustering, principal component analysis (PCA) and t-distributed stochastic neighbour embedding (t-SNE) are also introduced, alongside optimisation methods like gradient descent and its variants.

The course emphasises model evaluation, model selection, feature engineering, and ethical considerations in machine learning. Beyond foundational algorithms, the course focuses on the operationalisation of ML systems for real-world deployment. Students will gain hands-on experience with end-to-end ML workflows, including CI/CD for ML, REST API deployment, cloud-based training and inference using AWS SageMaker, real-time data streaming, experiment tracking, and responsible AI practices.

By integrating algorithmic understanding with production-grade deployment skills, the course prepares students for advanced topics in artificial intelligence and data science, ensuring a well-rounded understanding of both popular and powerful machine learning methodologies beyond just neural networks.

Instructor

Pritee Agrawal

Number of credits: 12