Machine Learning (CS60050) - Spring 2026

• Instructor: Sourangshu Bhattacharya
• Teaching Assistants: Suman Bera, Saptarshi Mandal, Vaishnovi Arun, Bibhudatta Bhanja, Dhyana Vardhan
• Class Schedule: Monday (8:00 - 9:55), Tuesday (12:00 - 12:55)
• Classroom: NR-412 (Nalanda Complex)

Announcements

References

• Thomas M. Mitchell, "Machine Learning." McGraw Hill (1997) -- Traditional book for concept learning.
• Christopher M. Bishop, "Pattern Recognition and Machine Learning." Springer (2006). Download Link -- Main book for most models
• Raschka, Sebastian, Yuxi Hayden Liu, and Vahid Mirjalili.
  Machine Learning with PyTorch and Scikit-Learn: Develop machine learning and deep learning models with Python.
  Packt Publishing Ltd, 2022. -- Practice Programming with github code link

Course Schedule

Week	Dates	Topic	Materials
Week 1	5/1, 6/1	Introduction to ML and concept learning	Introduction
Week 2	12/1, 13/1	Linear models, regression, classification	Linear Models
Week 3	19/1, 20/1	Support Vector Machines	SVM
Week 4	27/1	Probabilistic Models	Probabilistic Models
Week 5	2/2, 3/2	Ensemble Learning	Ensemble Learning
Week 6	9/2, 10/2	Boosting	Boosting
Week 7	2/3, 3/3	Cross Validation	Cross Validation
Week 8	9/3, 10/3	Neural Networks	Neural Networks
Week 9	16/3, 17/3	RNN & Transformers	RNN & Transformers
Week 10	23/3, 17/3	Clustering - k-Means, Hierarchical, Gaussian Mixture Models	Clustering SLides
Week 11 + 12	31/3, 6/4, 7/4	Graphical Models - Bayesian Networks, MRFs, Message Passing on Factor Graphs	Graphical Models SLides

Syllabus (Broad Indication)

• Foundations: Introduction & concept learning, Linear models for classification and regression.
• Kernels & Trees: SVMs and kernel methods, Decision Trees, Bagging, Random forests.
• Probabilistic ML: Naive Bayes, Gaussian Mixture Models, Graphical Models.
• Advanced Ensemble: Boosting and XGBoost.
• Neural Networks: Architectures, Backpropagation, and Deep Learning basics.
• Theory & Meta-Learning: Learning Theory, Metrics, Active, Transfer, and Multi-task learning.
• Ethics: Explainability and Trustworthiness in AI.
• Unsupervised: Clustering techniques and Latent Variable models.