APL405 Machine Learning in Mechanics

---

Course info

Credit: 3 Units (2-0-2).

Instructor: Dr. Md Rushdie Ibne Islam, Email: rushdie@am.iitd.ac.in, Office: 4B-19 (Block IV, top floor).

Class timings:

• Lecture: Tuesday, Thursday and Friday; 11 AM – 11:50 AM; LH 517
• Practical: Wednesday; 3 PM - 5 PM; LH 503

TAs: Tapas, Sawan and Rohan.

Office hours: Either stop by instructor's office (for quick discussion) or take an email appointment (for longer discussion).

Important note: Please include APL405 (without space) in the subject line for all email communications related to this course.

Syllabus

• Part I: Foundations
  • Introduction and mathematical preliminaries:
    • Review of linear algebra.
    • Probability theory.
    • Matrix calculus.
  • Setting of Supervised Learning.
  • k-Nearest Neighbors, Decision Trees.
  • Linear regression; Classification and Logistic Regression.
  • Probabilistic interpretation; Maximum Likelihood Estimation (MLE).
• Part II: Optimization and Model Selection
  • Regularization and Generalized Linear Models.
  • Cross-validation:
    • Training error vs generalization gap.
    • Bias-variance decomposition.
  • Loss functions and parameter optimization.
• Part III: Advanced Models and Learning Paradigms
  • Neural Networks and Deep Learning.
  • Kernel Ridge Regression; Theory of kernels; Support Vector Classification.
  • Ensemble Methods:
    • Bagging and Random Forests.
    • Boosting.
  • Generative Models and Unsupervised Learning:
    • Gaussian Mixture Models (GMM) with Expectation Maximization (EM).
    • k-means clustering.
    • Principal Component Analysis (PCA).

References

• Primary
  • Lindholm, A., Wahlström, N., Lindsten, F. and Schön, T.B. Machine Learning: A First Course for Engineers and Scientists.
  • Bishop, C. Pattern Recognition and Machine Learning.
• Additional
  • Brunton, S. L. and Kutz J.N. Data-Driven Science and Engineering: Machine Learning, Dynamical Systems and Control.
  • Murphy, K.P. Machine learning: A Probabilistic Perspective.

Grading

• Homework: 30%
• Minor: 15%
• Major: 15%
• Practical: 10%
• Project: 30%

If a student decides to audit the course, the passing grade will be 40%. Additionally, the student must write minor and major exams, and the attendance must be greater than or equal to 75% in lectures and tutorials separately.

Homeworks, minor and major

• Four homeworks will be assigned; two (homeworks #1 and #2) before and another two (homeworks #3 and #4) after the minor exam. Exact date, time and place for submission will be notified later.
• Minor and major exams will be scheduled based on the institute timetable.

Attendance

• Students are strongly encouraged to attend all the lectures sessions. Attendance will be recorded for each lecture session. If a student's attendance is less than 75%, the student will be awarded one grade less than the actual grade that he/she has earned. For example, a student who has got an A grade but has attendance less than 75% will be awarded an A (-) grade. However, please note that there are no marks for attendance.
• In case of absence due to unforeseen circumstances, such as medical reasons, the student must inform the instructor via email within a week from the date of absence.
• If a student cannot appear for an exam due to medical reasons, a medical certificate needs to be produced from IIT Delhi hospital for consideration of time extension or re-exam.