### Basic Information

**Course Code**: EE636**Course Name**: Matrix Computations**Course Offered In**: 2022-‘23**Semester Season**: Spring**Instructors**: Madhu Belur Sir**Prerequisites**: None**Difficulty (1 being easy and 5 being tough)**: 4

### Course Content

norms, SVD, concept of stability of algorithms, Ax=b solution procedures, LU factorization, QR factorization, Givens rotation, Householder reflections, flop count, error analysis and backward/forward stability of methods, eigenvalue/eigenvector computation, Schur forms, power-iteration, Hessenberg forms, symmetric/positive-definite matrices, normal matrices, conjugate gradient methods for solving Ax=b.

### Feedback on Lectures

Lectures were very detailed, sir gave deep intuitions on the various concepts taught in class. Following the lectures is very important as even though the proofs are available in the reference books, the intuition that sir gives is very useful and gives a very deep geometric understanding rather than just some out of the box algebraic manipulations.

### Feedback on Evaluations

Exam-policy:

- quiz 1: 7% (2nd week of Jan)
- midsem: 28% (during midsem week)
- quiz 2: 28% in end-March or early-April
- endsem: 37% (balance)
The exams were pretty easy if we have followed the lectures and attempted the tutorial problems.
### Study Material and Resources

Tut sheets were given regularly, they had pretty tough problems.

Books: (by and large: the first one, but others will help)

- Golub & van Loan: Matrix Computations
- Watkins: Fundamentals of Matrix Computations
- Trefethen: Numerical Linear Algebra
### Follow-up Courses

EE710: Large Sparse Matrix Computations

### Final Takeaway

The course is very interesting and also very useful in understanding controls system theory. If you’re interested in linear algebra and/or control theory, you’ll enjoy the course. Following the lectures and solving tut problems is very important.