Course Opening: Students, Goals, and Applications
Course Opening: Students, Goals, and Applications¶
Status: outline — expand with your first-session slides and demos.
Getting to know each other¶
Backgrounds (math vs CS), programming comfort, expectations
How the sessions run: concept + notebook demos
Goals of the course¶
Shared mathematical language for data: representation, distance, structure, randomness, learning
Applications first; math when it explains what you already saw
Enough fluency to read BHK and later SML / ML courses
Example applications (demo list)¶
Images as matrices (intro, segmentation, compression)
kNN and k-means on real datasets
Curse of dimensionality (why more features can hurt)
SVD / PCA (faces, compression, ranking)
Random graphs (networks, phase transition)
Learning foundations (bias–variance, linear regression, SGD, perceptron, MLP)
The kernel trick (one session, no SVM)
What is deferred¶
Bayesian methods, HMM, MCMC Statistical Machine Learning
SVM other ML course
Markov transition matrices, HMM, graphical models other courses
Sources¶
Course README and
intro.mdBlum–Hopcroft–Kannan, Foundations of Data Science (preface / Ch. 1)