Optimization Algorithms for Machine Learning

Wasserstein Barycnters

NEWS: During the lectures we will use the Julia programming language (please, install or update to version 1.1, or above). We will start using the Plots.jl library.

All the material is available on dropbox.

[02/04/2019]: Digit Classification Challenge [.pdf] Scripts: challenge.jl, xor2.jl Data: seavision_trainig.csv, small_5_6.csv
[26/03/2019]: Slides Classification Solution for homer.jl - T3.jl. mlp_example.jl - CNN.jl
[19/03/2019]: Slides Multilayer Neural Networks Script for homer.jl, T2-ex.jl.
Suggested reading: Chapter 11 of book [1]
[14/03/2019]: Slides Clustering Script for exercise1.jl, exercise2.jl. Data: Test_MNSIT_1_2_8.csv, MNIST_all.csv
Suggested reading: Chapter 9 of book [2]
[05/03/2019]: Slides Introduction Script for basic_regression.jl.
Suggested reading: Chapter 1 of book [1] and Chapter 1 and 2 of book [2]

Readings

Pretty nice video on YouTube:

General reading about Machine Learning topics:

Readings about the Flux Machine Learning Stack:

Readings about the Julia Programming Language:

Interesting Links

  • Runge phenomena: Polynomial interpolation at evenly-spaced nodes converges iff the function is analytic inside a football-shaped region.

Useful Links