Applied Machine Learning - Week 2

Monday the 2nd - Friday the 6th of May 2022

Monday 3rd of May (afternoon):
Lectures: Introduction to Loss Functions, Stochastic Gradient Descent, and Training/Validation (TP).
     Initial project start, and if time allows, I'll briefly introduce/remind about Principle Component Analysis.

Exercise: Use cross validation (CV) in your training and testing. As the advantage of CV is typically for smaller datasets, try using 10% (or 1%) of the Aleph b-jet data.
     Also check the performance on each k-fold, and determine the uncertainty in your performance.
     Finally, you should try to change your (classification) loss function and also rewrite your analysis to predict "energy" or "cTheta" from the other variables (i.e. regression).


Wednesday 5th of May (morning):
Lectures: Introduction to Unsupervised Learning: Clustering and Nearest Neighbor algorithms on astro data.

Exercise: The exercise consists of preprocessing and applying dimensionality reduction (v1) to the Cosmos2015 data.


Wednesday 5th of May (afternoon):
Lectures: Further work on Unsupervised Learning: Clustering and Nearest Neighbor algorithms on astro data.

Exercise: The exercise consists of preprocessing and applying dimensionality reduction (v2) to the Cosmos2015 data.
     The analysis will this time also include the Swift Properties and Swift Gamma Ray Bursts datasets.

Example solutions from week 2:
The following are example solutions and related code, which comes with absolutely no warrenty, that you may let yourself be inspired by:
  • Dimensionality Reduction Part I.
  • Dimensionality Reduction Part II.
    Last updated: 6th of May 2022 by Troels Petersen.