Project
Mgr. Radoslav Doktor
Project

Goal: Book on Current Trends in Machine learning

Areas

  • Online machine learning
  • Gradient boosting
  • Concept drift
  • Logistic regression
  • Multi-label classification
  • Belief networks
  • Animations in data analytics
  • Bayesian optimization
  • Kernel methods
  • Machine learning methods for quantile regression
  • Probabilistic multirelational learning
  • AutoML after AutoWeka and auto-scikit-learn
  • (Visual data mining without animations)
  • (Graph mining)
  • (Statistics and ML, Statistical test)
  • Gaussian processes
  • (Quantum ML (ICML 2020 Tutorial))
  • Novelty detection
  • and more . . .

Choose/suggest an area (group by 4). Each group: Write a chapter on a state-of-the-art. Present it in the lecture time.


Project Part 2

Each member of the group: Find two algorithms/methods/tools from the area, one written in R, one in Python. Use an OpenML dataset (preferably the same for all algorithms) to test the algorithms. Do not choose a dataset with accuracy (check OpenML) bellow  60% or higher than 90%.

A final version of a template (a jupyter notebook) is already in Study materials.

Upload the second part of the project to the corresponding homework vault (q.v. link down below)

DEADLINE  (EXTENDED)   Sunday, June 6



Two chapters from Encyclopedia of Machine learning by Claude Sammutt et al. (Springer 2012). See maybe some other for your inspiration!




Předchozí
Následující