Week 9: March 10-March 17

Interactive session, Wednesday March 10

• Back-prop example w/Marius Havgar

Weekly lecture:

Slides

Video Recordings:

1. Overview
2. scikit-learn applied to mandatory 2
   • Notebook
3. Overfitting and regularization
4. Regularization in scikit-learn
   • Notebook
5. Bias-variance tradeoff
6. Cross-validation
   • Notebook
7. Ensemble learning and Random forests

Readings:

Hal Daumé III, A course in Machine Learning

• Ch. 5 Practical issues, sec. 5.0-5.6 (p.55-66), except precision-recall curves, ROC curves and AUC curves.

Jurafsky and Martin, Speech and language processing, 3rd ed. draft, 30 Dec. 2020

• Ch. 5, sec. 5.5 Regularization
  • except the last paragraph starting with "Both L1 and L2..."
• Ch. 4, sec 4.7 "Evaluation: Precision, Recall, F-measure"
• Ch. 4, sec 4.8 "Test sets and Cross-validation"

Marsland

• Ch 2, sec 2.5 (Not the formulas)
• Ch 13: Introduction, 13.2 Bagging, 13.3 Random forest

Group sessions

• No new exercises. Work on Mandatory assignment 2 and earlier weekly exercises.
• Starting from March 1, UiO has opened for physical teaching in small groups of up to 20 people. We will therefore turn to physical group sessions as originally planned, starting from our teaching week 8. All attendees to the physical groups will have to sign up before the session to make sure there is room. You can read more about the booking system and about which groups which will turn physical here. We will in addition continue with some digital groups. You do not have to sign up to the digital groups

Recommended readings

Through the UiO library we have now got access to https://www.oreilly.com/library/view/temporary-access/ . Log in with UiO user name. They have many useful books in ML (and computing at large). In particular, they have published some of the best-selling books in ML, including some using scikit-learn. We recommend 

• Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow,
  by Aurélien Géron 

The following is also useful, cover some of the same, a little less technical:

• Introduction to Machine Learning with Python: A Guide for Data Scientists
  by Andreas C. Müller and Sarah Guido