Foundations 0/0

Foundations

Classical models

Perceptrons, support vector machines, decision trees, random forests, gradient-boosted trees and Gaussian classifiers. Gradient-boosted trees are still the usual first choice for tabular data, and margins, impurity and bagging come back in deep learning.

0 / 36 solved

Questions

0/36 solved
Question Module Tags Difficulty Success
1. Mistake-Driven Nudge Perceptron & SVM
perceptronconvergence
Easy —
2. The Widest Gap Perceptron & SVM
svmmargin
Easy —
3. Hinge and Squeeze Perceptron & SVM
svmhinge-loss +1
Medium —
4. The Chosen Few Perceptron & SVM
svmsupport-vectors +1
Medium —
5. Infinite Dimensions Trick Perceptron & SVM
kernel-trickrbf +1
Medium —
6. Turning the Dial Perceptron & SVM
svmregularization +1
Medium —
7. Widest Street Wins Perceptron & SVM
svmmargin +1
Medium —
8. Tight Against the Boundary Perceptron & SVM
svmkkt +2
Medium —
9. Normalized After the Fact Multiclass & Decision Trees
multiclassone-vs-rest +2
Medium —
10. Weight Matrix Headcount Multiclass & Decision Trees
multiclasssoftmax +1
Easy —
11. The Tree That Grew Too Deep Multiclass & Decision Trees
decision-treespruning +1
Easy —
12. Weighted Impurity of a Split Multiclass & Decision Trees
decision-treesgini +2
Medium —
13. Split Decisions, One Node at a Time Multiclass & Decision Trees
decision-treesgini +2
Medium —
14. Variance Cut Multiclass & Decision Trees
decision-treesregression-trees +1
Medium —
15. XOR Ambush Multiclass & Decision Trees
decision-treesgreedy-splitting +1
Medium —
16. Stuck At Zero Ensembles & Neural-Net Basics
reludying-relu +2
Medium 78%
17. Correlation Beats Strength Ensembles & Neural-Net Basics
random-forestbagging +2
Medium —
18. Untouched Trees Ensembles & Neural-Net Basics
ensemblesrandom-forest +1
Easy —
19. Eight Cores Forty Trees Ensembles & Neural-Net Basics
baggingboosting +2
Medium —
20. Counting The Zeros Ensembles & Neural-Net Basics
l1-regularizationlasso +2
Medium —
21. Ten Layers of Squash Ensembles & Neural-Net Basics
activation-functionssigmoid +2
Medium —
22. Straight Line Fails Ensembles & Neural-Net Basics
neural-networksxor +1
Medium —
23. Nine Hundred Ninety-Nine Problems Ensembles & Neural-Net Basics
neural-networksparameters +1
Medium —
24. Matched Pair Ensembles & Neural-Net Basics
neural-networkssoftmax +1
Medium —
25. 255 Leaves, 3000 Rows Gradient Boosting Internals
lightgbmleaf-wise-growth +1
Medium —
26. Slow And Steady Gradient Boosting Internals
gradient-boostinghyperparameters +1
Medium —
27. The Eight Times Rule Gradient Boosting Internals
lightgbmgoss +1
Hard —
28. Leaf Tax Gradient Boosting Internals
xgboostregularization +1
Hard —
29. The Prior Overrules the Peak Generative Models & Gaussians
gaussian-discriminant-analysispriors +2
Medium 50%
30. Two Words, One Posterior Generative Models & Gaussians
naive-bayesconditional-independence +2
Medium 63%
31. Priors and Peaks Generative Models & Gaussians
generative-modelsgaussian +1
Easy 83%
32. Reading the Covariance Matrix Generative Models & Gaussians
covariancegaussian
Easy 50%
33. Biased by Design Generative Models & Gaussians
mlegaussian
Easy 36%
34. Round Blobs vs Squished Blobs Generative Models & Gaussians
covariancegaussian
Medium 100%
35. Counting Covariance Parameters Generative Models & Gaussians
ldaqda +2
Medium 44%
36. When Lines Do the Job Generative Models & Gaussians
ldagaussian +1
Medium 100%