Module 1 Flashcards
1
Q
Curse of dimensionality
A
- increased computational complexity
- data sparsity
- overfitting
2
Q
Lazy learner
A
- stores training example
- postponed generalising beyond these data until an explicit request is made
3
Q
Eager Learner
A
- uses the training examples
- constructs a general, explicit description of the target function
4
Q
Bias-variance trade off
A
Low variance → high bias
High variance → low bias
5
Q
Accuracy
A
correct predictions / # test instances
6
Q
Baseline
A
Chance/random performance [lower bound]
7
Q
Upper bound
A
The best case (usually comparison to human)
