- We could define an unbiased hypothesis space by allowing arbitrary conjunctions, disjunctions, and negations.
- The number of possible hypothesis (ignoring semantic repetition) is ${2}^{\left|X\right|}$.
*The Candidate-Elimination algorithm, if applied to this hypothesis space, will be unable to generalize beyond the observed examples!*- So, in order to learn the concept we would need to present every single example.
- If we use a partially-learned concept, then
*exactly half*of the hypotheses will vote positive and half negative for every unseen example.

29 of 31 |