Distribution-free Testing for Halfspaces (Almost) Requires PAC Learning.
Xi ChenShyamal PatelPublished in: SODA (2022)
Keyphrases
- pac learning
- distribution free
- concept class
- vc dimension
- concept classes
- sample complexity
- learning theory
- uniform distribution
- membership queries
- target concept
- sample size
- agnostic learning
- dnf formulas
- upper bound
- decision lists
- learning problems
- lower bound
- theoretical analysis
- inductive inference
- learning algorithm
- pac model
- term dnf
- statistical queries
- concept learning
- active learning
- euclidean space
- feature selection