On PAC learning algorithms for rich Boolean function classes.
Lisa HellersteinRocco A. ServedioPublished in: Theor. Comput. Sci. (2007)
Keyphrases
- function classes
- vc dimension
- sample complexity
- learning algorithm
- pac model
- agnostic learning
- upper bound
- boolean functions
- pac learning
- uniform distribution
- sample size
- concept learning
- learning problems
- lower bound
- concept classes
- real valued
- generalization bounds
- data dependent
- theoretical analysis
- generalization error
- learning machines
- supervised learning
- statistical learning theory
- membership queries
- worst case
- concept class
- target function
- inductive inference
- training data
- noise tolerant
- linear functions
- machine learning algorithms
- active learning
- statistical queries
- computational learning theory
- euclidean space
- decision lists
- compression scheme
- generalization ability
- machine learning
- learning theory
- mistake bound