VC-Dimension Based Generalization Bounds for Relational Learning.
Ondrej KuzelkaYuyi WangSteven SchockaertPublished in: CoRR (2018)
Keyphrases
- relational learning
- generalization bounds
- vc dimension
- learning problems
- sample complexity
- inductive logic programming
- vapnik chervonenkis
- relational data
- upper bound
- information extraction
- sample size
- lower bound
- statistical learning theory
- inductive inference
- learning machines
- uniform convergence
- worst case
- phase transition
- background knowledge
- function classes
- model selection
- learning algorithm
- data dependent
- compression scheme
- data mining
- gaussian processes
- euclidean space
- learning theory
- learning tasks
- kernel methods
- semi supervised learning
- generalization ability
- linear classifiers
- training data
- special case
- machine learning
- theoretical analysis
- image compression