Dual VC Dimension Obstructs Sample Compression by Embeddings.
Zachary ChaseBogdan ChornomazSteve HannekeShay MoranAmir YehudayoffPublished in: CoRR (2024)
Keyphrases
- vc dimension
- sample size
- compression scheme
- euclidean space
- concept classes
- sample complexity
- upper bound
- image compression
- inductive inference
- statistical learning theory
- compression ratio
- distribution free
- compression algorithm
- data compression
- lower bound
- covering numbers
- generalization bounds
- empirical risk minimization
- concept class
- pac learnability
- vapnik chervonenkis dimension
- worst case
- vector space
- uniform convergence
- pac learning
- model selection
- generalization error
- low dimensional
- high dimensional
- support vector
- training set
- dimensionality reduction
- high dimensional data