Tight Lower Bounds on Worst-Case Guarantees for Zero-Shot Learning with Attributes.
Alessio MazzettoCristina MenghiniAndrew YuanEli UpfalStephen H. BachPublished in: NeurIPS (2022)
Keyphrases
- lower bound
- worst case
- upper bound
- theoretical guarantees
- np hard
- branch and bound algorithm
- branch and bound
- average case
- lower bounding
- error bounds
- lower and upper bounds
- running times
- upper and lower bounds
- objective function
- attribute values
- worst case bounds
- greedy algorithm
- approximation algorithms
- sample size
- approximation guarantees
- constant factor
- optimal solution
- knowledge transfer
- decision table
- generalization error
- linear programming relaxation
- concept lattice
- multi attribute
- image search
- learning theory
- computational complexity
- search algorithm