Sharp oracle bounds for monotone and convex regression through aggregation.
Pierre C. BellecAlexandre B. TsybakovPublished in: J. Mach. Learn. Res. (2015)
Keyphrases
- upper bound
- monotonicity constraints
- regression model
- convex combinations
- lower bound
- piecewise linear
- response variable
- data aggregation
- upper and lower bounds
- lower and upper bounds
- regression algorithm
- oracle database
- convex optimization
- worst case
- linear regression
- support vector regression
- gaussian processes
- convex hull
- model selection
- high quality
- regression analysis
- learning machines
- convex sets
- support vector
- ridge regression
- database
- lipschitz continuity
- regression function
- generalization bounds
- regression method
- convex relaxation
- aggregation operators
- partial least squares
- learning algorithm
- expected error
- vc dimension
- aggregating algorithm