Fast Bayesian optimization methods match SOTA at 100x–1000x speed
The methods target continuous and combinatorial inputs from about 10 to more than 6000 dimensions by using linear structure already in BO.
Stories per day · 30 days
1 RSS
Relatedbayesian-optimization1black-box-evaluation1cma-es1combinatorial-optimization1differential-privacy1quasi-concave-optimization1
3 stories, newest first · one card per story
Stories tagged #high-dimensional
Differentially private black-box evaluation extended to high dimensions
The method reduces evaluation to private quasi-concave optimization and replaces multiplicative sample overhead with additive overhead.
Minimax Additive Regression under Unknown Dependent Designs
Researchers prove minimax rates for additive regression when dimension grows and the design may be dependent.