HALK, a primal-dual kernel method with greedy compression, attains O(T^{-1/2}+alpha) expected suboptimality with zero average constraint violation and an explicit dictionary-size bound.
A multi-robo t control policy for information gathering in the presence of unknown hazards,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive Kernel Learning in Heterogeneous Networks
HALK, a primal-dual kernel method with greedy compression, attains O(T^{-1/2}+alpha) expected suboptimality with zero average constraint violation and an explicit dictionary-size bound.