Signature transforms approximate path-dependent nonlinear rewards as linear functionals, enabling the DisSigUCB algorithm with a high-probability regret bound of order O(sqrt((d+m)KT)).
Risk of transfer learning and its applications in finance,
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
roles
other 1polarities
unclear 1representative citing papers
Transfer learning achieves sample complexity O(m^{-(α+1)/d}) for d>3 via optimal transport, outperforming direct learning's O(m^{-p/d}) when target models are not smooth.
Proposes Mallows-type weights for parameter-transfer learning that are asymptotically optimal for target prediction risk and selectively weight informative sources without requiring correct source models.
citing papers explorer
-
Signature Approach for Contextual Bandits with Nonlinear and Path-dependent Rewards
Signature transforms approximate path-dependent nonlinear rewards as linear functionals, enabling the DisSigUCB algorithm with a high-probability regret bound of order O(sqrt((d+m)KT)).
-
Sample Complexity of Transfer Learning: An Optimal Transport Approach
Transfer learning achieves sample complexity O(m^{-(α+1)/d}) for d>3 via optimal transport, outperforming direct learning's O(m^{-p/d}) when target models are not smooth.
-
Generalized optimal parameter-transfer learning through Mallows-type model averaging
Proposes Mallows-type weights for parameter-transfer learning that are asymptotically optimal for target prediction risk and selectively weight informative sources without requiring correct source models.