The paper defines an 'unnecessary constraint' in kernel-based learning from logical constraints and shows it can be detected by logical consequence or by finding alternative Lagrange multipliers that preserve the optimal kernel expansion.
Machine learning 86(1), 57–88 (2012)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LO 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Conditions for Unnecessary Logical Constraints in Kernel Machines
The paper defines an 'unnecessary constraint' in kernel-based learning from logical constraints and shows it can be detected by logical consequence or by finding alternative Lagrange multipliers that preserve the optimal kernel expansion.