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.
In: Proceedings of the 20th International Conference on Machine Learning (ICML-03)
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.