DisjunctiveNet represents input-dependent mixed-integer rules as disjunctive constraints and applies hierarchical convex relaxations to create tractable linear layers that enforce exact rule satisfaction inside end-to-end neural networks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DisjunctiveNet: Neural Symbolic Learning via Differentiable Convexified Optimization Layers
DisjunctiveNet represents input-dependent mixed-integer rules as disjunctive constraints and applies hierarchical convex relaxations to create tractable linear layers that enforce exact rule satisfaction inside end-to-end neural networks.