Definable nonconvex parametric optimization problems admit an adjoint state formula under a qualification condition, selecting a conservative field for the value function without smoothness or uniqueness assumptions.
Optnet: Differentiable optimization as a layer in neural networks
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
Introduces PZOS partial zeroth-order algorithm for MPECs that exploits leader's white-box cost information to achieve lower variance than full black-box zeroth-order methods, with convergence to partial Goldstein stationary points and empirical gains on routing and security games.
SnareNet introduces a repair layer that navigates the range space of constraints plus adaptive relaxation training to enforce hard non-convex constraints on neural network outputs more reliably than prior methods.
An alternative complementarity formulation for primal-dual interior-point methods keeps linear systems spectrally bounded near the solution, enabling stable single-precision solves and differentiation for bilevel and end-to-end learning.
C3PO is a foundation model for bilevel pricing optimization that trains on simulated discrete choice data and retrieves elasticity priors from literature to improve revenue KPIs under business constraints.
citing papers explorer
-
The adjoint state method for parametric definable optimization without smoothness or uniqueness
Definable nonconvex parametric optimization problems admit an adjoint state formula under a qualification condition, selecting a conservative field for the value function without smoothness or uniqueness assumptions.
-
Black-Box Followers, White-Box Leaders: Partial Zeroth-Order Methods for MPECs
Introduces PZOS partial zeroth-order algorithm for MPECs that exploits leader's white-box cost information to achieve lower variance than full black-box zeroth-order methods, with convergence to partial Goldstein stationary points and empirical gains on routing and security games.
-
SnareNet: Flexible Repair Layers for Neural Networks with Hard Constraints
SnareNet introduces a repair layer that navigates the range space of constraints plus adaptive relaxation training to enforce hard non-convex constraints on neural network outputs more reliably than prior methods.
-
A Differentiable Interior-Point Method in Single Precision
An alternative complementarity formulation for primal-dual interior-point methods keeps linear systems spectrally bounded near the solution, enabling stable single-precision solves and differentiation for bilevel and end-to-end learning.
-
Causal-Aware Foundation-Model for Bilevel Optimization in Discrete Choice Settings
C3PO is a foundation model for bilevel pricing optimization that trains on simulated discrete choice data and retrieves elasticity priors from literature to improve revenue KPIs under business constraints.