Requiring LICQ/SCS/SOSC everywhere in bilevel optimization is non-prevalent and rigid, while holding almost everywhere is prevalent, but the distinction introduces fundamental difficulties.
Mixed citations
Title resolution pending
Mixed citation behavior. Most common role is background (60%).
34
Pith papers citing it
Background
60% of classified citations
citation-role summary
background 3
method 2
citation-polarity summary
representative citing papers
Local LMO is a new projection-free method that achieves the convergence rates of projected gradient descent for constrained optimization by using local linear minimization oracles over small balls.
Derives non-asymptotic error bounds for standard, defensive, and self-normalized importance sampling with random KDE proposals from geometrically ergodic Markov chains, separating n^{-1/2} Monte Carlo error from MIAE/MISE proposal error.
A diameter criterion tied to a potential function certifies convergence of difference inclusions, enabling discrete proofs for first-order optimization methods with diminishing steps.
In generalized contrastive learning with imbalanced classes, optimal representations collapse to class means whose angular geometry is determined by class proportions via convex optimization, and extreme imbalance causes all minority classes to collapse to one vector.
A physics-informed Bayesian model recovers user trajectories and radio maps from CSI measurements by using multipath feature distances as proxies for spatial displacements under known access point geometry.
A prediction market design that uses online learning to adaptively mix liquidity regimes from cost functions, achieving switching-regret bounds against the best hindsight sequence.
Optimistic bilevel optimization with manifold lower-level minimizers is differentiable if the optimistic selection is unique, yielding a pseudoinverse hyper-gradient and a convergent HG-MS algorithm whose rate depends on intrinsic manifold dimension.
Introduces Bayesian Sensitivity Value (BSV) for causal inference sensitivity analysis based on evidence-derived priors and Monte Carlo estimation, applied to diabetes treatment effects.
Policy iteration for discounted robust MDPs is strongly polynomial for L1 and L∞ uncertainty sets but hard for other Lp sets.
ABGD parametrizes piecewise linear functions as difference of max-affine functions and converges linearly to an epsilon-accurate solution with O(d max(sigma/epsilon,1)^2) samples under sub-Gaussian noise, which is minimax optimal up to logs.
Finite-size general security for DPSK QKD is achieved with positive key rates for 10^5 signals beyond 12 dB loss via variable-length entropy accumulation and conic optimization.
A modular reduction from budget-constrained contextual bandits with adversarial contexts to unconstrained bandits via surrogate rewards, yielding improved guarantees and an efficient algorithm based on SquareCB.
HyperCertificates combine closure certificates for lookahead with barrier and ranking functions to verify discrete-time systems against HyperLTL specifications.
Privatar uses horizontal frequency partitioning and distribution-aware minimal perturbation to enable private offloading of VR avatar reconstruction, supporting 2.37x more users with modest overhead.
Diffusion models solve noisy (non)linear inverse problems via approximated posterior sampling that blends diffusion steps with manifold gradients without strict consistency projection.
ICDN is a neural network that models log-demand from log-prices so elasticities can be derived exactly by differentiation, showing better out-of-sample performance than log-log benchmarks on beer sales data.
A projection-based algorithm for COCO achieves O(log T) regret and O(log T) CCV for strongly convex losses and O(sqrt(T)) for convex losses by leveraging self-contracted curves.
Establishes exponential convergence in Wasserstein distance for the mean-field limit and finite-particle approximation of a consensus-based method solving nonconvex bi-level optimization problems.
Learns regionally stable RNN models from input-output data by deriving LMI constraints from generalized sector conditions on deadzone activations and a barrier function to certify forward invariance on a compact set.
Introduces a fairness layer for deep learning models that guarantees output parity and an online primal-dual algorithm for aggregate fairness guarantees in streaming predictions with small batch sizes.
FLASH Policy uses sparse Legendre polynomial trajectory fitting and history-anchored flow matching to enable single-step inference for visuomotor control, reporting 31.4 ms per-episode latency and >=92% success on five simulated plus two real manipulation tasks.
Weighted rules extend stable model semantics to support probabilistic reasoning, model ranking, and statistical inference in answer set programs.
Stationary duality reduces composite cardinality optimization to simple cardinality, yielding dual problems with equivalent local solutions and global solutions under appropriate parameter selection.
citing papers explorer
-
On the Nature of Regularity Assumptions in Bilevel Optimization with Constrained Lower-level Problem
Requiring LICQ/SCS/SOSC everywhere in bilevel optimization is non-prevalent and rigid, while holding almost everywhere is prevalent, but the distinction introduces fundamental difficulties.
-
Local LMO: Constrained Gradient Optimization via a Local Linear Minimization Oracle
Local LMO is a new projection-free method that achieves the convergence rates of projected gradient descent for constrained optimization by using local linear minimization oracles over small balls.
-
Error Bounds for Importance Sampling with Estimated Proposal Distributions
Derives non-asymptotic error bounds for standard, defensive, and self-normalized importance sampling with random KDE proposals from geometrically ergodic Markov chains, separating n^{-1/2} Monte Carlo error from MIAE/MISE proposal error.
-
Convergence of difference inclusions via a diameter criterion
A diameter criterion tied to a potential function certifies convergence of difference inclusions, enabling discrete proofs for first-order optimization methods with diminishing steps.
-
Optimal Representations for Generalized Contrastive Learning with Imbalanced Datasets
In generalized contrastive learning with imbalanced classes, optimal representations collapse to class means whose angular geometry is determined by class proportions via convex optimization, and extreme imbalance causes all minority classes to collapse to one vector.
-
Annotation-Free Indoor Radio Mapping via Physics-Informed Trajectory Inference
A physics-informed Bayesian model recovers user trajectories and radio maps from CSI measurements by using multipath feature distances as proxies for spatial displacements under known access point geometry.
-
Adaptive Liquidity in Prediction Markets via Online Learning
A prediction market design that uses online learning to adaptively mix liquidity regimes from cost functions, achieving switching-regret bounds against the best hindsight sequence.
-
Select-then-differentiate: Solving Bilevel Optimization with Manifold Lower-level Solution Sets
Optimistic bilevel optimization with manifold lower-level minimizers is differentiable if the optimistic selection is unique, yielding a pseudoinverse hyper-gradient and a convergent HG-MS algorithm whose rate depends on intrinsic manifold dimension.
-
Bayesian Sensitivity of Causal Inference Estimators under Evidence-Based Priors
Introduces Bayesian Sensitivity Value (BSV) for causal inference sensitivity analysis based on evidence-derived priors and Monte Carlo estimation, applied to diabetes treatment effects.
-
On the Complexity of Discounted Robust MDPs with $L_p$ Uncertainty Sets
Policy iteration for discounted robust MDPs is strongly polynomial for L1 and L∞ uncertainty sets but hard for other Lp sets.
-
Locally Near Optimal Piecewise Linear Regression in High Dimensions via Difference of Max-Affine Functions
ABGD parametrizes piecewise linear functions as difference of max-affine functions and converges linearly to an epsilon-accurate solution with O(d max(sigma/epsilon,1)^2) samples under sub-Gaussian noise, which is minimax optimal up to logs.
-
Finite-size general security for differential phase shift keying via variable-length quantum key distribution
Finite-size general security for DPSK QKD is achieved with positive key rates for 10^5 signals beyond 12 dB loss via variable-length entropy accumulation and conic optimization.
-
Constrained Contextual Bandits with Adversarial Contexts
A modular reduction from budget-constrained contextual bandits with adversarial contexts to unconstrained bandits via surrogate rewards, yielding improved guarantees and an efficient algorithm based on SquareCB.
-
HyperCertificates: Verification of Discrete-time Dynamical Systems against HyperLTL Specifications
HyperCertificates combine closure certificates for lookahead with barrier and ranking functions to verify discrete-time systems against HyperLTL specifications.
-
Privatar: Scalable Privacy-preserving Multi-user VR via Secure Offloading
Privatar uses horizontal frequency partitioning and distribution-aware minimal perturbation to enable private offloading of VR avatar reconstruction, supporting 2.37x more users with modest overhead.
-
Diffusion Posterior Sampling for General Noisy Inverse Problems
Diffusion models solve noisy (non)linear inverse problems via approximated posterior sampling that blends diffusion steps with manifold gradients without strict consistency projection.
-
Integrable Elasticity via Neural Demand Potentials
ICDN is a neural network that models log-demand from log-prices so elasticities can be derived exactly by differentiation, showing better out-of-sample performance than log-log benchmarks on beer sales data.
-
Improved Guarantees for Constrained Online Convex Optimization via Self-Contraction
A projection-based algorithm for COCO achieves O(log T) regret and O(log T) CCV for strongly convex losses and O(sqrt(T)) for convex losses by leveraging self-contracted curves.
-
Convergence of Consensus-Based Particle Methods for Nonconvex Bi-Level Optimization
Establishes exponential convergence in Wasserstein distance for the mean-field limit and finite-particle approximation of a consensus-based method solving nonconvex bi-level optimization problems.
-
Learning the dynamics of nonlinear systems with regional stability guarantees through linear matrix inequality constraints
Learns regionally stable RNN models from input-output data by deriving LMI constraints from generalized sector conditions on deadzone activations and a barrier function to certify forward invariance on a compact set.
-
Differentiable Optimization Layers for Guaranteed Fairness in Deep Learning
Introduces a fairness layer for deep learning models that guarantees output parity and an online primal-dual algorithm for aggregate fairness guarantees in streaming predictions with small batch sizes.
-
FLASH: Efficient Visuomotor Policy via Sparse Sampling
FLASH Policy uses sparse Legendre polynomial trajectory fitting and history-anchored flow matching to enable single-step inference for visuomotor control, reporting 31.4 ms per-episode latency and >=92% success on five simulated plus two real manipulation tasks.
-
Weighted Rules under the Stable Model Semantics
Weighted rules extend stable model semantics to support probabilistic reasoning, model ranking, and statistical inference in answer set programs.
-
On the Stationary Duality of Structural Composite Cardinality Optimization
Stationary duality reduces composite cardinality optimization to simple cardinality, yielding dual problems with equivalent local solutions and global solutions under appropriate parameter selection.
-
Cooperative Multi-Static Target Localization for ISAC in Cluttered Industrial IoT Networks
A new cooperative ISAC localization algorithm reduces RMSE to about 45 cm after six iterations in cluttered IIoT by adaptively selecting informative nodes and suppressing persistent reflections.
-
On the Blessing of Pre-training in Weak-to-Strong Generalization
Pre-training provides a geometric warm start in a single-index model that enables weak-to-strong generalization up to a supervisor-limited bound, with empirical phase-transition evidence in LLMs.
-
The Distributionally Robust Cyclic Inventory Routing Problem
The authors create a distributionally robust formulation for the cyclic inventory routing problem that admits a deterministic reformulation via multi-point worst-case distributions and chance-constraint equivalents, solved by nested branch-and-price and tested on real automotive data.
-
An adaptive variance estimator for relative sparsity
A new adaptive variance estimator for relative sparsity coefficients is introduced that fully utilizes the prior asymptotic normality theorem and incorporates variable selection effects.
-
Decision-Focused Learning via Tangent-Space Projection of Prediction Error
PEAR computes regret gradients via tangent-space projection of prediction error, delivering top decision quality and efficiency on LP and QP tasks without solver differentiation.
-
Minimizing Collateral Damage in Activation Steering
Activation steering is cast as constrained optimization that minimizes collateral damage by weighting perturbations according to the empirical second-moment matrix of activations instead of assuming isotropy.
-
Representation-Guided Parameter-Efficient LLM Unlearning
REGLU guides LoRA-based unlearning via representation subspaces and orthogonal regularization to outperform prior methods on forget-retain trade-off in LLM benchmarks.
-
Robust Learning Meets Quasar-Convex Optimization: Inexact High-Order Proximal-Point Methods
Robust learning problems are formulated as quasar-convex optimization, and HiPPA is proposed as an inexact high-order proximal method with global and superlinear convergence guarantees.
-
Differentiable Optimization Layered Safety-Critical Control for Risk-Aware Navigation via Conformal Prediction
A layered differentiable optimization approach combines conformal prediction for risk-aware ellipsoids with control barrier functions and quadratic programming for safe robot navigation under sensor uncertainty.
- S2MAM: Semi-supervised Meta Additive Model for Robust Estimation and Variable Selection