Establishes maximal concentration bounds for stochastic approximation under heavy-tailed Markovian noise, with tails ranging from sub-Gaussian to heavier than Weibull depending on step sizes and contractivity properties, plus a truncation argument for unbounded noise.
Title resolution pending
16 Pith papers cite this work. Polarity classification is still indexing.
16
Pith papers citing it
citation-role summary
background 2
method 1
citation-polarity summary
years
2026 16representative citing papers
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
Any convex L-Lipschitz functional on a compact convex subset of a separable Hilbert space can be uniformly approximated to arbitrary accuracy by an explicit convex L-Lipschitz reconstruction from finitely many linear measurements, exactly implementable by a ReLU-MLP.
OptiVerse is a new benchmark spanning neglected optimization domains that shows LLMs suffer sharp accuracy drops on hard problems due to modeling and logic errors, with a Dual-View Auditor Agent proposed to improve performance.
Temperature scaling of density-matrix eigenvalues from LLM semantic embeddings optimizes proper-score calibration and corrects systematic overconfidence so entropy equals risk.
Gives necessary and sufficient exactness certificates based on primal and dual feasibility for when a violated-set closed-form correction equals the exact Euclidean projection in CBF safety filters.
SeqLoRA applies bilevel optimization to sequential LoRA adaptation for continual multi-concept text-to-image generation with theoretical bounds on forgetting and interference.
W-SparQ-BL models time-varying lower-level responses with multi-output GPs and sparse approximations to achieve sublinear dynamic regret in bilevel optimization under noise.
Proves that stabilizing switched linear systems with full-history controllers implies existence of memoryless homogeneous degree one controllers depending only on current state (and mode if applicable).
SNMPP builds a product-form neural influence kernel from a signed interaction network over event classes and a delay-aware monotonic temporal network to enable explicit discovery of inter-event relationships alongside strong prediction.
PowerStep delivers coordinate-wise adaptive optimization by nonlinearly transforming a momentum buffer under an lp-norm steepest-descent geometry, matching Adam convergence with half the memory and supporting aggressive quantization.
New extended-variable relaxations are derived for CGMESP that generalize prior bounds for CMESP and binary D-optimality and are tested numerically inside branch-and-bound.
POSCMs extend SCMs to settings where the causal graph itself is generated by latent context and can be intervened on, with conditional kernel-identifiability theorems and illustrative retina simulations.
BOOOM parametrizes Stiefel manifold optimization into Euclidean angle space using global Givens rotations and solves it with recursive modified pattern search for loss-agnostic black-box problems.
The paper proposes Trajectory Regularized Merging (TRM) to enable storage-free model merging in continual learning by optimizing in an augmented trajectory subspace with task alignment, prediction consistency, and gradient responsiveness objectives, claiming SOTA results.
citing papers explorer
-
Concentration of General Stochastic Approximation Under Heavy-Tailed Markovian Noise
Establishes maximal concentration bounds for stochastic approximation under heavy-tailed Markovian noise, with tails ranging from sub-Gaussian to heavier than Weibull depending on step sizes and contractivity properties, plus a truncation argument for unbounded noise.
-
Sobolev Regularized MMD Gradient Flow
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
-
Structure-Preserving Reconstruction of Convex Lipschitz Functionals on Hilbert Spaces from Finite Samples
Any convex L-Lipschitz functional on a compact convex subset of a separable Hilbert space can be uniformly approximated to arbitrary accuracy by an explicit convex L-Lipschitz reconstruction from finitely many linear measurements, exactly implementable by a ReLU-MLP.
-
OptiVerse: A Comprehensive Benchmark towards Optimization Problem Solving
OptiVerse is a new benchmark spanning neglected optimization domains that shows LLMs suffer sharp accuracy drops on hard problems due to modeling and logic errors, with a Dual-View Auditor Agent proposed to improve performance.
-
Eigenvalue Calibration for Semantic Embeddings of Large Language Models
Temperature scaling of density-matrix eigenvalues from LLM semantic embeddings optimizes proper-score calibration and corrects systematic overconfidence so entropy equals risk.
-
Exactness Certificates for Closed-Form CBF Safety-Filter Projections
Gives necessary and sufficient exactness certificates based on primal and dual feasibility for when a violated-set closed-form correction equals the exact Euclidean projection in CBF safety filters.
-
SeqLoRA: Bilevel Orthogonal Adaptation for Continual Multi-Concept Generation
SeqLoRA applies bilevel optimization to sequential LoRA adaptation for continual multi-concept text-to-image generation with theoretical bounds on forgetting and interference.
-
No-regret optimization of time-varying bilevel problems
W-SparQ-BL models time-varying lower-level responses with multi-output GPs and sparse approximations to achieve sublinear dynamic regret in bilevel optimization under noise.
-
Feedback Stabilization of Switched Systems: Memory is not needed
Proves that stabilizing switched linear systems with full-history controllers implies existence of memoryless homogeneous degree one controllers depending only on current state (and mode if applicable).
-
Structured Neural Marked Point Processes for Interpretable Event Interaction Modeling
SNMPP builds a product-form neural influence kernel from a signed interaction network over event classes and a delay-aware monotonic temporal network to enable explicit discovery of inter-event relationships alongside strong prediction.
-
PowerStep: Memory-Efficient Adaptive Optimization via $\ell_p$-Norm Steepest Descent
PowerStep delivers coordinate-wise adaptive optimization by nonlinearly transforming a momentum buffer under an lp-norm steepest-descent geometry, matching Adam convergence with half the memory and supporting aggressive quantization.
-
Extended-variable relaxations for the constrained generalized maximum-entropy sampling problem
New extended-variable relaxations are derived for CGMESP that generalize prior bounds for CMESP and binary D-optimality and are tested numerically inside branch-and-bound.
-
Partially Observed Structural Causal Models
POSCMs extend SCMs to settings where the causal graph itself is generated by latent context and can be intervened on, with conditional kernel-identifiability theorems and illustrative retina simulations.
-
BOOOM: Loss-Function-Agnostic Black-Box Optimization over Orthonormal Manifolds for Machine Learning and Statistical Inference
BOOOM parametrizes Stiefel manifold optimization into Euclidean angle space using global Givens rotations and solves it with recursive modified pattern search for loss-agnostic black-box problems.
-
Revitalizing the Beginning: Avoiding Storage Dependency for Model Merging in Continual Learning
The paper proposes Trajectory Regularized Merging (TRM) to enable storage-free model merging in continual learning by optimizing in an augmented trajectory subspace with task alignment, prediction consistency, and gradient responsiveness objectives, claiming SOTA results.
- Optimizing Trajectory-Trees in Belief Space: An Application from Model Predictive Control to Task and Motion Planning