generalization-bounds

Tag

Cards List
#generalization-bounds

The Sharp Tail of Uniform Stability

arXiv cs.LG · 2026-08-26 Cached

This paper presents a new logarithmic-free upper bound for the generalization gap in uniformly stable algorithms and constructs a deterministic learning problem that achieves optimal high-probability dependence, closing a gap in the literature.

0 favorites 0 likes
#generalization-bounds

When Does More Correct Data Hurt? Insertion-Stability and the Limits of Dimension-Based Theory

arXiv cs.LG · 2026-08-17 Cached

This paper investigates scenarios where adding correctly labeled data can harm model performance, introducing insertion-stability and examining the limits of dimension-based theory in machine learning generalization.

0 favorites 0 likes
#generalization-bounds

Drift and Dependence: Layer-wise Information-Theoretic Bounds for Replay-Based Continual Learning

arXiv cs.LG · 2026-08-13 Cached

This paper presents a layer-wise information-theoretic framework for replay-based continual learning, decomposing the generalization gap into replay-induced representation drift and optimization-dependence terms, with refinements via Wasserstein relaxation and SGLD instantiation.

0 favorites 0 likes
#generalization-bounds

PAC-Bayes Beyond Parameter Space: Behavioral Equivalence, Z-Information, and Exact Complexity Decomposition

arXiv cs.LG · 2026-08-13 Cached

This paper extends PAC-Bayes theory by decomposing its complexity measure using behavioral equivalence, introducing PAC-Bayes Z-information and exact structural decomposition beyond parameter space.

0 favorites 0 likes
#generalization-bounds

PAC--Bayes Bounds on Quotient Parameter Spaces: Geometry-induced Implicit-Bias Priors

arXiv cs.LG · 2026-07-22 Cached

This paper proposes performing PAC-Bayesian analysis on quotient parameter spaces to remove KL contributions from parameter symmetries, and constructs a geometry-induced prior that approximates the ideal posterior-matched prior, resulting in tighter generalization bounds. Experiments on Fourier regression and Query-Key attention show significant reductions in KL divergence and certificate values.

0 favorites 0 likes
#generalization-bounds

@ddkang: New research from Bridgewater AIA Labs, UIUC, and MIT: we prove what we believe to be the first non-vacuous generalizat…

X AI KOLs Timeline · 2026-07-20 Cached

Researchers from Bridgewater AIA Labs, UIUC, and MIT prove the first non-vacuous generalization bounds for reasoning LLMs trained with RLVR, providing provable accuracy lower bounds on unseen data to guide safe deployment.

0 favorites 0 likes
#generalization-bounds

Ghost in the Kernel: In-Context Learning with Efficient Transformers via Domain Generalization

arXiv cs.LG · 2026-07-02 Cached

This paper theoretically analyzes linear transformers for in-context learning under domain generalization, establishing dimension-independent convergence rates and proposing novel activation and loss designs for linearizing pretrained softmax LLMs.

0 favorites 0 likes
#generalization-bounds

Uncertainty Estimation and Generalization Bounds for Modern Deep Learning

arXiv cs.LG · 2026-06-15 Cached

This paper presents theoretical bounds for uncertainty estimation and generalization in modern deep learning models.

0 favorites 0 likes
#generalization-bounds

Regime-Arrival Uncertainty in Generalization Bounds under Distribution Shift

arXiv cs.LG · 2026-06-03 Cached

This paper introduces a theoretical framework for quantifying deployment risk when training and deployment distributions differ due to latent regime dynamics modeled as a Markov-switching process, providing exact decomposition and finite-sample bounds.

0 favorites 0 likes
#generalization-bounds

MMD-Balls as Credal Sets: A PAC-Bayesian Framework for Epistemic Uncertainty in Test-Time Adaptation

arXiv cs.LG · 2026-05-22 Cached

This paper develops a PAC-Bayesian framework for test-time adaptation that uses MMD-balls as credal sets, providing formal generalization bounds and separating epistemic from aleatoric uncertainty under distribution shift.

0 favorites 0 likes
#generalization-bounds

Generalization Bounds of Emergent Communications for Agentic AI Networking

arXiv cs.AI · 2026-05-12 Cached

This paper proposes an information-theoretic framework for emergent communication in Agentic AI Networking (AgentNet), addressing physical constraints and providing generalization bounds. Experimental validation on hardware prototypes demonstrates improved generalization performance compared to state-of-the-art solutions.

0 favorites 0 likes
← Back to home

Submit Feedback