On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

10 results for “supersampling”

Tighter Information-Theoretic Generalization Bounds from Supersamplesstat.ML

New tighter bounds for learning algorithms from Steinke & Zakynthinou's supersample setting.

problem Improving generalization bounds for machine learning algorithms.

method Information-theoretic approach using projected loss and Rademacher sequence.

result The new bounds are tighter than previous information-theoretic bounds.

information-theoretic bounds supersample setting machine learning generalization learning algorithms

2023-02-05abs ↗pdf ↗

Information-Theoretic Generalization Bounds for Sequential Decision Makingstat.ML

Develops a new framework for analyzing sequential decision-making problems using information theory.

problem Lack of information-theoretic generalization bounds for sequential decision-making problems.

method Introduces a sequential supersample framework that separates learner filtration from proof-side enlargement, controlling the generalization gap by sequential CMI.

result Establishes a sequential CMI that controls the generalization gap in sequential decision-making problems.

information theory sequential decision making generalization bounds supersample construction exchangeability

2026-05-12abs ↗pdf ↗

Information-Theoretic Generalization Bounds for Transductive Learning and its Applicationscs.LG

The paper establishes bounds for transductive learning using information theory.

problem Transductive learning generalization gap control.

method Information theory, PAC-Bayes, mutual information, conditional mutual information, different information measures.

result Established transductive information-theoretic and PAC-Bayesian bounds.

transductive learning information theory PAC-Bayes mutual information conditional mutual information

2023-11-08abs ↗pdf ↗

Generalization Bounds via Conditional

f

-Informationstat.ML

New bounds derived using conditional $f$ -information for machine learning models.

problem Improving generalization bounds in machine learning.

method Introducing novel information-theoretic generalization bounds via conditional

f

-information.

result Derives generalization bounds applicable to both bounded and unbounded loss functions.

machine learning generalization information theory conditional $f$-information supersample setting

2024-10-30abs ↗pdf ↗

Conditional Mutual Information-Based Generalization Bound for Meta Learningcs.LG

Meta-learning bound uses conditional mutual information.

problem Bounding generalization performance in meta-learning.

method Extends CMI framework to meta-learning with a meta-supersample.

result Explicit bound involving two CMI terms.

meta-learning generalization information-theoretic conditional mutual information bound

2020-10-21abs ↗pdf ↗

Adaptive Learning Rate Clipping Stabilizes Learningcs.LG

Artificial neural network training with stochastic gradient descent can be destabilized by "bad batches" with high losses. This is often problematic for training with small batch sizes, high order loss functions or unstably high learning rates. To stabilize learning, we have developed adaptive learning rate clipping (A…

2019-06-21abs ↗pdf ↗

A Hierarchical Sampling Framework for bounding the Generalization Error of Federated Learningcs.LG

Hierarchical Federated Learning bounds generalize using Wasserstein distance.

problem Bounding generalization error in Federated Learning with hierarchical sampling.

method Introduced a hierarchical sampling framework and derived generalization bounds using Wasserstein distance.

result Recover and strictly imply existing CMI bounds for bounded losses.

Federated Learning Hierarchical Sampling Generalization Error Wasserstein Distance Conditional Mutual Information

2026-05-05abs ↗pdf ↗

Towards the Generalization of Multi-view Learning: An Information-theoretical Analysisstat.ML

This paper analyzes multi-view learning using information theory to improve generalization.

problem Lack of theoretical understanding of multi-view learning's generalization behavior.

method Developed information-theoretic generalization bounds for multi-view learning.

result Capturing both consensus and complementary information maximizes representation disentanglement.

multi-view learning generalization information theory reconstruction classification

2025-01-28abs ↗pdf ↗

Generalization in Federated Learning: A Conditional Mutual Information Frameworkstat.ML

This work analyzes generalization in federated learning using information theory.

problem Generalization performance in federated learning is less explored compared to centralized learning.

method The work applies an information-theoretic analysis via the conditional mutual information (CMI) framework to study federated learning's two-level generalization.

result The work derives multiple CMI-based bounds, including hypothesis-based CMI bounds and fast-rate evaluated CMI bounds, which improve convergence rates for specific model aggregation strategies and structured loss functions.

Federated Learning Generalization Conditional Mutual Information Privacy Preserving Convergence Rate

2025-03-06abs ↗pdf ↗

Optimal SQ Lower Bounds for Learning Halfspaces with Massart Noisecs.LG

The study sets limits on how well halfspaces can be learned when labels are corrupted.

problem Learning halfspaces in the presence of Massart noise.

method Statistical query (SQ) lower bounds.

result No SQ algorithm can achieve misclassification error better than the corruption rate

η

with superpolynomial accuracy or a superpolynomial number of queries.

statistical query learning halfspaces Massart noise lower bounds supersampling

2022-01-24abs ↗pdf ↗