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.

169,051 papers · 148 categories

Trend · papers per month

count % of corpus

48 results for “perturbation framework”

Measuring the Robustness of Graph Propertiescs.SI

Proposes a new framework to measure graph robustness.

problem Lack of control over perturbation strength in existing methods.

method Introduces weighted node perturbation with controllable strength and preserved graph topology.

result Extends robustness measure from statistics to graph properties.

graph robustness perturbation framework graph topology robust statistics

2018-12-03abs ↗pdf ↗

Active Learning of Spin Network Modelscond-mat.dis-nn

Active learning reduces spin network inference complexity by 10^6-fold.

problem Difficulty in inferring direct interactions in complex networks.

method Information geometry framework to quantify inference difficulty and information gain from perturbations.

result Designed perturbations reduce sampling complexity by 10^6-fold across various network architectures.

active learning information geometry spin networks inference complexity complex networks

2019-03-25abs ↗pdf ↗

Perturbation Analysis of Learning Algorithms: A Unifying Perspective on Generation of Adversarial Examplescs.LG

Framework for generating adversarial examples from learning algorithms.

problem Adversarial perturbations leading to erroneous classification.

method Perturbation analysis framework based on convex programming.

result Closed-form solutions for new adversarial attacks.

adversarial examples learning algorithms perturbation analysis convex programming classification attacks

2018-12-15abs ↗pdf ↗

A Perturbation-Based Kernel Approximation Frameworkcs.LG

Paper develops a new kernel approximation framework.

problem High time and space complexity of kernel methods for large datasets.

method Perturbation-based kernel approximation framework using classical perturbation theory.

result Framework generalizes and improves upon existing methods.

kernel approximation Nyström method perturbation theory large datasets time complexity

2020-09-07abs ↗pdf ↗

Robustifying

\ell_\infty

Adversarial Training to the Union of Perturbation Modelscs.LG

Enhances robustness of AT frameworks to multiple perturbations without increasing training complexity.

problem Defending against the union of multiple perturbations in adversarial training.

method SNAP technique that augments a network with shaped noise to enhance robustness.

result 14%-to-20% improvement in adversarial accuracy for ResNet-18 on CIFAR-10.

Adversarial Training Robustness Union of Perturbations SNAP Curvature Reduction

2021-05-31abs ↗pdf ↗

SmoothFool: An Efficient Framework for Computing Smooth Adversarial Perturbationscs.LG

SmoothFool efficiently computes smooth adversarial perturbations for deep networks.

problem Vulnerability of deep neural networks to adversarial attacks with specific statistical properties.

method SmoothFool: a general and computationally efficient framework for computing smooth adversarial perturbations.

result Smoothness significantly enhances robustness against adversarial attacks and improves transferability.

adversarial attacks smooth perturbations deep learning robustness transferability

2019-10-08abs ↗pdf ↗

Quantifying perturbation impacts for large language modelscs.LG

DBPA assesses LLM perturbations using frequentist hypothesis testing.

problem Quantifying input perturbation impacts on LLM outputs.

method DBPA reformulates perturbation analysis as frequentist hypothesis testing, using Monte Carlo sampling for empirical null and alternative distributions.

result DBPA provides interpretable p-values and scalar effect sizes for LLM perturbations.

large language models perturbation analysis frequentist hypothesis testing Monte Carlo sampling semantic similarity

2024-12-01abs ↗pdf ↗

Differentially Private Empirical Risk Minimization with Input Perturbationstat.ML

We propose a novel framework for the differentially private ERM, input perturbation. Existing differentially private ERM implicitly assumed that the data contributors submit their private data to a database expecting that the database invokes a differentially private mechanism for publication of the learned model. In i…

2017-10-20abs ↗pdf ↗

Predicting Cellular Responses with Variational Causal Inference and Refined Relational Informationcs.LG

Novel framework predicts cell responses to perturbations using GRNs.

problem Predicting cellular responses to perturbations for drug discovery and personalized therapeutics.

method Graph variational Bayesian causal inference framework with refined GRNs and robust estimator.

result Enhanced model performance and robust estimation of perturbation effects.

cellular responses perturbation prediction graph variational Bayesian refined GRNs robust estimator

2022-09-30abs ↗pdf ↗

A model for cosmological perturbations in affine gravitygr-qc

We developed a perturbation model for affine gravity theories.

problem Cosmological perturbations in theories without metric.

method Segregated perturbations into symmetric and antisymmetric components, decomposing into irreducible elements.

result Fully addressed gauge freedom in affine gravity theories.

cosmological perturbations affine gravity gauge freedom perturbation theory theories without metric

2024-01-16abs ↗pdf ↗

Towards a Deep Unified Framework for Nuclear Reactor Perturbation Analysiscs.LG

In this paper, we take the first steps towards a novel unified framework for the analysis of perturbations in both the Time and Frequency domains. The identification of type and source of such perturbations is fundamental for monitoring reactor cores and guarantee safety while running at nominal conditions. A 3D Convol…

2018-07-26abs ↗pdf ↗

Universal adversarial examples in speech command classificationcs.LG

Paper finds universal speech command perturbations that fool models.

problem Existence of universal adversarial examples in speech command classification.

method Proposed a novel analytical framework for evaluating universal perturbations and a detailed distortion measurement method.

result Universal perturbations can fool speech command classification models across different models.

adversarial examples speech command classification universal perturbations audio domain machine learning

2019-11-22abs ↗pdf ↗

Multi-Label Adversarial Perturbationscs.LG

This paper tackles adversarial perturbations in multi-label classification problems.

problem Vulnerability and robustness of multi-label learning models under adversarial attacks.

method Proposes a general attacking framework and a ranking-based framework for generating multi-label adversarial perturbations.

result Demonstrates the effectiveness of the proposed frameworks and provides insights into the vulnerability of multi-label deep learning models.

adversarial perturbations multi-label classification deep learning ranking algorithms robustness

2019-01-02abs ↗pdf ↗

Graph Convolutional Neural Networks Sensitivity under Probabilistic Error Modelstat.ML

Paper analyzes GCNN sensitivity to probabilistic graph perturbations.

problem Investigating how GCNNs handle probabilistic graph errors.

method Establishes error bounds and linear relationships between GSO perturbations and GCNN outputs.

result GCNNs maintain stability under graph edge perturbations if GSO errors are bounded.

Graph Convolutional Neural Networks Probabilistic Graph Perturbations Graph Shift Operator Stability Analysis Machine Learning

2022-03-15abs ↗pdf ↗

Generalizable Adversarial Attacks with Latent Variable Perturbation Modellingcs.LG

This work generates diverse adversarial attacks for different domains using latent variable perturbation.

problem Adversarial attacks on deep neural networks are limited to a single perturbation.

method Frame adversarial attacks as learning a distribution of perturbations, enabling generation of diverse attacks.

result Framework generates competitive or superior adversarial attacks across diverse domains (images, text, graphs).

adversarial attacks latent variable domain-agnostic diverse perturbations zero-shot attack

2019-05-26abs ↗pdf ↗

Learning to Generate Noise for Multi-Attack Robustnesscs.LG

Meta framework generates noise to improve multi-attack robustness.

problem Extraneous defense against single type of adversarial perturbation.

method Meta-learning framework with Meta Noise Generator (MNG).

result Significantly outperforms baselines across multiple perturbations.

adversarial robustness meta-learning noise generation multi-attack

2020-06-22abs ↗pdf ↗

Robustness to Adversarial Perturbations in Learning from Incomplete Datastat.ML

Unified framework for learning from incomplete data under adversarial perturbations.

problem The role of unlabeled data in inference when the underlying distribution is adversarially perturbed.

method Unified learning framework combining Semi-Supervised Learning and Distributionally Robust Learning, with a novel generalization theory based on complexity measures.

result The method shows comparable performance to state-of-the-art on real-world datasets.

Adversarial Perturbations Incomplete Data Semi-Supervised Learning Distributionally Robust Learning Generalization Theory

2019-05-24abs ↗pdf ↗

Marginal Weighted Maximum Log-likelihood for Efficient Learning of Perturb-and-Map modelsstat.ML

Efficiently learns perturb-and-map models using weighted log-likelihood.

problem Structured output prediction with weighted Hamming losses.

method Generalizes perturb-and-MAP framework, uses dynamic graph cuts for MAP inference, and double stochastic gradient descent for efficient learning.

result Shows efficiency in learning log-supermodular models with weak supervision.

perturb-and-map log-supermodular weighted Hamming dynamic graph cuts weak supervision

2018-11-21abs ↗pdf ↗

Meta Dropout: Learning to Perturb Features for Generalizationcs.LG

Meta-dropout learns to perturb training examples for better generalization.

problem Achieving good generalization in unseen test examples.

method Meta-learning a noise generator to perturb latent features of training examples.

result Meta-dropout significantly improves generalization performance on few-shot classification datasets.

meta-learning generalization noise generation perturbation few-shot learning

2019-05-30abs ↗pdf ↗

A proposal of a methodological framework with experimental guidelines to investigate clustering stability on financial time seriesq-fin.ST

We present in this paper an empirical framework motivated by the practitioner point of view on stability. The goal is to both assess clustering validity and yield market insights by providing through the data perturbations we propose a multi-view of the assets' clustering behaviour. The perturbation framework is illust…

2015-09-17abs ↗pdf ↗

High Dimensional Inference with Random Maximum A-Posteriori Perturbationscs.LG

This paper presents a new approach, called perturb-max, for high-dimensional statistical inference that is based on applying random perturbations followed by optimization. This framework injects randomness to maximum a-posteriori (MAP) predictors by randomly perturbing the potential function for the input. A classic re…

2016-02-10abs ↗pdf ↗

A new gauge for gravitational perturbations of Kerr spacetimes I: The linearised theorygr-qc

New framework for gravitational perturbations of Kerr spacetimes, focusing on stability.

problem Stability of Kerr spacetimes to gravitational perturbations.

method New geometric framework with tailored null frames and gauge, reformulating Einstein equations.

result Derivation of linearised vacuum Einstein equations in the new framework.

Kerr spacetimes gravitational perturbations stability null frames geometric gauge

2022-11-01abs ↗pdf ↗

Fair Classification with Adversarial Perturbationscs.LG

New framework for fair classification in adversarial settings with provable guarantees.

problem Fairness in classification with adversarial perturbations of protected attributes.

method Optimization framework for learning fair classifiers with provable guarantees.

result Near-tightness of accuracy and fairness guarantees for multiple protected attributes and various hypothesis classes.

fair classification adversarial perturbations provable guarantees linear-fractional fairness protected attributes

2021-06-10abs ↗pdf ↗

Automatic Perturbation Analysis for Scalable Certified Robustness and Beyondcs.LG

Automates perturbation analysis for neural networks, enabling certified robustness on complex architectures.

problem Limited applicability of existing perturbation analysis methods to complex neural network architectures.

method Developed an automatic framework to generalize LiRPA algorithms to any neural network structure, enabling loss fusion and state-of-the-art certified defense results.

result Demonstrated LiRPA based certified defense on Tiny ImageNet and Downscaled ImageNet.

perturbation analysis neural networks certified robustness loss fusion automatic framework

2020-02-28abs ↗pdf ↗

Learning to Generate Image Source-Agnostic Universal Adversarial Perturbationscs.LG

New method generates universal adversarial perturbations across different image sources.

problem Certifying robustness of deep learning models with universal adversarial perturbations across various image sources.

method Few-shot learning approach using bilevel optimization and learning-to-optimize techniques.

result Improved attack success rate and faster performance compared to existing methods.

adversarial attacks universal perturbations few-shot learning meta-learning learning-to-optimize

2020-09-29abs ↗pdf ↗

BioBO: Biology-informed Bayesian Optimization for Perturbation Designstat.ML

BioBO optimizes gene perturbation design using Bayesian optimization with biological priors.

problem Efficient design of genomic perturbation experiments in drug discovery.

method Integrates Bayesian optimization with multimodal gene embeddings and enrichment analysis.

result Improves labeling efficiency by 25-40% and identifies top-performing perturbations more effectively.

Bayesian optimization genomic perturbation drug discovery biological prior pathway-level explanations

2025-09-24abs ↗pdf ↗

Towards non-perturbative BV-theory via derived differential geometryhep-th

Introduces a new geometric framework for non-perturbative BV-theory.

problem Non-perturbative generalization of BV-theory in infinite-dimensional spaces.

method Derived differential geometry and homotopical algebraic geometry.

result Concrete model of derived smooth stacks for encoding non-perturbative BV-theory.

BV-theory derived differential geometry non-perturbative formal stacks infinite-dimensional spaces

2023-07-27abs ↗pdf ↗

Perturbation is All You Need for Extrapolating Language Modelsstat.ML

A new method improves language model extrapolation without changing training.

problem Improving language model performance on sequences not seen during training.

method Perturbation-based training procedure for language models.

result The method consistently improves out-of-support prediction.

language models extrapolation perturbation next-token prediction model extrapolability

2026-05-05abs ↗pdf ↗

Perturbation-Assisted Sample Synthesis: A Novel Approach for Uncertainty Quantificationstat.ML

Novel framework uses synthetic data to quantify uncertainty in complex data.

problem Uncertainty quantification in complex, unstructured data.

method Perturbation-Assisted Sample Synthesis (PASS) and Perturbation-Assisted Inference (PAI) framework.

result Statistically guaranteed validity in inference, enhancing reliability of synthetic data.

uncertainty quantification synthetic data deep learning data perturbation statistical inference

2023-05-30abs ↗pdf ↗

Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Frameworkcs.LG

New framework improves adversarial robustness certification for various perturbations.

problem Certifying robustness against adversarial attacks in deep learning models.

method Unified functional optimization approach with non-Gaussian smoothing noise for multiple types of attacks.

result Achieves better certification results and identifies key trade-offs between accuracy and robustness.

adversarial robustness randomized smoothing functional optimization certification deep learning

2020-02-21abs ↗pdf ↗

Nonconvex proximal splitting: batch and incremental algorithmsmath.OC

Within the unmanageably large class of nonconvex optimization, we consider the rich subclass of nonsmooth problems that have composite objectives---this already includes the extensively studied convex, composite objective problems as a special case. For this subclass, we introduce a powerful, new framework that permits…

2011-09-01abs ↗pdf ↗

Perturbed Iterate Analysis for Asynchronous Stochastic Optimizationstat.ML

We introduce and analyze stochastic optimization methods where the input to each gradient update is perturbed by bounded noise. We show that this framework forms the basis of a unified approach to analyze asynchronous implementations of stochastic optimization algorithms.In this framework, asynchronous stochastic optim…

2015-07-24abs ↗pdf ↗

A PAC-Bayes Analysis of Adversarial Robustnesscs.LG

PAC-Bayesian bounds estimate adversarial robustness.

problem Estimating robustness to imperceptible input perturbations.

method PAC-Bayesian framework for averaging over hypotheses.

result General bounds valid for any type of adversarial attacks.

PAC-Bayes Adversarial Robustness Robustness Generalization Perturbation

2021-02-19abs ↗pdf ↗

The pseudo-differential perturbations of the Dirac operator and the Kastler-Kalau-Walze type theoremsmath.DG

Study perturbs Dirac operator on 4D manifolds, proving Kastler-Kalau-Walze theorems.

problem Analyzing perturbations of Dirac operator on compact manifolds.

method Defining pseudo-differential perturbations and proving Kastler-Kalau-Walze theorems.

result Proved Kastler-Kalau-Walze theorems for 4D compact manifolds with boundary.

Dirac operator pseudo-differential perturbations noncommutative geometry Kastler-Kalau-Walze theorems compact manifolds

2024-01-09abs ↗pdf ↗

Improved Image Wasserstein Attacks and Defensescs.LG

We improve image perturbation defenses using a better-defined Wasserstein threat model.

problem Real-world image perturbations are not pixel-independent, unlike

\ell_p

threat models.

method We rectify flaws in the Wasserstein threat model and explore stronger attacks and defenses.

result Current Wasserstein-robust models are ineffective against real-world perturbations.

image perturbations Wasserstein distance robustness defenses threat models

2020-04-26abs ↗pdf ↗

Theoretical Understanding of Learning from Adversarial Perturbationscs.LG

The study reveals how adversarial perturbations can include class features for generalization.

problem Understanding why adversarial examples deceive neural networks and transfer between networks.

method A one-hidden-layer network trained on mutually orthogonal samples.

result Adversarial perturbations, even of a few pixels, contain sufficient class features for generalization.

adversarial examples neural networks generalization class features decision boundary

2024-02-16abs ↗pdf ↗

Inverse Classification with Limited Budget and Maximum Number of Perturbed Samplescs.LG

New framework maximizes perturbed samples for inverse classification with budget constraints.

problem Maximizing perturbed samples for desired classification outcomes under budget constraints.

method Gradient methods, stochastic processes, Lagrangian relaxations, Gumbel trick.

result Stochastic process-based algorithms outperform in different budget settings.

inverse classification machine learning budget constraints perturbation classification accuracy

2020-09-29abs ↗pdf ↗

Learning Perturbations to Extrapolate Your LLMstat.ML

Learnable token perturbations boost extrapolation in LLMs.

problem Limited flexibility of current discrete perturbations in large language models.

method Learnable continuous latent vector transformations in embedding space, unbiased estimating equations, stochastic gradient descent optimization.

result Significant gains in out-of-domain settings over state-of-the-art methods.

large language models extrapolation perturbations continuous vectors embedding space

2026-05-13abs ↗pdf ↗

Learning Maximum-A-Posteriori Perturbation Models for Structured Prediction in Polynomial Timecs.LG

MAP perturbation models have emerged as a powerful framework for inference in structured prediction. Such models provide a way to efficiently sample from the Gibbs distribution and facilitate predictions that are robust to random noise. In this paper, we propose a provably polynomial time randomized algorithm for learn…

2018-05-21abs ↗pdf ↗

Causal Structure and Representation Learning with Biomedical Applicationscs.LG

Framework for causal discovery using multi-modal data.

problem Failure of representation learning in causal tasks.

method Statistical and computational framework combining representation learning and causal inference.

result Effective use of observational and perturbational data for causal discovery.

representation learning causal inference multi-modal data biomedical applications observational data

2025-11-06abs ↗pdf ↗

Learning Causal States Under Partial Observability and Perturbationcs.LG

Enhances RL in partially observable, noisy environments by uncovering causal states.

problem Making decisions based on incomplete and noisy observations in partially observable Markov decision processes (P

^2

OMDPs).

method Causal State Representation under Asynchronous Diffusion Model (CaDiff) framework, incorporating a novel asynchronous diffusion model (ADM) and a new bisimulation metric.

result Enhances returns by at least 14.18% compared to baselines on Roboschool tasks.

Reinforcement Learning Partial Observability Perturbation Causal States Diffusion Models

2025-11-29abs ↗pdf ↗

Self-Concordant Perturbations for Linear Banditsstat.ML

Unified algorithm for linear bandits with improved regret bound.

problem Adversarial linear bandits with improved regret.

method Self-concordant perturbations in FTPL framework.

result Regret bound of

\mathcal{O}(d\sqrt{n \ln n})

for hypercube and

\ell_2

ball.

linear bandits self-concordant perturbations FTPL regret bound adversarial setting

2025-10-28abs ↗pdf ↗

Robust normalizing flows using Bernstein-type polynomialscs.LG

This paper improves flow models to better handle perturbations in real-world data.

problem Flow models amplify initial errors in perturbed data, leading to poor generalization.

method Utilizes Bernstein-type polynomials to construct Normalizing Flows (NF) for higher robustness.

result Proposed NF framework provides theoretical upper bounds and practical advantages.

Normalizing Flows Bernstein polynomials robustness perturbations data perturbation

2021-02-06abs ↗pdf ↗

On the Reducibility of Submodular Functionscs.LG

The scalability of submodular optimization methods is critical for their usability in practice. In this paper, we study the reducibility of submodular functions, a property that enables us to reduce the solution space of submodular optimization problems without performance loss. We introduce the concept of reducibility…

2016-01-04abs ↗pdf ↗

μP

^2

: Effective Sharpness Aware Minimization Requires Layerwise Perturbation Scalingcs.LG

New method $μP^2$ improves neural network training by scaling perturbations layerwise.

problem Improving neural network performance as models scale up.

method Layerwise perturbation scaling in the infinite-width limit of neural networks.

result Layerwise perturbation scaling ensures all layers are effectively perturbed in the limit.

neural networks sharpness aware minimization infinite-width limit layerwise perturbation maximal update and perturbation parameterization

2024-10-31abs ↗pdf ↗

Perturbations and Stability of Black Ellipsoidsgr-qc

We study the perturbations of two classes of static black ellipsoid solutions of four dimensional vacuum Einstein equations. Such solutions are described by generic off--diagonal metrics which are generated by anholonomic transforms of diagonal metrics. The analysis is performed in the approximation of small eccentrici…

2002-06-05abs ↗pdf ↗

Higher-Order Singular-Value Derivatives of Rectangular Real Matricesstat.ML

New framework for higher-order singular-value derivatives of rectangular matrices.

problem Challenging to derive higher-order Fréchet derivatives of singular values in real rectangular matrices.

method Using Kato's analytic perturbation theory for self-adjoint operators and embedding rectangular matrices into block self-adjoint operators.

result Closed-form expressions for the

n

-th order spectral variations of singular values.

singular values rectangular matrices Fréchet derivatives Kato's theory Hessian

2025-06-04abs ↗pdf ↗

Faster Perturbed Stochastic Gradient Methods for Finding Local Minimamath.OC

Paper proposes faster method to find local minima in nonconvex optimization.

problem Escaping saddle points and finding local minima in nonconvex optimization.

method LENA (Last stEp shriNkAge) framework for faster perturbed stochastic gradient methods.

result LENA finds

(ε, ε_{H})

-approximate local minima within

ilde O(ε^{-3} + ε_{H}^{-6})

evaluations.

nonconvex optimization local minima saddle points stochastic gradient perturbed methods

2021-10-25abs ↗pdf ↗