Research
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

66132197263 · Jun 202019922001200920182026
48 results for labeling discrepancy

A new discrepancy method improves efficiency and accuracy in unsupervised domain adaptation.

problem Measuring the difference between source and target domains without labeled target data.
method Source-guided discrepancy (S-disc) that uses source domain labels for efficient computation and tighter generalization bounds.
result S-disc provides a tighter generalization error bound than existing discrepancies.

Private algorithms adapt from public to private domains with minimal labeled data.

problem Adapting from a public source domain to a private target domain with few labeled data.
method Differentially private discrepancy minimization algorithms based on Frank-Wolfe and Mirror-Descent methods.
result Effective adaptation with strong generalization and privacy guarantees.

Study shows cross-domain X-ray prediction performance discrepancies and label shifts.

problem Quantifying generalization limits across different X-ray datasets.
method Large-scale study on multiple X-ray datasets, focusing on performance and label shifts.
result Interesting discrepancies found between model performance and agreement, and concept similarity across tasks.

This work investigates how GCNs should handle local structure discrepancies in testing nodes.

problem GCNs assume homophily but real graphs often have discrepancies in local structure.
method Using causal graph analysis, the study intervenes the graph structure to assess the local structure's impact on predictions.
result The method effectively enhances GCN predictions by eliminating local structure discrepancies.

Stein discrepancy improves UDA performance in low-data scenarios.

problem Improving model performance on unlabeled target domains with limited data.
method Proposes a novel UDA framework using Stein discrepancy, an asymmetric measure that depends on the target distribution through its score function.
result Consistently outperforms prior UDA approaches under limited target data across multiple benchmarks.

Active learning algorithms propose which unlabeled objects should be queried for their labels to improve a predictive model the most. We study active learners that minimize generalization bounds and uncover relationships between these bounds that lead to an improved approach to active learning. In particular we show th…

2017-06-08abs ↗pdf ↗

Proposes TFDF to learn transferable and discriminative features for unsupervised domain adaptation.

problem Difficult to induce supervised classifier without labeled data in unsupervised domain adaptation.
method TFDF optimizes transferability and discriminability by aligning distributions and minimizing class confusion.
result TFDF achieves better performance on real-world datasets compared to existing methods.

Unified analysis of generalization and sample complexity for semi-supervised domain adaptation.

problem Theoretical foundations of domain adaptation remain underexplored, especially for modern approaches.
method Unified theoretical study of domain adaptation algorithms based on domain alignment, considering joint learning of feature transformations and shared classifiers in a semi-supervised setting.
result Unified theoretical analysis of domain adaptation algorithms, providing generalization bounds and sample complexity bounds for MMD and adversarial models.

CoSCA improves unsupervised domain adaptation by better aligning ambiguous target samples.

problem Missing alignment of ambiguous target samples in unsupervised domain adaptation.
method CoSCA explicitly incorporates intra- and inter-class domain discrepancy, estimating label hypotheses and optimizing a contrastive loss with MMD for better global alignment.
result CoSCA outperforms state-of-the-art approaches in producing more discriminative features.

New method tackles label noise on imbalanced datasets by considering class-specific uncertainty.

problem Label noise and class imbalance in imbalanced datasets.
method Epistemic and aleatoric uncertainty-aware class-specific noise modeling.
result Proposed ULC framework improves performance on imbalanced datasets.

New bounds for contrastive learning handle domain shifts and generalization.

problem Domain shifts and generalization challenges in downstream tasks.
method Novel generalization bounds accounting for both domain shift and generalization.
result Performance of contrastively learned representations depends on statistical discrepancy between pretraining and downstream distributions.

Domain adaptation is transfer learning which aims to generalize a learning model across training and testing data with different distributions. Most previous research tackle this problem in seeking a shared feature representation between source and target domains while reducing the mismatch of their data distributions.…

2017-04-13abs ↗pdf ↗

W2S FT often outperforms weak teachers due to low intrinsic dimensionality.

problem Understanding why weak-to-strong finetuning outperforms weak models.
method Analyzing W2S in ridgeless regression setting, focusing on variance reduction.
result Weak teacher's variance is inherited by strong student in shared feature subspace, reduced in discrepancy subspace.

A new method compares image classifiers using adaptive sampling of natural images.

problem Evaluation of image classifiers on small, fixed test sets may not generalize to real-world images.
method Adaptive sampling from a large corpus of unlabeled images to maximize classifier discrepancies measured by WordNet hierarchy.
result Human labeling of model-dependent image sets reveals relative classifier performance.

Paper tackles noise-robust domain adaptation in noisy environments.

problem Learning machines struggle with domain adaptation in noisy environments.
method The paper proposes offline curriculum learning and proxy distribution based margin discrepancy to mitigate label and feature noise.
result The proposed algorithm significantly outperforms state-of-the-art methods in noisy environments.

Paper tackles backwards-compatible data adaptation for confounded covariate and label shifts.

problem Adapt covariates to predict labels confounded with covariate shifts.
method Proposes confounded shift framework based on minimizing divergence between source and target conditional distributions, conditioning on confounders.
result Demonstrates approach on synthetic and real datasets, achieving backwards-compatible data adaptation.

This work proposes a novel method for semi-supervised learning from partially labeled massive network-structured datasets, i.e., big data over networks. We model the underlying hypothesis, which relates data points to labels, as a graph signal, defined over some graph (network) structure intrinsic to the dataset. Follo…

2016-12-05abs ↗pdf ↗

Method generates natural ECGs with 25 interpretable features.

problem Lack of labeled ECGs for supervised learning and automatic diagnostics.
method Variational autoencoder for ECG generation and feature extraction.
result Low Maximum Mean Discrepancy (0.00383) indicates good ECG generation quality.

We study the task of unsupervised domain adaptation, where no labeled data from the target domain is provided during training time. To deal with the potential discrepancy between the source and target distributions, both in features and labels, we exploit a copula-based regression framework. The benefits of this approa…

2017-09-29abs ↗pdf ↗

A semi-supervised learning method using predefined class centroids for image classification.

problem Reducing the need for labeled data in deep learning.
method Use a small number of labeled samples and data augmentation on unlabeled samples. Constrain all samples to predefined evenly-distributed class centroids (PEDCC) using loss functions.
result Achieves state-of-the-art results with minimal labeled data.

This paper introduces localized discrepancy theories for unsupervised domain adaptation.

problem Improving generalization bounds for unsupervised domain adaptation.
method Localized discrepancies defined on the hypothesis space after localization, leading to smaller and asymmetric values.
result Improved generalization bounds and sample complexity reduction.

New discrepancy function compares discrete probability measures considering space geometry.

problem Comparing discrete probability measures in a geometrically meaningful way.
method Proposes the Fourier Discrepancy Function, proving convexity, differentiability, and providing gradient formula.
result Proves the Fourier Discrepancy is convex, twice differentiable, and provides an explicit gradient formula.

The article introduces practical estimators for kernel discrepancies.

problem Estimating kernel discrepancies accurately and efficiently.
method Presented various estimators for MMD, HSIC, and KSD, including V-statistics, U-statistics, and incomplete U-statistics. Stressed the importance of kernel bandwidth and introduced adaptive estimators.
result Adaptive estimators combining multiple estimators with various kernels address the problem of kernel selection.

Improves unsupervised domain adaptation by mixing source and target domains.

problem Improves unsupervised domain adaptation by mixing source and target domains.
method Enforces training constraints across domains using mixup formulation and feature-level consistency regularizer.
result Significantly improves state-of-the-art performance on image classification and human activity recognition tasks.

The paper tackles hypothesis testing for likelihood-free inference with a new kernel-based approach.

problem Testing hypotheses with limited labeled data in likelihood-free inference.
method Kernel-based tests using maximum mean discrepancy (MMD) for non-parametric density comparison.
result Existence of an asymmetric trade-off between labeled and unlabeled data samples.

Proposes PHD to measure domain discrepancy for complex models.

problem Insufficient domain discrepancy measures for complex models.
method Introduces PHD, a novel discrepancy measure for complex models.
result PHD is computationally efficient and applicable to multi-class classification.

The paper explores how neural networks generalize differently from natural and medical images.

problem Discrepancies in generalization error between natural and medical images.
method Established and empirically validated a generalization scaling law with respect to intrinsic dataset properties.
result Higher intrinsic 'label sharpness' of medical images leads to higher adversarial vulnerability.

Paper defines class discrepancy for machine learning problems and provides coresets.

problem Addressing discrepancies in machine learning models.
method Defines class discrepancy, provides techniques for bounding discrepancy, and develops coresets and streaming sketches.
result Establishes coresets of size O(sqrt{d}/epsilon) for various machine learning problems.

Sliced kernelized Stein discrepancy improves goodness-of-fit tests and model learning in high dimensions.

problem The curse-of-dimensionality in kernelized Stein discrepancy (KSD).
method Sliced Stein discrepancy and its scalable variants using optimal one-dimensional projections.
result Significantly outperforms KSD and baselines in goodness-of-fit tests and improves model learning.

Study shows the corrected Akaike criterion is inadmissible for estimating Kullback-Leibler discrepancy.

problem Inadmissibility of the corrected Akaike information criterion for estimating Kullback-Leibler discrepancy.
method Loss estimation framework to demonstrate inadmissibility and provide improved estimators.
result Improved estimators of Kullback-Leibler discrepancy are provided and perform well in reduced-rank situations.

Framework synthesizes geological images minimizing patch distribution discrepancy.

problem Synthesizing realistic geological images from a single exemplar.
method Uses kernel discrepancies and generative neural networks for efficient synthesis.
result Synthesized images match visual patterns and spatial statistics of the exemplar.