Cover Image

Gradient-Weight Alignment as a Train-Time Proxy for Generalization in Classification Tasks

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hölzl, Florian A., Rueckert, Daniel, Kaissis, Georgios
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

How Much Is One Recurrence Worth? Iso-Depth Scaling Laws for Looped Language Models
by: Schwethelm, Kristian, et al.
Published: (2026)

Incentivising the federation: gradient-based metrics for data selection and valuation in private decentralised training
by: Usynin, Dmitrii, et al.
Published: (2023)

ChEX: Interactive Localization and Region Description in Chest X-rays
by: Müller, Philip, et al.
Published: (2024)

Laplace Sample Information: Data Informativeness Through a Bayesian Lens
by: Kaiser, Johannes, et al.
Published: (2025)

Weakly Supervised Object Detection in Chest X-Rays with Differentiable ROI Proposal Networks and Soft ROI Pooling
by: Müller, Philip, et al.
Published: (2024)

Kernel Normalized Convolutional Networks
by: Nasirigerdeh, Reza, et al.
Published: (2022)

Survival In-Context: Amortized Bayesian Survival Analysis via Prior-Fitted Networks
by: Seletkov, Dmitrii, et al.
Published: (2026)

Machine Unlearning for Medical Imaging
by: Nasirigerdeh, Reza, et al.
Published: (2024)

Beyond the Calibration Point: Mechanism Comparison in Differential Privacy
by: Kaissis, Georgios, et al.
Published: (2024)

Efficient numeracy in language models through single-token number embeddings
by: Kreitner, Linus, et al.
Published: (2025)

Improved Localized Machine Unlearning Through the Lens of Memorization
by: Torkzadehmahani, Reihaneh, et al.
Published: (2024)

Differentially Private Active Learning: Balancing Effective Data Selection and Privacy
by: Schwethelm, Kristian, et al.
Published: (2024)

Complex-valued Federated Learning with Differential Privacy and MRI Applications
by: Riess, Anneliese, et al.
Published: (2021)

On Arbitrary Predictions from Equally Valid Models
by: Lockfisch, Sarah, et al.
Published: (2025)

Unintended Memorization of Sensitive Information in Fine-Tuned Language Models
by: Szep, Marton, et al.
Published: (2026)

Private, fair and accurate: Training large-scale, privacy-preserving AI models in medical imaging
by: Arasteh, Soroosh Tayebi, et al.
Published: (2023)

Training Gradient Boosted Decision Trees on Tabular Data Containing Label Noise for Classification Tasks
by: Eisenbürger, Anita, et al.
Published: (2024)

Weighting What Matters: Boosting Sample Efficiency in Medical Report Generation via Token Reweighting
by: Weers, Alexander, et al.
Published: (2026)

TOFU: Towards Obfuscated Federated Updates by Encoding Weight Updates into Gradients from Proxy Data
by: Garg, Isha, et al.
Published: (2022)

Proxy-RLHF: Decoupling Generation and Alignment in Large Language Model with Proxy
by: Zhu, Yu, et al.
Published: (2024)

Your Privacy Depends on Others: Collusion Vulnerabilities in Individual Differential Privacy
by: Kaiser, Johannes, et al.
Published: (2026)

Task Weighting through Gradient Projection for Multitask Learning
by: Bohn, Christian, et al.
Published: (2024)

Path-Weighted Integrated Gradients for Interpretable Dementia Classification
by: Kamalov, Firuz, et al.
Published: (2025)

Counterfactual Influence as a Distributional Quantity
by: Meeus, Matthieu, et al.
Published: (2025)

A Training-Time Diagnostic for Generalization via the Log-Alignment Ratio
by: Shehper, Ali, et al.
Published: (2026)

Enhancing the Utility of Privacy-Preserving Cancer Classification using Synthetic Data
by: Osuala, Richard, et al.
Published: (2024)

Multimodal Graph-based Classification of Esophageal Motility Disorders
by: Geiger, Alexander, et al.
Published: (2026)

Bilinear Coordinate Alignment for Training-Free Task-Vector Transfer
by: Son, Jungyong, et al.
Published: (2026)

Step-resolved data attribution for looped transformers
by: Kaissis, Georgios, et al.
Published: (2026)

Sensitivity, Specificity, and Consistency: A Tripartite Evaluation of Privacy Filters for Synthetic Data Generation
by: Koeken, Adil, et al.
Published: (2025)

Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data
by: Schoepf, Stefan, et al.
Published: (2025)

Robust-Multi-Task Gradient Boosting
by: Emami, Seyedsaman, et al.
Published: (2025)

Improving Performance in Classification Tasks with LCEN and the Weighted Focal Differentiable MCC Loss
by: Seber, Pedro, et al.
Published: (2026)

Influence-Preserving Proxies for Gradient-Based Data Selection in LLM Fine-tuning
by: Chen, Sirui, et al.
Published: (2026)

Is your algorithm unlearning or untraining?
by: Triantafillou, Eleni, et al.
Published: (2026)

Forward Direct Feedback Alignment for Online Gradient Estimates of Spiking Neural Networks
by: Bacho, Florian, et al.
Published: (2024)

Sobolev Training of End-to-End Optimization Proxies
by: Rosemberg, Andrew W., et al.
Published: (2025)

Learning Multi-Indicator Weights for Data Selection: A Joint Task-Model Adaptation Framework with Efficient Proxies
by: Song, Jingze, et al.
Published: (2026)

Pretrain-Test Task Alignment Governs Generalization in In-Context Learning
by: Letey, Mary I., et al.
Published: (2025)

Robust Weight Imprinting: Insights from Neural Collapse and Proxy-Based Aggregation
by: Westerhoff, Justus, et al.
Published: (2025)