Latest Articles for Data Privacy

Examining how fine-tuning increases the risk of revealing sensitive training data.

A look into reconstruction attacks and their impact on data privacy in machine learning.

A method for collaborative analysis without sharing sensitive patient data.

A new system enables faster CNN training on devices with limited memory.

Examining the challenges and solutions in Collaborative Machine Learning for better privacy and safety.

A new method to improve machine learning models affected by poor data.

A novel cache attack exploits replacement policies to leak sensitive information.

This article examines machine unlearning in large language models.

CoDream enables organizations to collaborate securely without sharing sensitive data.

Discover how ESFL improves machine learning efficiency while protecting privacy.

New methods to safeguard sensitive data against unauthorized access in machine learning.

Addressing privacy concerns in machine learning with effective techniques.

Explore how synthetic datasets enhance machine learning performance and model selection.

Addressing the challenge of privacy in data-driven decision-making for healthcare.

Exploring how larger batch sizes improve differential privacy in machine learning.

FedReview improves federated learning by rejecting harmful model updates.

Examining the challenges of differential privacy in online learning systems.

Exploring the privacy and security risks linked to Large Language Models.

FedUV improves model performance in federated learning on non-IID data.

Exploring local differential privacy methods for secure graph analysis.

AerisAI enhances AI collaboration while protecting data privacy through decentralized methods.

Exploring differential privacy methods in reinforcement learning to protect sensitive data.

New methods secure data in AI while ensuring effective computations.

This article presents a method for clients with diverse objectives in federated bandit learning.

Discussing privacy and fairness in machine learning through differential privacy and worst-group risk.

New algorithms enhance privacy and accuracy in sparse data scenarios.

A new method combines federated learning and secure computation to protect gaze data privacy.

BasedAI uses encryption to ensure privacy while enhancing language model performance.

A look at how data analysis can maintain individual privacy.

A method to remove unwanted skills from language models while keeping essential functions intact.

A novel method enhances energy load predictions while ensuring data privacy.

Asyn2F improves asynchronous federated learning for better model training and data privacy.

A new approach improves machine learning accuracy while ensuring data privacy.

A new approach to image representation with differential privacy through captioning.

A new method enhances federated learning efficiency using client update strategies.

Examining federated unlearning and its challenges in machine learning privacy.

Research shows how LLMs can expose training data, raising privacy concerns.

This article discusses privacy solutions for Max Cover and Set Cover problems.

A look into the risks of data poisoning in federated learning systems.

A new framework merges large and small models to prioritize user data protection.