A Metric Differential Privacy Mechanism for Sentence Embeddings

This paper considers the problem of secure data aggregation (mainly summation) in a distributed setting, while ensuring differential privacy of the result. We study secure multiparty addition protocols using well known security schemes: Shamir's secret sharing, perturbation-based, and various encryptions. We supplement our study with our new enhanced encryption scheme EFT, which is efficient and fault tolerant. Differential privacy of the final result is achieved by either distributed Laplace or Geometric mechanism (respectively DLPA or DGPA), while approximated differential privacy is achieved by diluted mechanisms. Distributed random noise is generated collectively by all participants, which draw random variables from one of several distributions: Gamma, Gauss, Geometric, or their diluted versions. We introduce a new distributed privacy mechanism with noise drawn from the Laplace distribution, which achieves smaller redundant noise with efficiency. We compare complexity and security characteristics of the protocols with different differential privacy mechanisms and security schemes. More importantly, we implemented all protocols and present an experimental comparison on their performance and scalability in a real distributed environment. Based on the evaluations, we identify our security scheme and Laplace DLPA as the most efficient for secure distributed data aggregation with privacy.

Differential privacy is often used in location privacy protection because of its strict reasoning and proof privacy guarantee. When users make continuous location query, it will cause noise superposition, which leads to the decline of query accuracy. At present, although differential privacy based on rule tree structure can reduce the query error, it will generate a lot of invalid zero nodes. The data structure is too large, and more improvement in the query accuracy can be further investigated. In this paper, we proposed a differential privacy location privacy protection method based on generative adversary network. Firstly, the definition of location data privacy protection under differential privacy mechanism is given, and then resume density aware network under differential privacy mechanism. Based on density aware network, the privacy protection problem of location data can be transformed into the distribution of fitting trajectory length. Finally, we use Markov chain to generate a new trajectory, and introduce the generative adversarial network to construct a set differential privacy protection method. Compared with other methods to improve the accuracy of differential privacy query, this method can effectively reduce the problem of query accuracy decline caused by noise superposition in continuous query, and can adapt to lbs location query service in different density environments.KeywordsGenerative adversarial networkLocation differencePrivacy protectionMarkov chainDensity aware network

Differential privacy mechanisms have been proposed to guarantee the privacy of individuals in various types of statistical information. When constructing a probabilistic mechanism to satisfy differential privacy, it is necessary to consider the impact of an arbitrary record on its statistics, i.e., sensitivity, but there are situations where sensitivity is difficult to derive. In this paper, we first summarize the situations in which it is difficult to derive sensitivity in general, and then propose a definition equivalent to the conventional definition of differential privacy to deal with them. This definition considers neighboring datasets as in the conventional definition. Therefore, known differential privacy mechanisms can be applied. Next, as an example of the difficulty in deriving sensitivity, we focus on the t-test, a basic tool in statistical analysis, and show that a concrete differential privacy mechanism can be constructed in practice. Our proposed definition can be treated in the same way as the conventional differential privacy definition, and can be applied to cases where it is difficult to derive sensitivity.

BlendCSE: Blend contrastive learnings for sentence embeddings with rich semantics and transferability

With growing applications of Machine Learning in daily lives Natural Language Processing (NLP) has emerged as a heavily researched area. Finding its applications in tasks ranging from simple Q/A chatbots to Fully fledged conversational AI, NLP models are vital. Word and Sentence embedding are one of the most common starting points of any NLP task. A word embedding represents a given word in a predefined vector-space while maintaining vector relations with similar or dis-similar entities. As such different pretrained embedding such as Word2Vec, GloVe, fasttext have been developed. These embedding generated on millions of words are however very large in terms of size. Having embedding with floating point precision also makes the downstream evaluation slow. In this paper we present a novel method to convert continuous embedding to its binary representation, thus reducing the overall size of the embedding while keeping the semantic and relational knowledge intact. This will facilitate an option of porting such big embedding onto devices where space is limited. We also present different approaches suitable for different downstream tasks based on the requirement of contextual and semantic information. Experiments have shown comparable result in downstream tasks with 7 to 15 times reduction in file size and about 5 % change in evaluation parameters.

The increasing complexity and dynamic nature of cyber threats in zero-trust architectures necessitates a more adaptive approach to differential privacy mechanisms. Current static privacy solutions fail to adequately address evolving threat landscapes, leading to potential vulnerabilities and reduced system efficiency. This research presents a novel adaptive differential privacy framework that dynamically adjusts privacy parameters based on real-time threat assessment within zero-trust environments. Our solution introduces an intelligent privacy budget optimization algorithm that continuously evaluates threat levels and automatically recalibrates privacy mechanisms to maintain optimal protection while minimizing performance overhead. Through extensive experimental evaluation using real-world datasets and simulated attack scenarios, we demonstrate that our adaptive approach achieves a 47% improvement in privacy preservation compared to static mechanisms, while maintaining system performance within acceptable thresholds. The framework successfully detects and responds to 94% of emerging threats within milliseconds, dynamically adjusting privacy parameters to counter identified risks. Our results show that the proposed solution effectively balances privacy protection, system performance, and threat responsiveness in zero-trust architectures. Additionally, we provide comprehensive implementation guidelines and identified key challenges for deploying adaptive differential privacy mechanisms in production environments. This research contributes to the advancement of privacy-preserving systems by introducing a practical, scalable solution for managing differential privacy in dynamic threat landscapes.

Artificial Intelligence (AI) has attracted a large amount of attention in recent years. However, several new problems, such as privacy violations, security issues, or effectiveness, have been emerging. Differential privacy has several attractive properties that make it quite valuable for AI, such as privacy preservation, security, randomization, composition, and stability. Therefore, this paper presents differential privacy mechanisms for multi-agent systems, reinforcement learning, and knowledge transfer based on those properties, which proves that current AI can benefit from differential privacy mechanisms. In addition, the previous usage of differential privacy mechanisms in private machine learning, distributed machine learning, and fairness in models is discussed, bringing several possible avenues to use differential privacy mechanisms in AI. The purpose of this paper is to deliver the initial idea of how to integrate AI with differential privacy mechanisms and to explore more possibilities to improve AIs performance.

Personal data privacy is a great concern by governments across the world as citizens generate huge amount of data continuously and industries using this for betterment of user centric services. There must be a reasonable balance between data privacy and utility of data. Differential privacy is a promise by data collector to the customer’s personal privacy. Centralised Differential Privacy (CDP) is performing output perturbation of user’s data by applying required privacy budget. This promises the inclusion or exclusion of individual’s data in data set not going to create significant change for a statistical query output and it offers -Differential privacy guarantee. CDP is holding a strong belief on trusted data collector and applying global sensitivity of the data. Local Differential Privacy (LDP) helps user to locally perturb his data and there by guaranteeing privacy even with untrusted data collector. Many differential privacy algorithms handles parameters like privacy budget, sensitivity and data utility in different ways and mostly trying to keep trade-off between privacy and utility of data. This paper evaluates differential privacy algorithms in regard to the privacy support it offers according to the sensitivity of the data. Generalized application of privacy budget is found ineffective in comparison to the sensitivity based usage of privacy budget.

Word and sentence embeddings are useful feature representations in natural language processing. However, intrinsic evaluation for embeddings lags far behind, and there has been no significant update since the past decade. Word and sentence similarity tasks have become the de facto evaluation method. It leads models to overfit to such evaluations, negatively impacting embedding models' development. This paper first points out the problems using semantic similarity as the gold standard for word and sentence embedding evaluations. Further, we propose a new intrinsic evaluation method called EvalRank, which shows a much stronger correlation with downstream tasks. Extensive experiments are conducted based on 60+ models and popular datasets to certify our judgments. Finally, the practical evaluation toolkit is released for future benchmarking purposes.

Gradient leakage attacks are considered one of the wickedest privacy threats in deep learning as attackers covertly spy gradient updates during iterative training without compromising model training quality, and yet secretly reconstruct sensitive training data using leaked gradients with high attack success rate. Although deep learning with differential privacy is a defacto standard for publishing deep learning models with differential privacy guarantee, we show that differentially private algorithms with fixed privacy parameters are vulnerable against gradient leakage attacks. This paper investigates alternative approaches to gradient leakage resilient deep learning with differential privacy (DP). <i>First</i>, we analyze existing implementation of deep learning with differential privacy, which use fixed noise variance to injects constant noise to the gradients in all layers using fixed privacy parameters. Despite the DP guarantee provided, the method suffers from low accuracy and is vulnerable to gradient leakage attacks. <i>Second</i>, we present a gradient leakage resilient deep learning approach with differential privacy guarantee by using dynamic privacy parameters. Unlike fixed-parameter strategies that result in constant noise variance, different dynamic parameter strategies present alternative techniques to introduce adaptive noise variance and adaptive noise injection which are closely aligned to the trend of gradient updates during differentially private model training. <i>Finally</i>, we describe four complementary metrics to evaluate and compare alternative approaches. Extensive experiments on six benchmark datasets show that differentially private deep learning with dynamic privacy parameters outperforms the deep learning using fixed DP parameters, and existing adaptive clipping approaches in all aspects: compelling accuracy performance, strong differential privacy guarantee, and high attack resilience.

New generation head-mounted displays, such as VR and AR glasses, are coming into the market with already integrated eye tracking and are expected to enable novel ways of human-computer interaction in numerous applications. However, since eye movement properties contain biometric information, privacy concerns have to be handled properly. Privacy-preservation techniques such as differential privacy mechanisms have recently been applied to eye movement data obtained from such displays. Standard differential privacy mechanisms; however, are vulnerable due to temporal correlations between the eye movement observations. In this work, we propose a novel transform-coding based differential privacy mechanism to further adapt it to the statistics of eye movement feature data and compare various low-complexity methods. We extend the Fourier perturbation algorithm, which is a differential privacy mechanism, and correct a scaling mistake in its proof. Furthermore, we illustrate significant reductions in sample correlations in addition to query sensitivities, which provide the best utility-privacy trade-off in the eye tracking literature. Our results provide significantly high privacy without any essential loss in classification accuracies while hiding personal identifiers.

Prompting techniques unlock valuable insights from pre-trained models, leading to a remarkable performance in Natural Language Processing (NLP). Inspired by NLP advancements, various graph prompt methods have emerged to transfer knowledge from pre-trained graph models to diverse downstream graph tasks. However, designing appropriate prompts for different downstream tasks remains a significant challenge due to substantial differences between pre-training graph tasks and downstream graph tasks. Moreover, existing graph prompt methods prevent transferability across different domains as they independently pre-train a model for each dataset. To address these challenges, we propose Graph Pretraining and Task-Specific Prompting (GTSP). To the best of our knowledge, GTSP is the first unified pre-training framework to handle cross-domain graph data and multiple types of graph tasks simultaneously. The key idea of our framework is to utilize similar label features between pre-training tasks and downstream tasks to seamlessly integrate the downstream task prompt into the pretrained model. We then propose a task-specific prompting architecture and design a label similarity matrix as prompt input. With GTSP, the pre-trained model can adapt to various types of downstream graph tasks, effectively avoiding the challenges of forcibly altering graph task forms. Experimental analysis conducted on benchmark datasets demonstrates the superiority of our method.

OPTDP: Towards optimal personalized trajectory differential privacy for trajectory data publishing

Representation Learning and Learning from Limited Labeled Data for Community Question Answering

There has been a recent wave of work assessing the fairness of machine learning models in general, and more specifically, on natural language processing (NLP) models built using machine learning techniques. While much work has highlighted biases embedded in state-of-the-art language models, and more recent efforts have focused on how to debias, research assessing the fairness and performance of biased/debiased models on downstream prediction tasks has been limited. Moreover, most prior work has emphasized bias along a single dimension such as gender or race. In this work, we benchmark multiple NLP models with regards to their fairness and predictive performance across a variety of NLP tasks. In particular, we assess intersectional bias - fairness across multiple demographic dimensions. The results show that while current debiasing strategies fare well in terms of the fairness-accuracy trade-off (generally preserving predictive power in debiased models), they are unable to effectively alleviate bias in downstream tasks. Furthermore, this bias is often amplified across demographic dimensions. We conclude with implications for future NLP debiasing research.