Searcharxiv⌕ Search

arXiv subjects

Ling Liu

Publications and source records attributed to Ling Liu.

At least 145 records · Page 8Linked to original sources
arXiv2020

A Framework for Evaluating Gradient Leakage Attacks in Federated Learning

Federated learning (FL) is an emerging distributed machine learning framework for collaborative model training with a network of clients (edge devices). FL offers default client privacy by allowing clients to keep their sensitive data on local devices and to only share local training parameter updates with the federated server. However, recent studies have shown that even sharing local parameter updates from a client to the federated server may be susceptible to gradient leakage attacks and intrude the client privacy regarding its training data. In this paper, we present a principled framework for evaluating and comparing different forms of client privacy leakage attacks. We first provide formal and experimental analysis to show how adversaries can reconstruct the private local training data by simply analyzing the shared parameter update from local training (e.g., local gradient or weight update vector). We then analyze how different hyperparameter configurations in federated learning and different settings of the attack algorithm may impact on both attack effectiveness and attack cost. Our framework also measures, evaluates, and analyzes the effectiveness of client privacy leakage attacks under different gradient compression ratios when using communication efficient FL protocols. Our experiments also include some preliminary mitigation strategies to highlight the importance of providing a systematic attack evaluation framework towards an in-depth understanding of the various forms of client privacy leakage threats in federated learning and developing theoretical foundations for attack mitigation.

cs.LG↗
arXiv2020

TOG: Targeted Adversarial Objectness Gradient Attacks on Real-time Object Detection Systems

The rapid growth of real-time huge data capturing has pushed the deep learning and data analytic computing to the edge systems. Real-time object recognition on the edge is one of the representative deep neural network (DNN) powered edge systems for real-world mission-critical applications, such as autonomous driving and augmented reality. While DNN powered object detection edge systems celebrate many life-enriching opportunities, they also open doors for misuse and abuse. This paper presents three Targeted adversarial Objectness Gradient attacks, coined as TOG, which can cause the state-of-the-art deep object detection networks to suffer from object-vanishing, object-fabrication, and object-mislabeling attacks. We also present a universal objectness gradient attack to use adversarial transferability for black-box attacks, which is effective on any inputs with negligible attack time cost, low human perceptibility, and particularly detrimental to object detection edge systems. We report our experimental measurements using two benchmark datasets (PASCAL VOC and MS COCO) on two state-of-the-art detection algorithms (YOLO and SSD). The results demonstrate serious adversarial vulnerabilities and the compelling need for developing robust object detection systems.

cs.LG↗
arXiv2019

Differentially Private Model Publishing for Deep Learning

Deep learning techniques based on neural networks have shown significant success in a wide range of AI tasks. Large-scale training datasets are one of the critical factors for their success. However, when the training datasets are crowdsourced from individuals and contain sensitive information, the model parameters may encode private information and bear the risks of privacy leakage. The recent growing trend of the sharing and publishing of pre-trained models further aggravates such privacy risks. To tackle this problem, we propose a differentially private approach for training neural networks. Our approach includes several new techniques for optimizing both privacy loss and model accuracy. We employ a generalization of differential privacy called concentrated differential privacy(CDP), with both a formal and refined privacy loss analysis on two different data batching methods. We implement a dynamic privacy budget allocator over the course of training to improve model accuracy. Extensive experiments demonstrate that our approach effectively improves privacy loss accounting, training efficiency and model quality under a given privacy budget.

cs.CR↗
arXiv2019

Enhancing the understanding of hydrogen evolution and oxidation reaction on Pt(111) through ab initio simulations on electrode/electrolyte kinetics

The hydrogen oxidation reaction (HOR) and hydrogen evolution reaction (HER) play an important role in hydrogen based energy conversion. Recently, the frustrating performance in alkaline media raised debates on the relevant mechanism, especially on the role of surface hydroxyl (OH*). We present a full pH range electrode/electrolyte kinetics simulation for HER/HOR on Pt(111), with the potential-related rate constants been calculated with density functional theory methods. The polarization curves agree well with the experimental observations. The stability of OH* is found to be unlikely an effective activity descriptor since it is irrelevant to the onset potential of HOR/HER. Degree of rate control analyses reveal that the alkaline current is controlled jointly by Tafel and Volmer steps, while the acidic current solely by Tafel step, which explains the observed pH-dependent kinetics. Therefore, it is also possible to reduce the overpotential of alkaline HER/HOR by accelerating the Tafel step besides tuning the hydrogen binding energy.

cond-mat.mtrl-sci↗
arXiv2019

Effects of Differential Privacy and Data Skewness on Membership Inference Vulnerability

Membership inference attacks seek to infer the membership of individual training instances of a privately trained model. This paper presents a membership privacy analysis and evaluation system, called MPLens, with three unique contributions. First, through MPLens, we demonstrate how membership inference attack methods can be leveraged in adversarial machine learning. Second, through MPLens, we highlight how the vulnerability of pre-trained models under membership inference attack is not uniform across all classes, particularly when the training data itself is skewed. We show that risk from membership inference attacks is routinely increased when models use skewed training data. Finally, we investigate the effectiveness of differential privacy as a mitigation technique against membership inference attacks. We discuss the trade-offs of implementing such a mitigation strategy with respect to the model complexity, the learning task complexity, the dataset complexity and the privacy parameter settings. Our empirical results reveal that (1) minority groups within skewed datasets display increased risk for membership inference and (2) differential privacy presents many challenging trade-offs as a mitigation technique to membership inference risk.

cs.CR↗
arXiv2019

Demystifying Learning Rate Policies for High Accuracy Training of Deep Neural Networks

Learning Rate (LR) is an important hyper-parameter to tune for effective training of deep neural networks (DNNs). Even for the baseline of a constant learning rate, it is non-trivial to choose a good constant value for training a DNN. Dynamic learning rates involve multi-step tuning of LR values at various stages of the training process and offer high accuracy and fast convergence. However, they are much harder to tune. In this paper, we present a comprehensive study of 13 learning rate functions and their associated LR policies by examining their range parameters, step parameters, and value update parameters. We propose a set of metrics for evaluating and selecting LR policies, including the classification confidence, variance, cost, and robustness, and implement them in LRBench, an LR benchmarking system. LRBench can assist end-users and DNN developers to select good LR policies and avoid bad LR policies for training their DNNs. We tested LRBench on Caffe, an open source deep learning framework, to showcase the tuning optimization of LR policies. Evaluated through extensive experiments, we attempt to demystify the tuning of LR policies by identifying good LR policies with effective LR value ranges and step sizes for LR update schedules.

cs.LG↗
arXiv2019

Denoising and Verification Cross-Layer Ensemble Against Black-box Adversarial Attacks

Deep neural networks (DNNs) have demonstrated impressive performance on many challenging machine learning tasks. However, DNNs are vulnerable to adversarial inputs generated by adding maliciously crafted perturbations to the benign inputs. As a growing number of attacks have been reported to generate adversarial inputs of varying sophistication, the defense-attack arms race has been accelerated. In this paper, we present MODEF, a cross-layer model diversity ensemble framework. MODEF intelligently combines unsupervised model denoising ensemble with supervised model verification ensemble by quantifying model diversity, aiming to boost the robustness of the target model against adversarial examples. Evaluated using eleven representative attacks on popular benchmark datasets, we show that MODEF achieves remarkable defense success rates, compared with existing defense methods, and provides a superior capability of repairing adversarial inputs and making correct predictions with high accuracy in the presence of black-box attacks.

cs.LG↗
arXiv2019

Secure and Utility-Aware Data Collection with Condensed Local Differential Privacy

Local Differential Privacy (LDP) is popularly used in practice for privacy-preserving data collection. Although existing LDP protocols offer high utility for large user populations (100,000 or more users), they perform poorly in scenarios with small user populations (such as those in the cybersecurity domain) and lack perturbation mechanisms that are effective for both ordinal and non-ordinal item sequences while protecting sequence length and content simultaneously. In this paper, we address the small user population problem by introducing the concept of Condensed Local Differential Privacy (CLDP) as a specialization of LDP, and develop a suite of CLDP protocols that offer desirable statistical utility while preserving privacy. Our protocols support different types of client data, ranging from ordinal data types in finite metric spaces (numeric malware infection statistics), to non-ordinal items (OS versions, transaction categories), and to sequences of ordinal and non-ordinal items. Extensive experiments are conducted on multiple datasets, including datasets that are an order of magnitude smaller than those used in existing approaches, which show that proposed CLDP protocols yield high utility. Furthermore, case studies with Symantec datasets demonstrate that our protocols accurately support key cybersecurity-focused tasks of detecting ransomware outbreaks, identifying targeted and vulnerable OSs, and inspecting suspicious activities on infected machines.

cs.CR↗
arXiv2019

On adaptivity of wavelet thresholding estimators with negatively super-additive dependent noise

This paper considers the nonparametric regression model with negatively super-additive dependent (NSD) noise and investigates the convergence rates of thresholding estimators. It is shown that the term-by-term thresholding estimator achieves nearly optimal and the block thresholding estimator attains optimal (or nearly optimal) convergence rates over Besov spaces. Additionally, some numerical simulations are implemented to substantiate the validity and adaptivity of the thresholding estimators with the presence of NSD noise.

math.ST↗
arXiv2019

Cross-Layer Strategic Ensemble Defense Against Adversarial Examples

Deep neural network (DNN) has demonstrated its success in multiple domains. However, DNN models are inherently vulnerable to adversarial examples, which are generated by adding adversarial perturbations to benign inputs to fool the DNN model to misclassify. In this paper, we present a cross-layer strategic ensemble framework and a suite of robust defense algorithms, which are attack-independent, and capable of auto-repairing and auto-verifying the target model being attacked. Our strategic ensemble approach makes three original contributions. First, we employ input-transformation diversity to design the input-layer strategic transformation ensemble algorithms. Second, we utilize model-disagreement diversity to develop the output-layer strategic model ensemble algorithms. Finally, we create an input-output cross-layer strategic ensemble defense that strengthens the defensibility by combining diverse input transformation based model ensembles with diverse output verification model ensembles. Evaluated over 10 attacks on ImageNet dataset, we show that our strategic ensemble defense algorithms can achieve high defense success rates and are more robust with high attack prevention success rates and low benign false negative rates, compared to existing representative defense methods.

cs.LG↗
arXiv2019

Decentralized Trust Management: Risk Analysis and Trust Aggregation

Decentralized trust management is used as a referral benchmark for assisting decision making by human or intelligence machines in open collaborative systems. During any given period of time, each participant may only interact with a few of other participants. Simply relying on direct trust may frequently resort to random team formation. Thus, trust aggregation becomes critical. It can leverage decentralized trust management to learn about indirect trust of every participant based on past transaction experiences. This paper presents alternative designs of decentralized trust management and their efficiency and robustness from three perspectives. First, we study the risk factors and adverse effects of six common threat models. Second, we review the representative trust aggregation models and trust metrics. Third, we present an in-depth analysis and comparison of these reference trust aggregation methods with respect to effectiveness and robustness. We show our comparative study results through formal analysis and experimental evaluation. This comprehensive study advances the understanding of adverse effects of present and future threats and the robustness of different trust metrics. It may also serve as a guideline for research and development of next generation trust aggregation algorithms and services in the anticipation of risk factors and mischievous threats.

cs.SI↗
arXiv2019

Deep Neural Network Ensembles against Deception: Ensemble Diversity, Accuracy and Robustness

Ensemble learning is a methodology that integrates multiple DNN learners for improving prediction performance of individual learners. Diversity is greater when the errors of the ensemble prediction is more uniformly distributed. Greater diversity is highly correlated with the increase in ensemble accuracy. Another attractive property of diversity optimized ensemble learning is its robustness against deception: an adversarial perturbation attack can mislead one DNN model to misclassify but may not fool other ensemble DNN members consistently. In this paper we first give an overview of the concept of ensemble diversity and examine the three types of ensemble diversity in the context of DNN classifiers. We then describe a set of ensemble diversity measures, a suite of algorithms for creating diversity ensembles and for performing ensemble consensus (voted or learned) for generating high accuracy ensemble output by strategically combining outputs of individual members. This paper concludes with a discussion on a set of open issues in quantifying ensemble diversity for robust deep learning.

cs.LG↗
arXiv2019

Delivering Scientific Influence Analysis as a Service on Research Grants Repository

Research grants have played an important role in seeding and promoting fundamental research projects worldwide. There is a growing demand for developing and delivering scientific influence analysis as a service on research grant repositories. Such analysis can provide insight on how research grants help foster new research collaborations, encourage cross-organizational collaborations, influence new research trends, and identify technical leadership. This paper presents the design and development of a grants-based scientific influence analysis service, coined as GImpact. It takes a graph-theoretic approach to design and develop large scale scientific influence analysis over a large research-grant repository with three original contributions. First, we mine the grant database to identify and extract important features for grants influence analysis and represent such features using graph theoretic models. For example, we extract an institution graph and multiple associated aspect-based collaboration graphs, including a discipline graph and a keyword graph. Second, we introduce self-influence and co-influence algorithms to compute two types of collaboration relationship scores based on the number of grants and the types of grants for institutions. We compute the self-influence scores to reflect the grant based research collaborations among institutions and compute multiple co-influence scores to model the various types of cross-institution collaboration relationships in terms of disciplines and subject areas. Third, we compute the overall scientific influence score for every pair of institutions by introducing a weighted sum of the self-influence score and the multiple co-influence scores and conduct an influence-based clustering analysis. We evaluate GImpact using a real grant database, consisting of 2512 institutions and their grants received over a period of 14 years...

cs.SI↗
arXiv2019

A Comparative Measurement Study of Deep Learning as a Service Framework

Big data powered Deep Learning (DL) and its applications have blossomed in recent years, fueled by three technological trends: a large amount of digitized data openly accessible, a growing number of DL software frameworks in open source and commercial markets, and a selection of affordable parallel computing hardware devices. However, no single DL framework, to date, dominates in terms of performance and accuracy even for baseline classification tasks on standard datasets, making the selection of a DL framework an overwhelming task. This paper takes a holistic approach to conduct empirical comparison and analysis of four representative DL frameworks with three unique contributions. First, given a selection of CPU-GPU configurations, we show that for a specific DL framework, different configurations of its hyper-parameters may have a significant impact on both performance and accuracy of DL applications. Second, to the best of our knowledge, this study is the first to identify the opportunities for improving the training time performance and the accuracy of DL frameworks by configuring parallel computing libraries and tuning individual and multiple hyper-parameters. Third, we also conduct a comparative measurement study on the resource consumption patterns of four DL frameworks and their performance and accuracy implications, including CPU and memory usage, and their correlations to varying settings of hyper-parameters under different configuration combinations of hardware, parallel computing libraries. We argue that this measurement study provides in-depth empirical comparison and analysis of four representative DL frameworks, and offers practical guidance for service providers to deploying and delivering DL as a Service (DLaaS) and for application developers and DLaaS consumers to select the right DL frameworks for the right DL workloads.

cs.PF↗
arXiv2019

Security and Privacy on Blockchain

Blockchain offers an innovative approach to storing information, executing transactions, performing functions, and establishing trust in an open environment. Many consider blockchain as a technology breakthrough for cryptography and cybersecurity, with use cases ranging from globally deployed cryptocurrency systems like Bitcoin, to smart contracts, smart grids over the Internet of Things, and so forth. Although blockchain has received growing interests in both academia and industry in the recent years, the security and privacy of blockchains continue to be at the center of the debate when deploying blockchain in different applications. This paper presents a comprehensive overview of the security and privacy of blockchain. To facilitate the discussion, we first introduce the notion of blockchains and its utility in the context of Bitcoin like online transactions. Then we describe the basic security properties that are supported as the essential requirements and building blocks for Bitcoin like cryptocurrency systems, followed by presenting the additional security and privacy properties that are desired in many blockchain applications. Finally, we review the security and privacy techniques for achieving these security properties in blockchain-based systems, including representative consensus algorithms, hash chained storage, mixing protocols, anonymous signatures, non-interactive zero-knowledge proof, and so forth. We conjecture that this survey can help readers to gain an in-depth understanding of the security and privacy of blockchain with respect to concept, attributes, techniques and systems.

cs.CR↗
arXiv2019

On the Polarization of Rényi Entropy

Existing polarization theories have mostly been concerned with Shannon's information measures, such as Shannon entropy and mutual information, and some related measures such as the Bhattacharyya parameter. In this work, we extend polarization theories to a more general information measure, namely, the Rényi entropy. Our study shows that under conditional Rényi entropies of different orders, the same synthetic sub-channel may exhibit opposite extremal states. This result reveals more insights into the polarization phenomenon on the micro scale (probability pairs) rather than on the average scale.

cs.IT↗
arXiv2019

Utility-aware and privacy-preserving mobile query services

Location-based queries enable fundamental services for mobile road network travelers. While the benefits of location-based services (LBS) are numerous, exposure of mobile travelers' location information to untrusted LBS providers may lead to privacy breaches. In this paper, we propose StarCloak, a utility-aware and attack-resilient approach to building a privacy-preserving query system for mobile users traveling on road networks. StarCloak has several desirable properties. First, StarCloak supports user-defined k-user anonymity and l-segment indistinguishability, along with user-specified spatial and temporal utility constraints, for utility-aware and personalized location privacy. Second, unlike conventional solutions which are indifferent to underlying road network structure, StarCloak uses the concept of stars and proposes cloaking graphs for effective location cloaking on road networks. Third, StarCloak achieves strong attack-resilience against replay and query injection-based attacks through randomized star selection and pruning. Finally, to enable scalable query processing with high throughput, StarCloak makes cost-aware star selection decisions by considering query evaluation and network communication costs. We evaluate StarCloak on two real-world road network datasets under various privacy and utility constraints. Results show that StarCloak achieves improved query success rate and throughput, reduced anonymization time and network usage, and higher attack-resilience in comparison to XStar, its most relevant competitor.

cs.CR↗