arXiv subjects

Prasant Mohapatra

Publications and source records attributed to Prasant Mohapatra.

48 records · Page 3Linked to original sources

arXiv2019

Online and Bandit Algorithms for Nonstationary Stochastic Saddle-Point Optimization

Saddle-point optimization problems are an important class of optimization problems with applications to game theory, multi-agent reinforcement learning and machine learning. A majority of the rich literature available for saddle-point optimization has focused on the offline setting. In this paper, we study nonstationary versions of stochastic, smooth, strongly-convex and strongly-concave saddle-point optimization problem, in both online (or first-order) and multi-point bandit (or zeroth-order) settings. We first propose natural notions of regret for such nonstationary saddle-point optimization problems. We then analyze extragradient and Frank-Wolfe algorithms, for the unconstrained and constrained settings respectively, for the above class of nonstationary saddle-point optimization problems. We establish sub-linear regret bounds on the proposed notions of regret in both the online and bandit setting.

math.OC↗

arXiv2019

Suspicion-Free Adversarial Attacks on Clustering Algorithms

Clustering algorithms are used in a large number of applications and play an important role in modern machine learning-- yet, adversarial attacks on clustering algorithms seem to be broadly overlooked unlike supervised learning. In this paper, we seek to bridge this gap by proposing a black-box adversarial attack for clustering models for linearly separable clusters. Our attack works by perturbing a single sample close to the decision boundary, which leads to the misclustering of multiple unperturbed samples, named spill-over adversarial samples. We theoretically show the existence of such adversarial samples for the K-Means clustering. Our attack is especially strong as (1) we ensure the perturbed sample is not an outlier, hence not detectable, and (2) the exact metric used for clustering is not known to the attacker. We theoretically justify that the attack can indeed be successful without the knowledge of the true metric. We conclude by providing empirical results on a number of datasets, and clustering algorithms. To the best of our knowledge, this is the first work that generates spill-over adversarial samples without the knowledge of the true metric ensuring that the perturbed sample is not an outlier, and theoretically proves the above.

cs.LG↗

arXiv2019

Multi-Point Bandit Algorithms for Nonstationary Online Nonconvex Optimization

Bandit algorithms have been predominantly analyzed in the convex setting with function-value based stationary regret as the performance measure. In this paper, motivated by online reinforcement learning problems, we propose and analyze bandit algorithms for both general and structured nonconvex problems with nonstationary (or dynamic) regret as the performance measure, in both stochastic and non-stochastic settings. First, for general nonconvex functions, we consider nonstationary versions of first-order and second-order stationary solutions as a regret measure, motivated by similar performance measures for offline nonconvex optimization. In the case of second-order stationary solution based regret, we propose and analyze online and bandit versions of the cubic regularized Newton's method. The bandit version is based on estimating the Hessian matrices in the bandit setting, based on second-order Gaussian Stein's identity. Our nonstationary regret bounds in terms of second-order stationary solutions have interesting consequences for avoiding saddle points in the bandit setting. Next, for weakly quasi convex functions and monotone weakly submodular functions we consider nonstationary regret measures in terms of function-values; such structured classes of nonconvex functions enable one to consider regret measure defined in terms of function values, similar to convex functions. For this case of function-value, and first-order stationary solution based regret measures, we provide regret bounds in both the low- and high-dimensional settings, for some scenarios.

stat.ML↗

arXiv2019

Strong Black-box Adversarial Attacks on Unsupervised Machine Learning Models

Machine Learning (ML) and Deep Learning (DL) models have achieved state-of-the-art performance on multiple learning tasks, from vision to natural language modelling. With the growing adoption of ML and DL to many areas of computer science, recent research has also started focusing on the security properties of these models. There has been a lot of work undertaken to understand if (deep) neural network architectures are resilient to black-box adversarial attacks which craft perturbed input samples that fool the classifier without knowing the architecture used. Recent work has also focused on the transferability of adversarial attacks and found that adversarial attacks are generally easily transferable between models, datasets, and techniques. However, such attacks and their analysis have not been covered from the perspective of unsupervised machine learning algorithms. In this paper, we seek to bridge this gap through multiple contributions. We first provide a strong (iterative) black-box adversarial attack that can craft adversarial samples which will be incorrectly clustered irrespective of the choice of clustering algorithm. We choose 4 prominent clustering algorithms, and a real-world dataset to show the working of the proposed adversarial algorithm. Using these clustering algorithms we also carry out a simple study of cross-technique adversarial attack transferability.

cs.LG↗

arXiv2019

Wireless Access to Ultimate Virtual Reality 360-Degree Video At Home

Virtual reality 360-degree videos will become the first prosperous online VR application. VR 360 videos are data-hungry and latency-sensitive that pose unique challenges to the networking infrastructure. In this paper, we focus on the ultimate VR 360 that satisfies human eye fidelity. The ultimate VR 360 requires downlink 1.5 Gbps for viewing and uplink 6.6 Gbps for live broadcasting, with round-trip time of less than 8.3 ms. On the other hand, wireless access to VR 360 services is preferred over wire-line transmission because of the better user experience and the safety concern (e.g., tripping hazard). We explore in this paper whether the most advanced wireless technologies from both cellular communications and WiFi communications support the ultimate VR 360. Specifically, we consider 5G in cellular communications, IEEE 802.11ac (operating in 5GHz) and IEEE 802.11ad (operating in 60GHz) in WiFi communications. According to their performance specified in their standards and/or empirical measurements, we have the following findings: (1) Only 5G has the potential to support both the the ultimate VR 360 viewing and live broadcasting. However, it is difficult for 5G to support multiple users of the ultimate VR live broadcasting at home; (2) IEEE 802.11ac supports the ultimate VR 360 viewing but fails to support the ultimate VR 360 live broadcasting because it does not meet the data rate requirement of the ultimate VR 360 live broadcasting; (3) IEEE 802.11ad fails to support the ultimate VR 360, because its current implementation incurs very high latency. Our preliminary results indicate that more advanced wireless technologies are needed to fully support multiple ultimate VR 360 users at home.

cs.HC↗

arXiv2018

Edge Cloud Offloading Algorithms: Issues, Methods, and Perspectives

Mobile devices supporting the "Internet of Things" (IoT), often have limited capabilities in computation, battery energy, and storage space, especially to support resource-intensive applications involving virtual reality (VR), augmented reality (AR), multimedia delivery and artificial intelligence (AI), which could require broad bandwidth, low response latency and large computational power. Edge cloud or edge computing is an emerging topic and technology that can tackle the deficiency of the currently centralized-only cloud computing model and move the computation and storage resource closer to the devices in support of the above-mentioned applications. To make this happen, efficient coordination mechanisms and "offloading" algorithms are needed to allow the mobile devices and the edge cloud to work together smoothly. In this survey paper, we investigate the key issues, methods, and various state-of-the-art efforts related to the offloading problem. We adopt a new characterizing model to study the whole process of offloading from mobile devices to the edge cloud. Through comprehensive discussions, we aim to draw an overall "big picture" on the existing efforts and research directions. Our study also indicates that the offloading algorithms in edge cloud have demonstrated profound potentials for future technology and application development.

cs.NI↗

arXiv2017

LeakSemantic: Identifying Abnormal Sensitive Network Transmissions in Mobile Applications

Mobile applications (apps) often transmit sensitive data through network with various intentions. Some transmissions are needed to fulfill the app's functionalities. However, transmissions with malicious receivers may lead to privacy leakage and tend to behave stealthily to evade detection. The problem is twofold: how does one unveil sensitive transmissions in mobile apps, and given a sensitive transmission, how does one determine if it is legitimate? In this paper, we propose LeakSemantic, a framework that can automatically locate abnormal sensitive network transmissions from mobile apps. LeakSemantic consists of a hybrid program analysis component and a machine learning component. Our program analysis component combines static analysis and dynamic analysis to precisely identify sensitive transmissions. Compared to existing taint analysis approaches, LeakSemantic achieves better accuracy with fewer false positives and is able to collect runtime data such as network traffic for each transmission. Based on features derived from the runtime data, machine learning classifiers are built to further differentiate between the legal and illegal disclosures. Experiments show that LeakSemantic achieves 91% accuracy on 2279 sensitive connections from 1404 apps.

cs.CR↗

arXiv2016

When to Reset Your Keys: Optimal Timing of Security Updates via Learning

Cybersecurity is increasingly threatened by advanced and persistent attacks. As these attacks are often designed to disable a system (or a critical resource, e.g., a user account) repeatedly, it is crucial for the defender to keep updating its security measures to strike a balance between the risk of being compromised and the cost of security updates. Moreover, these decisions often need to be made with limited and delayed feedback due to the stealthy nature of advanced attacks. In addition to targeted attacks, such an optimal timing policy under incomplete information has broad applications in cybersecurity. Examples include key rotation, password change, application of patches, and virtual machine refreshing. However, rigorous studies of optimal timing are rare. Further, existing solutions typically rely on a pre-defined attack model that is known to the defender, which is often not the case in practice. In this work, we make an initial effort towards achieving optimal timing of security updates in the face of unknown stealthy attacks. We consider a variant of the influential FlipIt game model with asymmetric feedback and unknown attack time distribution, which provides a general model to consecutive security updates. The defender's problem is then modeled as a time associative bandit problem with dependent arms. We derive upper confidence bound based learning policies that achieve low regret compared with optimal periodic defense strategies that can only be derived when attack time distributions are known.

cs.LG↗

arXiv2016

Trust Exploitation and Attention Competition: A Game Theoretical Model

The proliferation of Social Network Sites (SNSs) has greatly reformed the way of information dissemination, but also provided a new venue for hosts with impure motivations to disseminate malicious information. Social trust is the basis for information dissemination in SNSs. Malicious hosts judiciously and dynamically make the balance between maintaining its social trust and selfishly maximizing its malicious gain over a long time-span. Studying the optimal response strategies for each malicious host could assist to design the best system maneuver so as to achieve the targeted level of overall malicious activities. In this paper, we propose an interaction-based social trust model, and formulate the maximization of long-term malicious gains of multiple competing hosts as a non-cooperative differential game. Through rigorous analysis, optimal response strategies are identified and the best system maneuver mechanism is presented. Extensive numerical studies further verify the analytical results.

cs.GT↗

arXiv2016

FlowIntent: Detecting Privacy Leakage from User Intention to Network Traffic Mapping

The exponential growth of mobile devices has raised concerns about sensitive data leakage. In this paper, we make the first attempt to identify suspicious location-related HTTP transmission flows from the user's perspective, by answering the question: Is the transmission user-intended? In contrast to previous network-level detection schemes that mainly rely on a given set of suspicious hostnames, our approach can better adapt to the fast growth of app market and the constantly evolving leakage patterns. On the other hand, compared to existing system-level detection schemes built upon program taint analysis, where all sensitive transmissions as treated as illegal, our approach better meets the user needs and is easier to deploy. In particular, our proof-of-concept implementation (FlowIntent) captures sensitive transmissions missed by TaintDroid, the state-of-the-art dynamic taint analysis system on Android platforms. Evaluation using 1002 location sharing instances collected from more than 20,000 apps shows that our approach achieves about 91% accuracy in detecting illegitimate location transmissions.

cs.CR↗

arXiv2013

Using Social Information for Flow Allocation in MANETs

Adhoc networks enable communication between distributed, mobile wireless nodes without any supporting infrastructure. In the absence of centralized control, such networks require node interaction, and are inherently based on cooperation between nodes. In this paper, we use social and behavioral trust of nodes to form a flow allocation optimization problem. We initialize trust using information gained from users' social relationships (from social networks) and update the trusts metric over time based on observed node behaviors. We conduct analysis of social trust using real data sets and used it as a parameter for performance evaluation of our frame work in ns-3. Based on our approach we obtain a significant improvement in both detection rate and packet delivery ratio using social trust information when compared to behavioral trust alone. Further, we observe that social trust is critical in the event of mobility and plays a crucial role in bootstrapping the computation of trust.

cs.NI↗

arXiv2010

Practical Rate and Route Adaptation with Efficient Link Quality Estimation for IEEE 802.11b/g Multi-Hop Networks

Accurate and fast packet delivery rate (PDR) estimation, used in evaluating wireless link quality, is a prerequisite to increase the performance of mobile, multi-hop and multi-rate wireless ad hoc networks. Unfortunately, contemporary PDR estimation methods, i.e. beacon-based packet counting in Estimated Transmission Time and Expected Transmission Count metrics, have unsatisfactory performance. Therefore, in this paper we propose a novel PDR estimation method based on SNR profiles. We classify all possible link quality estimation methods and compare them analytically against our design. Results show that it leads to a more efficient link quality estimation. Further investigations with the prototype implementation of our method in IEEE 802.11b/g testbeds reveal that the accuracy of PDR estimation in mobile scenarios can be improved up to 50% in comparison to generic packet-based PDR. Experiments with the same prototype on link and routing layers for different measurement scenarios show that it leads to a better rate adaptation and route selection in the form of end-to-end throughput increase compared to traditional packet counting methods.

cs.NI↗