Search

Multi-Object Tracking (MOT) poses significant challenges in computer vision. Despite its wide application in robotics, autonomous driving, and smart manufacturing, there is limited literature addressing the specific challenges of running MOT on embedded devices. State-of-the-art MOT trackers designed for high-end GPUs often experience low processing rates (<11fps) when deployed on embedded devices. Existing MOT frameworks for embedded devices proposed strategies such as fusing the detector model with the feature embedding model to reduce inference latency or combining different trackers to improve tracking accuracy, but tend to compromise one for the other. This paper introduces HopTrack, a real-time multi-object tracking system tailored for embedded devices. Our system employs a novel discretized static and dynamic matching approach along with an innovative content-aware dynamic sampling technique to enhance tracking accuracy while meeting the real-time requirement. Compared with the best high-end GPU modified baseline Byte (Embed) and the best existing baseline on embedded devices MobileNet-JDE, HopTrack achieves a processing speed of up to 39.29 fps on NVIDIA AGX Xavier with a multi-object tracking accuracy (MOTA) of up to 63.12% on the MOT16 benchmark, outperforming both counterparts by 2.15% and 4.82%, respectively. Additionally, the accuracy improvement is coupled with the reduction in energy consumption (20.8%), power (5%), and memory usage (8%), which are crucial resources on embedded devices. HopTrack is also detector agnostic allowing the flexibility of plug-and-play.

Edge computing is deemed a promising technique to execute latency-sensitive applications by offloading computation-intensive tasks to edge servers. Extensive research has been conducted in the field of end-device to edge server task offloading for several goals, including latency minimization, energy optimization, and resource optimization. However, few of them consider our mobile computing devices (smartphones, tablets, and laptops) to be edge devices. In this paper, we propose a novel multi-tier edge computing framework, which we refer to as M-TEC, that aims to optimize latency, reduce the probability of failure, and optimize cost while accounting for the sporadic failure of personally owned devices and the changing network conditions. We conduct experiments with a real testbed and a real commercial CBRS 4G network, and the results indicate that M-TEC is capable of reducing the end-to-end latency of applications by at least 8\% compared to the best baseline under a variety of network conditions, while providing reliable performance at an affordable cost., Comment: 12 pages

Deep learning has shown incredible potential across a vast array of tasks and accompanying this growth has been an insatiable appetite for data. However, a large amount of data needed for enabling deep learning is stored on personal devices and recent concerns on privacy have further highlighted challenges for accessing such data. As a result, federated learning (FL) has emerged as an important privacy-preserving technology enabling collaborative training of machine learning models without the need to send the raw, potentially sensitive, data to a central server. However, the fundamental premise that sending model updates to a server is privacy-preserving only holds if the updates cannot be "reverse engineered" to infer information about the private training data. It has been shown under a wide variety of settings that this premise for privacy does {\em not} hold. In this survey paper, we provide a comprehensive literature review of the different privacy attacks and defense methods in FL. We identify the current limitations of these attacks and highlight the settings in which FL client privacy can be broken. We dissect some of the successful industry applications of FL and draw lessons for future successful adoption. We survey the emerging landscape of privacy regulation for FL. We conclude with future directions for taking FL toward the cherished goal of generating accurate models while preserving the privacy of the data from its participants., Comment: Submitted to ACM Computing Surveys

Agreement protocols are crucial in various emerging applications, spanning from distributed (blockchains) oracles to fault-tolerant cyber-physical systems. In scenarios where sensor/oracle nodes measure a common source, maintaining output within the convex range of correct inputs, known as convex validity, is imperative. Present asynchronous convex agreement protocols employ either randomization, incurring substantial computation overhead, or approximate agreement techniques, leading to high $\mathcal{\tilde{O}}(n^3)$ communication for an $n$-node system. This paper introduces Delphi, a deterministic protocol with $\mathcal{\tilde{O}}(n^2)$ communication and minimal computation overhead. Delphi assumes that honest inputs are bounded, except with negligible probability, and integrates agreement primitives from literature with a novel weighted averaging technique. Experimental results highlight Delphi's superior performance, showcasing a significantly lower latency compared to state-of-the-art protocols. Specifically, for an $n=160$-node system, Delphi achieves an 8x and 3x improvement in latency within CPS and AWS environments, respectively., Comment: 14 pages, 8 figures, Accepted to DSN 2024

Given the recent impact of Deep Reinforcement Learning in training agents to win complex games like StarCraft and DoTA(Defense Of The Ancients) - there has been a surge in research for exploiting learning based techniques for professional wargaming, battlefield simulation and modeling. Real time strategy games and simulators have become a valuable resource for operational planning and military research. However, recent work has shown that such learning based approaches are highly susceptible to adversarial perturbations. In this paper, we investigate the robustness of an agent trained for a Command and Control task in an environment that is controlled by an active adversary. The C2 agent is trained on custom StarCraft II maps using the state of the art RL algorithms - A3C and PPO. We empirically show that an agent trained using these algorithms is highly susceptible to noise injected by the adversary and investigate the effects these perturbations have on the performance of the trained agent. Our work highlights the urgent need to develop more robust training algorithms especially for critical arenas like the battlefield., Comment: Accepted to appear in the Journal Of Defense Modeling and Simulation (JDMS)

Federated learning is a decentralized learning paradigm introduced to preserve privacy of client data. Despite this, prior work has shown that an attacker at the server can still reconstruct the private training data using only the client updates. These attacks are known as data reconstruction attacks and fall into two major categories: gradient inversion (GI) and linear layer leakage attacks (LLL). However, despite demonstrating the effectiveness of these attacks in breaching privacy, prior work has not investigated the usefulness of the reconstructed data for downstream tasks. In this work, we explore data reconstruction attacks through the lens of training and improving models with leaked data. We demonstrate the effectiveness of both GI and LLL attacks in maliciously training models using the leaked data more accurately than a benign federated learning strategy. Counter-intuitively, this bump in training quality can occur despite limited reconstruction quality or a small total number of leaked images. Finally, we show the limitations of these attacks for downstream training, individually for GI attacks and for LLL attacks., Comment: Accepted to CVPR 2024

Many of our critical infrastructure systems and personal computing systems have a distributed computing systems structure. The incentives to attack them have been growing rapidly as has their attack surface due to increasing levels of connectedness. Therefore, we feel it is time to bring in rigorous reasoning to secure such systems. The distributed system security and the game theory technical communities can come together to effectively address this challenge. In this article, we lay out the foundations from each that we can build upon to achieve our goals. Next, we describe a set of research challenges for the community, organized into three categories -- analytical, systems, and integration challenges, each with "short term" time horizon (2-3 years) and "long term" (5-10 years) items. This article was conceived of through a community discussion at the 2022 NSF SaTC PI meeting., Comment: Accepted to appear at the IEEE Security & Privacy Magazine

While prior domain generalization (DG) benchmarks consider train-test dataset heterogeneity, we evaluate Federated DG which introduces federated learning (FL) specific challenges. Additionally, we explore domain-based heterogeneity in clients' local datasets - a realistic Federated DG scenario. Prior Federated DG evaluations are limited in terms of the number or heterogeneity of clients and dataset diversity. To address this gap, we propose an Federated DG benchmark methodology that enables control of the number and heterogeneity of clients and provides metrics for dataset difficulty. We then apply our methodology to evaluate 14 Federated DG methods, which include centralized DG methods adapted to the FL context, FL methods that handle client heterogeneity, and methods designed specifically for Federated DG. Our results suggest that despite some progress, there remain significant performance gaps in Federated DG particularly when evaluating with a large number of clients, high client heterogeneity, or more realistic datasets. Please check our extendable benchmark code here: https://github.com/inouye-lab/FedDG_Benchmark.

Modern Byzantine Fault-Tolerant State Machine Replication (BFT-SMR) solutions focus on reducing communication complexity, improving throughput, or lowering latency. This work explores the energy efficiency of BFT-SMR protocols. First, we propose a novel SMR protocol that optimizes for the steady state, i.e., when the leader is correct. This is done by reducing the number of required signatures per consensus unit and the communication complexity by order of the number of nodes n compared to the state-of-the-art BFT-SMR solutions. Concretely, we employ the idea that a quorum (collection) of signatures on a proposed value is avoidable during the failure-free runs. Second, we model and analyze the energy efficiency of protocols and argue why the steady-state needs to be optimized. Third, we present an application in the cyber-physical system (CPS) setting, where we consider a partially connected system by optionally leveraging wireless multicasts among neighbors. We analytically determine the parameter ranges for when our proposed protocol offers better energy efficiency than communicating with a baseline protocol utilizing an external trusted node. We present a hypergraph-based network model and generalize previous fault tolerance results to the model. Finally, we demonstrate our approach's practicality by analyzing our protocol's energy efficiency through experiments on a CPS test bed. In particular, we observe as high as 64% energy savings when compared to the state-of-the-art SMR solution for n=10 settings using BLE., Comment: Appearing in Middleware 2023

Secure aggregation promises a heightened level of privacy in federated learning, maintaining that a server only has access to a decrypted aggregate update. Within this setting, linear layer leakage methods are the only data reconstruction attacks able to scale and achieve a high leakage rate regardless of the number of clients or batch size. This is done through increasing the size of an injected fully-connected (FC) layer. However, this results in a resource overhead which grows larger with an increasing number of clients. We show that this resource overhead is caused by an incorrect perspective in all prior work that treats an attack on an aggregate update in the same way as an individual update with a larger batch size. Instead, by attacking the update from the perspective that aggregation is combining multiple individual updates, this allows the application of sparsity to alleviate resource overhead. We show that the use of sparsity can decrease the model size overhead by over 327$\times$ and the computation time by 3.34$\times$ compared to SOTA while maintaining equivalent total leakage rate, 77% even with $1000$ clients in aggregation., Comment: Accepted to CVPR 2023

Federated learning was introduced to enable machine learning over large decentralized datasets while promising privacy by eliminating the need for data sharing. Despite this, prior work has shown that shared gradients often contain private information and attackers can gain knowledge either through malicious modification of the architecture and parameters or by using optimization to approximate user data from the shared gradients. However, prior data reconstruction attacks have been limited in setting and scale, as most works target FedSGD and limit the attack to single-client gradients. Many of these attacks fail in the more practical setting of FedAVG or if updates are aggregated together using secure aggregation. Data reconstruction becomes significantly more difficult, resulting in limited attack scale and/or decreased reconstruction quality. When both FedAVG and secure aggregation are used, there is no current method that is able to attack multiple clients concurrently in a federated learning setting. In this work we introduce LOKI, an attack that overcomes previous limitations and also breaks the anonymity of aggregation as the leaked data is identifiable and directly tied back to the clients they come from. Our design sends clients customized convolutional parameters, and the weight gradients of data points between clients remain separate even through aggregation. With FedAVG and aggregation across 100 clients, prior work can leak less than 1% of images on MNIST, CIFAR-100, and Tiny ImageNet. Using only a single training round, LOKI is able to leak 76-86% of all data samples., Comment: To appear in the IEEE Symposium on Security & Privacy (S&P) 2024

Weather and soil conditions are particularly important when it comes to farming activities. Study of these factors and their role in nutrient and nitrate absorption rates can lead to useful insights with benefits for both the crop yield and the protection of the environment through the more controlled use of fertilizers and chemicals. There is a paucity of public data from rural, agricultural sensor networks. This is partly due to the unique challenges faced during the deployment and maintenance of IoT networks in rural agricultural areas. As part of a 5-year project called WHIN we have been deploying and collecting sensor data from production and experimental agricultural farms in and around Purdue University in Indiana. Here we release a dataset comprising soil sensor data from a representative sample of 3 nodes across 3 production farms, each for 5 months. We correlate this data with the weather data and draw some insights about the absorption of rain in the soil. We provide the dataset at: https://purduewhin.ecn.purdue.edu/dataset2021.

As we increase the number of personal computing devices that we carry (mobile devices, tablets, e-readers, and laptops) and these come equipped with increasing resources, there is a vast potential computation power that can be utilized from those devices. Edge computing promises to exploit these underlying computation resources closer to users to help run latency-sensitive applications such as augmented reality and video analytics. However, one key missing piece has been how to incorporate personally owned unmanaged devices into a usable edge computing system. The primary challenges arise due to the heterogeneity, lack of interference management, and unpredictable availability of such devices. In this paper we propose an orchestration framework IBDASH, which orchestrates application tasks on an edge system that comprises a mix of commercial and personal edge devices. IBDASH targets reducing both end-to-end latency of execution and probability of failure for applications that have dependency among tasks, captured by directed acyclic graphs (DAGs). IBDASH takes memory constraints of each edge device and network bandwidth into consideration. To assess the effectiveness of IBDASH, we run real application tasks on real edge devices with widely varying capabilities.We feed these measurements into a simulator that runs IBDASH at scale. Compared to three state-of-the-art edge orchestration schemes, LAVEA, Petrel, and LaTS, and two intuitive baselines, IBDASH reduces the end-to-end latency and probability of failure, by 14% and 41% on average respectively. The main takeaway from our work is that it is feasible to combine personal and commercial devices into a usable edge computing platform, one that delivers low latency and predictable and high availability.

Smart manufacturing systems are being deployed at a growing rate because of their ability to interpret a wide variety of sensed information and act on the knowledge gleaned from system observations. In many cases, the principal goal of the smart manufacturing system is to rapidly detect (or anticipate) failures to reduce operational cost and eliminate downtime. This often boils down to detecting anomalies within the sensor date acquired from the system. The smart manufacturing application domain poses certain salient technical challenges. In particular, there are often multiple types of sensors with varying capabilities and costs. The sensor data characteristics change with the operating point of the environment or machines, such as, the RPM of the motor. The anomaly detection process therefore has to be calibrated near an operating point. In this paper, we analyze four datasets from sensors deployed from manufacturing testbeds. We evaluate the performance of several traditional and ML-based forecasting models for predicting the time series of sensor data. Then, considering the sparse data from one kind of sensor, we perform transfer learning from a high data rate sensor to perform defect type classification. Taken together, we show that predictive failure classification can be achieved, thus paving the way for predictive maintenance., Comment: arXiv admin note: substantial text overlap with arXiv:2102.05814

Efficient and adaptive computer vision systems have been proposed to make computer vision tasks, such as image classification and object detection, optimized for embedded or mobile devices. These solutions, quite recent in their origin, focus on optimizing the model (a deep neural network, DNN) or the system by designing an adaptive system with approximation knobs. In spite of several recent efforts, we show that existing solutions suffer from two major drawbacks. First, the system does not consider energy consumption of the models while making a decision on which model to run. Second, the evaluation does not consider the practical scenario of contention on the device, due to other co-resident workloads. In this work, we propose an efficient and adaptive video object detection system, Virtuoso, which is jointly optimized for accuracy, energy efficiency, and latency. Underlying Virtuoso is a multi-branch execution kernel that is capable of running at different operating points in the accuracy-energy-latency axes, and a lightweight runtime scheduler to select the best fit execution branch to satisfy the user requirement. To fairly compare with Virtuoso, we benchmark 15 state-of-the-art or widely used protocols, including Faster R-CNN (FRCNN), YOLO v3, SSD, EfficientDet, SELSA, MEGA, REPP, FastAdapt, and our in-house adaptive variants of FRCNN+, YOLO+, SSD+, and EfficientDet+ (our variants have enhanced efficiency for mobiles). With this comprehensive benchmark, Virtuoso has shown superiority to all the above protocols, leading the accuracy frontier at every efficiency level on NVIDIA Jetson mobile GPUs. Specifically, Virtuoso has achieved an accuracy of 63.9%, which is more than 10% higher than some of the popular object detection models, FRCNN at 51.1%, and YOLO at 49.5%., Comment: 28 pages, 15 figures, 4 tables, ACM-TODAES

Federated learning---multi-party, distributed learning in a decentralized environment---is vulnerable to model poisoning attacks, even more so than centralized learning approaches. This is because malicious clients can collude and send in carefully tailored model updates to make the global model inaccurate. This motivated the development of Byzantine-resilient federated learning algorithms, such as Krum, Bulyan, FABA, and FoolsGold. However, a recently developed untargeted model poisoning attack showed that all prior defenses can be bypassed. The attack uses the intuition that simply by changing the sign of the gradient updates that the optimizer is computing, for a set of malicious clients, a model can be diverted from the optima to increase the test error rate. In this work, we develop TESSERACT---a defense against this directed deviation attack, a state-of-the-art model poisoning attack. TESSERACT is based on a simple intuition that in a federated learning setting, certain patterns of gradient flips are indicative of an attack. This intuition is remarkably stable across different learning algorithms, models, and datasets. TESSERACT assigns reputation scores to the participating clients based on their behavior during the training phase and then takes a weighted contribution of the clients. We show that TESSERACT provides robustness against even a white-box version of the attack., Comment: 12 pages

Federated learning allows a large number of devices to jointly learn a model without sharing data. In this work, we enable clients with limited computing power to perform action recognition, a computationally heavy task. We first perform model compression at the central server through knowledge distillation on a large dataset. This allows the model to learn complex features and serves as an initialization for model fine-tuning. The fine-tuning is required because the limited data present in smaller datasets is not adequate for action recognition models to learn complex spatio-temporal features. Because the clients present are often heterogeneous in their computing resources, we use an asynchronous federated optimization and we further show a convergence bound. We compare our approach to two baseline approaches: fine-tuning at the central server (no clients) and fine-tuning using (heterogeneous) clients using synchronous federated averaging. We empirically show on a testbed of heterogeneous embedded devices that we can perform action recognition with comparable accuracy to the two baselines above, while our asynchronous learning strategy reduces the training time by 40%, relative to synchronous learning., Comment: 13 pages, 12 figures

While previous distribution shift detection approaches can identify if a shift has occurred, these approaches cannot localize which specific features have caused a distribution shift -- a critical step in diagnosing or fixing any underlying issue. For example, in military sensor networks, users will want to detect when one or more of the sensors has been compromised, and critically, they will want to know which specific sensors might be compromised. Thus, we first define a formalization of this problem as multiple conditional distribution hypothesis tests and propose both non-parametric and parametric statistical tests. For both efficiency and flexibility, we then propose to use a test statistic based on the density model score function (i.e. gradient with respect to the input) -- which can easily compute test statistics for all dimensions in a single forward and backward pass. Any density model could be used for computing the necessary statistics including deep density models such as normalizing flows or autoregressive models. We additionally develop methods for identifying when and where a shift occurs in multivariate time-series data and show results for multiple scenarios using realistic attack models on both simulated and real world data., Comment: NeurIPS 2020 Camera Ready

We consider a security game in a setting consisting of two players (an attacker and a defender), each with a given budget to allocate towards attack and defense, respectively, of a set of nodes. Each node has a certain value to the attacker and the defender, along with a probability of being successfully compromised, which is a function of the investments in that node by both players. For such games, we characterize the optimal investment strategies by the players at the (unique) Nash Equilibrium. We then investigate the impacts of behavioral probability weighting on the investment strategies; such probability weighting, where humans overweight low probabilities and underweight high probabilities, has been identified by behavioral economists to be a common feature of human decision-making. We show via numerical experiments that behavioral decision-making by the defender causes the Nash Equilibrium investments in each node to change (where the defender overinvests in the high-value nodes and underinvests in the low-value nodes)., Comment: Accepted to appear at European Control Conference 2021

IoT systems have been facing increasingly sophisticated technical problems due to the growing complexity of these systems and their fast deployment practices. Consequently, IoT managers have to judiciously detect failures (anomalies) in order to reduce their cyber risk and operational cost. While there is a rich literature on anomaly detection in many IoT-based systems, there is no existing work that documents the use of ML models for anomaly detection in digital agriculture and in smart manufacturing systems. These two application domains pose certain salient technical challenges. In agriculture the data is often sparse, due to the vast areas of farms and the requirement to keep the cost of monitoring low. Second, in both domains, there are multiple types of sensors with varying capabilities and costs. The sensor data characteristics change with the operating point of the environment or machines, such as, the RPM of the motor. The inferencing and the anomaly detection processes therefore have to be calibrated for the operating point. In this paper, we analyze data from sensors deployed in an agricultural farm with data from seven different kinds of sensors, and from an advanced manufacturing testbed with vibration sensors. We evaluate the performance of ARIMA and LSTM models for predicting the time series of sensor data. Then, considering the sparse data from one kind of sensor, we perform transfer learning from a high data rate sensor. We then perform anomaly detection using the predicted sensor data. Taken together, we show how in these two application domains, predictive failure classification can be achieved, thus paving the way for predictive maintenance.

Adversarial examples (AEs) are images that can mislead deep neural network (DNN) classifiers via introducing slight perturbations into original images. This security vulnerability has led to vast research in recent years because it can introduce real-world threats into systems that rely on neural networks. Yet, a deep understanding of the characteristics of adversarial examples has remained elusive. We propose a new way of achieving such understanding through a recent development, namely, invertible neural models with Lipschitz continuous mapping functions from the input to the output. With the ability to invert any latent representation back to its corresponding input image, we can investigate adversarial examples at a deeper level and disentangle the adversarial example's latent representation. Given this new perspective, we propose a fast latent space adversarial example generation method that could accelerate adversarial training. Moreover, this new perspective could contribute to new ways of adversarial example detection.

We consider a security setting in which the Cyber-Physical System (CPS) is composed of subnetworks where each subnetwork is under ownership of one defender. Such CPS can be represented by an attack graph where the defenders are required to invest (subject to a budget constraint) on the graph's edges in order to protect their critical assets (where each defender's critical asset has a certain value to the defender if compromised). We model such CPS using Hybrid Input-Output Automaton (HIOA) where each subnetwork is represented by a HIOA module. We first establish the building blocks needed in our setting. We then present our model that characterizes the continuous time evolution of the investments and discrete transitions between different states (where each state represents different condition and/or perturbation) within the system. Finally, we provide a real-world CPS example to validate our modeling.

We model the behavioral biases of human decision-making in securing interdependent systems and show that such behavioral decision-making leads to a suboptimal pattern of resource allocation compared to non-behavioral (rational) decision-making. We provide empirical evidence for the existence of such behavioral bias model through a controlled subject study with 145 participants. We then propose three learning techniques for enhancing decision-making in multi-round setups. We illustrate the benefits of our decision-making model through multiple interdependent real-world systems and quantify the level of gain compared to the case in which the defenders are behavioral. We also show the benefit of our learning techniques against different attack models. We identify the effects of different system parameters on the degree of suboptimality of security outcomes due to behavioral decision-making., Comment: Accepted to appear at the 16th ACM Asia Conference on Computer and Communications Security (ASIACCS), 2021. arXiv admin note: text overlap with arXiv:2004.01958

In recent years, edge computing has become a popular choice for latency-sensitive applications like facial recognition and augmented reality because it is closer to the end users compared to the cloud. Although infrastructure providers are working toward creating managed edge networks, personal devices such as laptops and tablets, which are widely available and are underutilized, can also be used as potential edge devices. We call such devices Unmanaged Edge Devices (UEDs). Scheduling application tasks on such an unmanaged edge system is not straightforward because of three fundamental reasons-heterogeneity in the computational capacity of the UEDs, uncertainty in the availability of the UEDs (due to devices leaving the system), and interference among multiple tasks sharing a UED. In this paper, we present I-BOT, an interference-based orchestration scheme for latency-sensitive tasks on an Unmanaged Edge Platform (UEP). It minimizes the completion time of applications and is bandwidth efficient. I-BOT brings forth three innovations. First, it profiles and predicts the interference patterns of the tasks to make scheduling decisions. Second, it uses a feedback mechanism to adjust for changes in the computational capacity of the UEDs and a prediction mechanism to handle their sporadic exits. Third, it accounts for input dependence of tasks in its scheduling decision (such as, two tasks requiring the same input data). To evaluate I-BOT, we run end-to-end simulations with applications representing autonomous driving, composed of multiple tasks. We compare to two basic baselines (random and round-robin) and two state-of-the-arts, Lavea [SEC-2017] and Petrel [MSN-2018]. Compared to these baselines, I-BOT significantly reduces the average service time of application tasks. This reduction is more pronounced in dynamic heterogeneous environments, which would be the case in a UEP., Comment: 14 pages, 13 figures

Advanced video analytic systems, including scene classification and object detection, have seen widespread success in various domains such as smart cities and autonomous transportation. With an ever-growing number of powerful client devices, there is incentive to move these heavy video analytics workloads from the cloud to mobile devices to achieve low latency and real-time processing and to preserve user privacy. However, most video analytic systems are heavyweight and are trained offline with some pre-defined latency or accuracy requirements. This makes them unable to adapt at runtime in the face of three types of dynamism -- the input video characteristics change, the amount of compute resources available on the node changes due to co-located applications, and the user's latency-accuracy requirements change. In this paper we introduce ApproxDet, an adaptive video object detection framework for mobile devices to meet accuracy-latency requirements in the face of changing content and resource contention scenarios. To achieve this, we introduce a multi-branch object detection kernel (layered on Faster R-CNN), which incorporates a data-driven modeling approach on the performance metrics, and a latency SLA-driven scheduler to pick the best execution branch at runtime. We couple this kernel with approximable video object tracking algorithms to create an end-to-end video object detection system. We evaluate ApproxDet on a large benchmark video dataset and compare quantitatively to AdaScale and YOLOv3. We find that ApproxDet is able to adapt to a wide variety of contention and content characteristics and outshines all baselines, e.g., it achieves 52% lower latency and 11.1% higher accuracy over YOLOv3., Comment: Accepted to appear at the 18th ACM Conference on Embedded Networked Sensor Systems (SenSys), 2020

We study the problem of learning representations that are private yet informative, i.e., provide information about intended "ally" targets while hiding sensitive "adversary" attributes. We propose Exclusion-Inclusion Generative Adversarial Network (EIGAN), a generalized private representation learning (PRL) architecture that accounts for multiple ally and adversary attributes unlike existing PRL solutions. While centrally-aggregated dataset is a prerequisite for most PRL techniques, data in real-world is often siloed across multiple distributed nodes unwilling to share the raw data because of privacy concerns. We address this practical constraint by developing D-EIGAN, the first distributed PRL method that learns representations at each node without transmitting the source data. We theoretically analyze the behavior of adversaries under the optimal EIGAN and D-EIGAN encoders and the impact of dependencies among ally and adversary tasks on the optimization objective. Our experiments on various datasets demonstrate the advantages of EIGAN in terms of performance, robustness, and scalability. In particular, EIGAN outperforms the previous state-of-the-art by a significant accuracy margin (47% improvement), and D-EIGAN's performance is consistently on par with EIGAN under different network settings., Comment: In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS) 2022

The recent advancement of the Internet of Things (IoT) enables the possibility of data collection from diverse environments using IoT devices. However, despite the rapid advancement of low-power communication technologies, the deployment of IoT networks still faces many challenges. In this paper, we propose a hybrid, low-power, wide-area network (LPWAN) structure that can achieve wide-area communication coverage and low power consumption on IoT devices by utilizing both sub-GHz long-range radio and 2.4 GHz short-range radio. Specifically, we constructed a low-power mesh network with LoRa, a physical-layer standard that can provide long-range (kilometers) point-to-point communication using custom time-division multiple access (TDMA). Furthermore, we extended the capabilities of the mesh network by enabling ANT, an ultra-low-power, short-range communication protocol to satisfy data collection in dense device deployments. Third, we demonstrate the performance of the hybrid network with two real-world deployments at the Purdue University campus and at the university-owned farm. The results suggest that both networks have superior advantages in terms of cost, coverage, and power consumption vis-\`a-vis other IoT solutions, like LoRaWAN.

Decades of continuous scaling has reduced the energy of unit computing to virtually zero, while energy-efficient communication has remained the primary bottleneck in achieving fully energy-autonomous IoT nodes. This paper presents and analyzes the trade-offs between the energies required for communication and computation in a wireless sensor network, deployed in a mesh architecture over a 2400-acre university campus, and is targeted towards multi-sensor measurement of temperature, humidity and water nitrate concentration for smart agriculture. Several scenarios involving In-Sensor-Analytics (ISA), Collaborative Intelligence (CI) and Context-Aware-Switching (CAS) of the cluster-head during CI has been considered. A real-time co-optimization algorithm has been developed for minimizing the energy consumption in the network, hence maximizing the overall battery lifetime of individual nodes. Measurement results show that the proposed ISA consumes ~467X lower energy as compared to traditional Bluetooth Low Energy (BLE) communication, and ~69,500X lower energy as compared with Long Range (LoRa) communication. When the ISA is implemented in conjunction with LoRa, the lifetime of the node increases from a mere 4.3 hours to 66.6 days with a 230 mAh coin cell battery, while preserving more than 98% of the total information. The CI and CAS algorithms help in extending the worst-case node lifetime by an additional 50%, thereby exhibiting an overall network lifetime of ~104 days, which is >90% of the theoretical limits as posed by the leakage currents present in the system, while effectively transferring information sampled every second. A web-based monitoring system was developed to archive the measured data in a continuous manner, and to report anomalies in the measured data., Comment: 15 pages with author info, 18 figures

The field of IoT has blossomed and is positively influencing many application domains. In this paper, we bring out the unique challenges this field poses to research in computer systems and networking. The unique challenges arise from the unique characteristics of IoT systems such as the diversity of application domains where they are used and the increasingly demanding protocols they are being called upon to run (such as, video and LIDAR processing) on constrained resources (on-node and network). We show how these open challenges can benefit from foundations laid in other areas, such as, 5G cellular protocols, ML model reduction, and device-edge-cloud offloading. We then discuss the unique challenges for reliability, security, and privacy posed by IoT systems due to their salient characteristics which include heterogeneity of devices and protocols, dependence on the physical environment, and the close coupling with humans. We again show how the open research challenges benefit from reliability, security, and privacy advancements in other areas. We conclude by providing a vision for a desirable end state for IoT systems., Comment: Invited paper to the IEEE IoT Journal

We study the security of large-scale cyber-physical systems (CPS) consisting of multiple interdependent subsystems, each managed by a different defender. Defenders invest their security budgets with the goal of thwarting the spread of cyber attacks to their critical assets. We model the security investment decisions made by the defenders as a security game. While prior work has used security games to analyze such scenarios, we propose behavioral security games, in which defenders exhibit characteristics of human decision making that have been identified in behavioral economics as representing typical human cognitive biases. This is important as many of the critical security decisions in our target class of systems are made by humans. We provide empirical evidence for our behavioral model through a controlled subject experiment. We then show that behavioral decision making leads to a suboptimal pattern of resource allocation compared to non-behavioral decision making. We illustrate the effects of behavioral decision making using two representative real-world interdependent CPS. In particular, we identify the effects of the defenders' security budget availability and distribution, the degree of interdependency among defenders, and collaborative defense strategies, on the degree of suboptimality of security outcomes due to behavioral decision making. In this context, the adverse effects of behavioral decision making are most severe with moderate defense budgets. Moreover, the impact of behavioral suboptimal decision making is magnified as the degree of the interdependency between subnetworks belonging to different defenders increases. We also observe that selfish defense decisions together with behavioral decisions significantly increase security risk., Comment: 32 pages

We consider the problem of distributed inference where agents in a network observe a stream of private signals generated by an unknown state, and aim to uniquely identify this state from a finite set of hypotheses. We focus on scenarios where communication between agents is costly, and takes place over channels with finite bandwidth. To reduce the frequency of communication, we develop a novel event-triggered distributed learning rule that is based on the principle of diffusing low beliefs on each false hypothesis. Building on this principle, we design a trigger condition under which an agent broadcasts only those components of its belief vector that have adequate innovation, to only those neighbors that require such information. We prove that our rule guarantees convergence to the true state exponentially fast almost surely despite sparse communication, and that it has the potential to significantly reduce information flow from uninformative agents to informative agents. Next, to deal with finite-precision communication channels, we propose a distributed learning rule that leverages the idea of adaptive quantization. We show that by sequentially refining the range of the quantizers, every agent can learn the truth exponentially fast almost surely, while using just $1$ bit to encode its belief on each hypothesis. For both our proposed algorithms, we rigorously characterize the trade-offs between communication-efficiency and the learning rate., Comment: Accepted for publication in the IEEE Transactions of Signal Processing as a regular paper

We study the problem of designing a distributed observer for an LTI system over a time-varying communication graph. The limited existing work on this topic imposes various restrictions either on the observation model or on the sequence of communication graphs. In contrast, we propose a single-time-scale distributed observer that works under mild assumptions. Specifically, our communication model only requires strong-connectivity to be preserved over non-overlapping, contiguous intervals that are even allowed to grow unbounded over time. We show that under suitable conditions that bound the growth of such intervals, joint observability is sufficient to track the state of any discrete-time LTI system exponentially fast, at any desired rate. In fact, we also establish finite-time convergence based on our approach. Finally, we develop a variant of our algorithm that is provably robust to worst-case adversarial attacks, provided the sequence of graphs is sufficiently connected over time. The key to our approach is the notion of a "freshness-index" that keeps track of the age-of-information being diffused across the network. Such indices enable nodes to reject stale estimates of the state, and, in turn, contribute to stability of the error dynamics., Comment: arXiv admin note: text overlap with arXiv:1810.06151

We consider a system consisting of multiple interdependent assets, and a set of defenders, each responsible for securing a subset of the assets against an attacker. The interdependencies between the assets are captured by an attack graph, where an edge from one asset to another indicates that if the former asset is compromised, an attack can be launched on the latter asset. Each edge has an associated probability of successful attack, which can be reduced via security investments by the defenders. In such scenarios, we investigate the security investments that arise under certain features of human decision-making that have been identified in behavioral economics. In particular, humans have been shown to perceive probabilities in a nonlinear manner, typically overweighting low probabilities and underweighting high probabilities. We show that suboptimal investments can arise under such weighting in certain network topologies. We also show that pure strategy Nash equilibria exist in settings with multiple (behavioral) defenders, and study the inefficiency of the equilibrium investments by behavioral defenders compared to a centralized socially optimal solution.

A set of about 80 researchers, practitioners, and federal agency program managers participated in the NSF-sponsored Grand Challenges in Resilience Workshop held on Purdue campus on March 19-21, 2019. The workshop was divided into three themes: resilience in cyber, cyber-physical, and socio-technical systems. About 30 attendees in all participated in the discussions of cyber resilience. This article brings out the substantive parts of the challenges and solution approaches that were identified in the cyber resilience theme. In this article, we put forward the substantial challenges in cyber resilience in a few representative application domains and outline foundational solutions to address these challenges. These solutions fall into two broad themes: resilience-by-design and resilience-by-reaction. We use examples of autonomous systems as the application drivers motivating cyber resilience. We focus on some autonomous systems in the near horizon (autonomous ground and aerial vehicles) and also a little more distant (autonomous rescue and relief). For resilience-by-design, we focus on design methods in software that are needed for our cyber systems to be resilient. In contrast, for resilience-by-reaction, we discuss how to make systems resilient by responding, reconfiguring, or recovering at runtime when failures happen. We also discuss the notion of adaptive execution to improve resilience, execution transparently and adaptively among available execution platforms (mobile/embedded, edge, and cloud). For each of the two themes, we survey the current state, and the desired state and ways to get there. We conclude the paper by looking at the research challenges we will have to solve in the short and the mid-term to make the vision of resilient autonomous systems a reality.

Cyberphysical systems (CPS) are ubiquitous in our personal and professional lives, and they promise to dramatically improve micro-communities (e.g., urban farms, hospitals), macro-communities (e.g., cities and metropolises), urban structures (e.g., smart homes and cars), and living structures (e.g., human bodies, synthetic genomes). The question that we address in this article pertains to designing these CPS systems to be resilient-from-the-ground-up, and through progressive learning, resilient-by-reaction. An optimally designed system is resilient to both unique attacks and recurrent attacks, the latter with a lower overhead. Overall, the notion of resilience can be thought of in the light of three main sources of lack of resilience, as follows: exogenous factors, such as natural variations and attack scenarios; mismatch between engineered designs and exogenous factors ranging from DDoS (distributed denial-of-service) attacks or other cybersecurity nightmares, so called "black swan" events, disabling critical services of the municipal electrical grids and other connected infrastructures, data breaches, and network failures; and the fragility of engineered designs themselves encompassing bugs, human-computer interactions (HCI), and the overall complexity of real-world systems. In the paper, our focus is on design and deployment innovations that are broadly applicable across a range of CPS application areas., Comment: 36 pages, 2 figures, NSF-supported workshop on Grand Challenges in Resilience, held at Purdue, March 20-21, 2019

Storage has become a constrained resource on smartphones. Gaming is a popular activity on mobile devices and the explosive growth in the number of games coupled with their growing size contributes to the storage crunch. Even where storage is plentiful, it takes a long time to download and install a heavy app before it can be launched. This paper presents AppStreamer, a novel technique for reducing the storage requirements or startup delay of mobile games, and heavy mobile apps in general. AppStreamer is based on the intuition that most apps do not need the entirety of its files (images, audio and video clips, etc.) at any one time. AppStreamer can, therefore, keep only a small part of the files on the device, akin to a "cache", and download the remainder from a cloud storage server or a nearby edge server when it predicts that the app will need them in the near future. AppStreamer continuously predicts file blocks for the near future as the user uses the app, and fetches them from the storage server before the user sees a stall due to missing resources. We implement AppStreamer at the Android file system layer. This ensures that the apps require no source code or modification, and the approach generalizes across apps. We evaluate AppStreamer using two popular games: Dead Effect 2, a 3D first-person shooter, and Fire Emblem Heroes, a 2D turn-based strategy role-playing game. Through a user study, 75% and 87% of the users respectively find that AppStreamer provides the same quality of user experience as the baseline where all files are stored on the device. AppStreamer cuts down the storage requirement by 87% for Dead Effect 2 and 86% for Fire Emblem Heroes., Comment: 12 pages; EWSN 2020

Adversarial examples (AEs) are images that can mislead deep neural network (DNN) classifiers via introducing slight perturbations into original images. Recent work has shown that detecting AEs can be more effective against AEs than preventing them from being generated. However, the state-of-the-art AE detection still shows a high false positive rate, thereby rejecting a considerable amount of normal images. To address this issue, we propose HAWKEYE, which is a separate neural network that analyzes the output layer of the DNN, and detects AEs. HAWKEYE's AE detector utilizes a quantized version of an input image as a reference, and is trained to distinguish the variation characteristics of the DNN output on an input image from the DNN output on its reference image. We also show that cascading our AE detectors that are trained for different quantization step sizes can drastically reduce a false positive rate, while keeping a detection rate high.

Videos take a lot of time to transport over the network, hence running analytics on the live video on embedded or mobile devices has become an important system driver. Considering that such devices, e.g., surveillance cameras or AR/VR gadgets, are resource constrained, creating lightweight deep neural networks (DNNs) for embedded devices is crucial. None of the current approximation techniques for object classification DNNs can adapt to changing runtime conditions, e.g., changes in resource availability on the device, the content characteristics, or requirements from the user. In this paper, we introduce ApproxNet, a video object classification system for embedded or mobile clients. It enables novel dynamic approximation techniques to achieve desired inference latency and accuracy trade-off under changing runtime conditions. It achieves this by enabling two approximation knobs within a single DNN model, rather than creating and maintaining an ensemble of models (e.g., MCDNN [MobiSys-16]. We show that ApproxNet can adapt seamlessly at runtime to these changes, provides low and stable latency for the image and video frame classification problems, and show the improvement in accuracy and latency over ResNet [CVPR-16], MCDNN [MobiSys-16], MobileNets [Google-17], NestDNN [MobiCom-18], and MSDNet [ICLR-18]., Comment: This paper has been accepted to appear in ACM Transactions on Sensor Networks in 2021

YouTube is the leading social media platform for sharing videos. As a result, it is plagued with misleading content that includes staged videos presented as real footages from an incident, videos with misrepresented context and videos where audio/video content is morphed. We tackle the problem of detecting such misleading videos as a supervised classification task. We develop UCNet - a deep network to detect fake videos and perform our experiments on two datasets - VAVD created by us and publicly available FVC [8]. We achieve a macro averaged F-score of 0.82 while training and testing on a 70:30 split of FVC, while the baseline model scores 0.36. We find that the proposed model generalizes well when trained on one dataset and tested on the other., Comment: Accepted at European Conference on Information Retrieval(ECIR) 2019. 7 Pages

The performance of most error-correction algorithms that operate on genomic sequencer reads is dependent on the proper choice of its configuration parameters, such as the value of k in k-mer based techniques. In this work, we target the problem of finding the best values of these configuration parameters to optimize error correction. We perform this in a data-driven manner, due to the observation that different configuration parameters are optimal for different datasets, i.e., from different instruments and organisms. We use language modeling techniques from the Natural Language Processing (NLP) domain in our algorithmic suite, Athena, to automatically tune the performance-sensitive configuration parameters. Through the use of N-Gram and Recurrent Neural Network (RNN) language modeling, we validate the intuition that the EC performance can be computed quantitatively and efficiently using the perplexity metric, prevalent in NLP. After training the language model, we show that the perplexity metric calculated for runtime data has a strong negative correlation with the correction of the erroneous NGS reads. Therefore, we use the perplexity metric to guide a hill climbing-based search, converging toward the best $k$-value. Our approach is suitable for both de novo and comparative sequencing (resequencing), eliminating the need for a reference genome to serve as the ground truth. This is important because the use of a reference genome often carries forward the biases along the stages of the pipeline., Comment: 10 main pages, 7 references and appendices

We study the problem of collaboratively estimating the state of a discrete-time LTI process by a network of sensor nodes interacting over a time-varying directed communication graph. Existing approaches to this problem either (i) make restrictive assumptions on the dynamical model, or (ii) make restrictive assumptions on the sequence of communication graphs, or (iii) require multiple consensus iterations between consecutive time-steps of the dynamics, or (iv) require higher-dimensional observers. In this paper, we develop a distributed observer that operates on a single time-scale, is of the same dimension as that of the state, and works under mild assumptions of joint observability of the sensing model, and joint strong-connectivity of the sequence of communication graphs. Our approach is based on the notion of a novel "freshness-index" that keeps track of the age-of-information being diffused across the network. In particular, such indices enable nodes to reject stale information regarding the state of the system, and in turn, help achieve stability of the estimation error dynamics. Based on the proposed approach, the estimate of each node can be made to converge to the true state exponentially fast, at any desired convergence rate. In fact, we argue that finite-time convergence can also be achieved through a suitable selection of the observer gains. Our proof of convergence is self-contained, and employs simple arguments from linear system theory and graph theory.

Distributed monitoring plays a crucial role in managing the activities of cloud-based datacenters. System administrators have long relied on monitoring systems such as Nagios and Ganglia to obtain status alerts on their desktop-class machines. However, the popularity of mobile devices is pushing the community to develop datacenter monitoring solutions for smartphone-class devices. Here we lay out desirable characteristics of such smartphone-based monitoring and identify quantitatively the shortcomings from directly applying existing solutions to this domain. Then we introduce a possible design that addresses some of these shortcomings and provide results from an early prototype, called MAVIS, using one month of monitoring data from approximately 3,000 machines hosted by Purdue's central IT organization., Comment: ACM Classification (2012): Data center networks; System management; Ubiquitous and mobile computing systems and tools

The widespread proliferation of sensor nodes in the era of Internet of Things (IoT) coupled with increasing sensor fidelity and data acquisition modality is expected to generate 3+ Exabytes of data per day by 2018. Since most of these IoT devices will be wirelessly connected at the last few feet, wireless communication is an integral part of the future IoT scenario. The ever-shrinking size of unit computation (Moore's Law) and continued improvements in efficient communication (Shannon's Law) is expected to harness the true potential of the IoT revolution and produce dramatic societal impact. However, reducing size of IoT nodes and lack of significant improvement in energy-storage density leads to reducing energy-availability. Moreover, smaller size and energy means less resources available for securing IoT nodes, making the energy-sparse low-cost leaf nodes of the network as prime targets for attackers. In this paper, we survey six prominent wireless technologies with respect to the three dimensions - security, energy efficiency, and communication capacity. We point out the state-of-the-art, open issues, and the road ahead for promising research directions., Comment: Awaiting publication in IEEE Internet Computing magazine