Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00072
Yangming Zhao, Jingyuan Fan, Lu Su, Tongyu Song, Sheng Wang, C. Qiao
More and more applications learn from the data collected by the edge devices. Conventional learning methods, such as gathering all the raw data to train an ultimate model in a centralized way, or training a target model in a distributed manner under the parameter server framework, suffer a high communication cost. In this paper, we design Select Neighbors and Parameters (SNAP), a communication efficient distributed machine learning framework, to mitigate the communication cost. A distinct feature of SNAP is that the edge servers act as peers to each other. Specifically, in SNAP, every edge server hosts a copy of the global model, trains it with the local data, and periodically updates the local parameters based on the weighted sum of the parameters from its neighbors (i.e., peers) only (i.e., without pulling the parameters from all other edge servers). Different from most of the previous works on consensus optimization in which the weight matrix to update parameter values is predefined, we propose a scheme to optimize the weight matrix based on the network topology, and hence the convergence rate can be improved. Another key idea in SNAP is that only the parameters which have been changed significantly since the last iteration will be sent to the neighbors. Both theoretical analysis and simulations show that SNAP can achieve the same accuracy performance as the centralized training method. Compared to the state-of-the-art communication-aware distributed learning scheme TernGrad, SNAP incurs a significantly lower (99.6% lower) communication cost.
{"title":"SNAP: A Communication Efficient Distributed Machine Learning Framework for Edge Computing","authors":"Yangming Zhao, Jingyuan Fan, Lu Su, Tongyu Song, Sheng Wang, C. Qiao","doi":"10.1109/ICDCS47774.2020.00072","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00072","url":null,"abstract":"More and more applications learn from the data collected by the edge devices. Conventional learning methods, such as gathering all the raw data to train an ultimate model in a centralized way, or training a target model in a distributed manner under the parameter server framework, suffer a high communication cost. In this paper, we design Select Neighbors and Parameters (SNAP), a communication efficient distributed machine learning framework, to mitigate the communication cost. A distinct feature of SNAP is that the edge servers act as peers to each other. Specifically, in SNAP, every edge server hosts a copy of the global model, trains it with the local data, and periodically updates the local parameters based on the weighted sum of the parameters from its neighbors (i.e., peers) only (i.e., without pulling the parameters from all other edge servers). Different from most of the previous works on consensus optimization in which the weight matrix to update parameter values is predefined, we propose a scheme to optimize the weight matrix based on the network topology, and hence the convergence rate can be improved. Another key idea in SNAP is that only the parameters which have been changed significantly since the last iteration will be sent to the neighbors. Both theoretical analysis and simulations show that SNAP can achieve the same accuracy performance as the centralized training method. Compared to the state-of-the-art communication-aware distributed learning scheme TernGrad, SNAP incurs a significantly lower (99.6% lower) communication cost.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125414784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00122
Yusaku Hayamizu, K. Matsuzono, H. Asaeda
Applications of Information-Centric Networking (ICN) technology to future internet of things (IoT) and distributed edge/fog computing are widely discussed in various research committees. In this paper, we demonstrate a real-time video streaming scenario using CeforeSim, an NS-3 based ICN simulator. CeforeSim is based on Cefore, an open-source implementation of ICN, which is compliant with the CCNx packet format standardized by the IRTF ICN Research Group (ICNRG). The virtual interfaces provisioned in CeforeSim expedite seamless interaction between the simulated nodes and physical nodes that run the Cefore applications, thereby affording performance evaluations in various scenarios, such as handover of mobile nodes, large-scale sensor networks, and distributed edge/fog computing with the real environments.
{"title":"Real-Time Video Streaming using CeforeSim: Simulator to the Real World","authors":"Yusaku Hayamizu, K. Matsuzono, H. Asaeda","doi":"10.1109/ICDCS47774.2020.00122","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00122","url":null,"abstract":"Applications of Information-Centric Networking (ICN) technology to future internet of things (IoT) and distributed edge/fog computing are widely discussed in various research committees. In this paper, we demonstrate a real-time video streaming scenario using CeforeSim, an NS-3 based ICN simulator. CeforeSim is based on Cefore, an open-source implementation of ICN, which is compliant with the CCNx packet format standardized by the IRTF ICN Research Group (ICNRG). The virtual interfaces provisioned in CeforeSim expedite seamless interaction between the simulated nodes and physical nodes that run the Cefore applications, thereby affording performance evaluations in various scenarios, such as handover of mobile nodes, large-scale sensor networks, and distributed edge/fog computing with the real environments.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"182 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126775652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00082
Jiacheng Shang, Jie Wu
With the rapid popularity of cameras on various devices, video chat has become one of the major ways for communication, such as online meetings. However, the recent progress of face reenactment techniques enables attackers to generate fake facial videos and use others’ identities. To protect video chats against fake facial videos, we propose a new defense system to significantly raise the bar for face reenactment-assisted attacks. Compared with existing works, our system has three major strengths. First, our system does not require extra hardware or intense computational resources. Second, it follows the normal video chat process and does not significantly degrade the user experience. Third, our system does not need to collect training data from attackers and new users, which means it can be quickly launched on new devices. We developed a prototype and conducted comprehensive evaluations. Experimental results show that our system can provide an average true acceptance rate of at least 92.5% for legitimate users and reject the attacker with mean accuracy of at least 94.4% for a single detection.
{"title":"Protecting Real-time Video Chat against Fake Facial Videos Generated by Face Reenactment","authors":"Jiacheng Shang, Jie Wu","doi":"10.1109/ICDCS47774.2020.00082","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00082","url":null,"abstract":"With the rapid popularity of cameras on various devices, video chat has become one of the major ways for communication, such as online meetings. However, the recent progress of face reenactment techniques enables attackers to generate fake facial videos and use others’ identities. To protect video chats against fake facial videos, we propose a new defense system to significantly raise the bar for face reenactment-assisted attacks. Compared with existing works, our system has three major strengths. First, our system does not require extra hardware or intense computational resources. Second, it follows the normal video chat process and does not significantly degrade the user experience. Third, our system does not need to collect training data from attackers and new users, which means it can be quickly launched on new devices. We developed a prototype and conducted comprehensive evaluations. Experimental results show that our system can provide an average true acceptance rate of at least 92.5% for legitimate users and reject the attacker with mean accuracy of at least 94.4% for a single detection.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124975215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00102
Zhen Xiao, Tao Chen, Yang Liu, Zhenjiang Li
Mobile phones nowadays are equipped with at least dual microphones. We find when a user is typing on a phone, the sounds generated from the vibration caused by finger’s tapping on the screen surface can be captured by both microphones, and these recorded sounds alone are informative enough to infer the user’s keystrokes. This ability can be leveraged to enable useful application designs, while it also raises a crucial privacy risk that the private information typed by users on mobile phones has a great potential to be leaked through such a recognition ability. In this paper, we address two key design issues and demonstrate, more importantly alarm people, that this risk is possible, which could be related to many of us when we use our mobile phones. We implement our proposed techniques in a prototype system and conduct extensive experiments. The evaluation results indicate promising successful rates for more than 4000 keystrokes from different users on various types of mobile phones.
{"title":"Mobile Phones Know Your Keystrokes through the Sounds from Finger’s Tapping on the Screen","authors":"Zhen Xiao, Tao Chen, Yang Liu, Zhenjiang Li","doi":"10.1109/ICDCS47774.2020.00102","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00102","url":null,"abstract":"Mobile phones nowadays are equipped with at least dual microphones. We find when a user is typing on a phone, the sounds generated from the vibration caused by finger’s tapping on the screen surface can be captured by both microphones, and these recorded sounds alone are informative enough to infer the user’s keystrokes. This ability can be leveraged to enable useful application designs, while it also raises a crucial privacy risk that the private information typed by users on mobile phones has a great potential to be leaked through such a recognition ability. In this paper, we address two key design issues and demonstrate, more importantly alarm people, that this risk is possible, which could be related to many of us when we use our mobile phones. We implement our proposed techniques in a prototype system and conduct extensive experiments. The evaluation results indicate promising successful rates for more than 4000 keystrokes from different users on various types of mobile phones.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133914314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00048
Jian Kang, D. Lin
Waiting in a long queue at a traffic light has been a common and frustrating experience of the majority of daily commuters, which not only wastes valuable time but also pollutes our environments. With the advances in autonomous vehicles and their collaboration capabilities, the previous jamming intersection has a great potential to be turned into weaving traffic flows that no longer need to stop. Towards this envision, we propose a novel autonomous vehicle traffic coordination system called DASH. Specifically, DASH has a comprehensive model to represent intersections and vehicle status. It can constantly process a large volume of vehicle information of various kinds, resolve scheduling conflicts of all vehicles coming towards the intersection, and generate the optimal travel plan for each individual vehicle in real time to guide vehicles passing intersections in a safe and highly efficient way. Unlike existing works on the autonomous traffic control which are limited to certain types of intersections and lack considerations of practicability, our proposed DASH algorithm is universal for any kind of intersections yields the near-maximum throughput while still ensuring riding comfort that prevents sudden stop and acceleration. We have conducted extensive experiments to evaluate the DASH system in the scenarios of different types of intersections and different traffic flows. Our experimental results demonstrate its practicality, effectiveness, and efficiency.
{"title":"DASH: A Universal Intersection Traffic Management System for Autonomous Vehicles","authors":"Jian Kang, D. Lin","doi":"10.1109/ICDCS47774.2020.00048","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00048","url":null,"abstract":"Waiting in a long queue at a traffic light has been a common and frustrating experience of the majority of daily commuters, which not only wastes valuable time but also pollutes our environments. With the advances in autonomous vehicles and their collaboration capabilities, the previous jamming intersection has a great potential to be turned into weaving traffic flows that no longer need to stop. Towards this envision, we propose a novel autonomous vehicle traffic coordination system called DASH. Specifically, DASH has a comprehensive model to represent intersections and vehicle status. It can constantly process a large volume of vehicle information of various kinds, resolve scheduling conflicts of all vehicles coming towards the intersection, and generate the optimal travel plan for each individual vehicle in real time to guide vehicles passing intersections in a safe and highly efficient way. Unlike existing works on the autonomous traffic control which are limited to certain types of intersections and lack considerations of practicability, our proposed DASH algorithm is universal for any kind of intersections yields the near-maximum throughput while still ensuring riding comfort that prevents sudden stop and acceleration. We have conducted extensive experiments to evaluate the DASH system in the scenarios of different types of intersections and different traffic flows. Our experimental results demonstrate its practicality, effectiveness, and efficiency.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129839250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00025
Christos Profentzas, M. Almgren, O. Landsiedel
With the rise of the Internet of Things (IoT), billions of devices ranging from simple sensors to smart-phones will participate in billions of micropayments. However, current centralized solutions are unable to handle a massive number of micropayments from untrusted devices.Blockchains are promising technologies suitable for solving some of these challenges. Particularly, permissionless blockchains such as Ethereum and Bitcoin have drawn the attention of the research community. However, the increasingly large-scale deployments of blockchain reveal some of their scalability limitations. Prominent proposals to scale the payment system include off-chain protocols such as payment channels. However, the leading proposals assume powerful nodes with an always-on connection and frequent synchronization. These assumptions require in practice significant communication, memory, and computation capacity, whereas IoT devices face substantial constraints in these areas. Existing approaches also do not capture the logic and process of IoT, where applications need to process locally collected sensor data to allow for full use of IoT micro-payments.In this paper, we present TinyEVM, a novel system to generate and execute off-chain smart contracts based on sensor data. TinyEVM’s goal is to enable IoT devices to perform micro-payments and, at the same time, address the device constraints. We investigate the trade-offs of executing smart contracts on low-power IoT devices using TinyEVM. We test our system with 7,000 publicly verified smart contracts, where TinyEVM achieves to deploy 93% of them without any modification. Finally, we evaluate the execution of off-chain smart contracts in terms of run-time performance, energy, and memory requirements on IoT devices. Notably, we find that low-power devices can deploy a smart contract in 215 ms on average, and they can complete an off-chain payment in 584 ms on average.
{"title":"TinyEVM: Off-Chain Smart Contracts on Low-Power IoT Devices","authors":"Christos Profentzas, M. Almgren, O. Landsiedel","doi":"10.1109/ICDCS47774.2020.00025","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00025","url":null,"abstract":"With the rise of the Internet of Things (IoT), billions of devices ranging from simple sensors to smart-phones will participate in billions of micropayments. However, current centralized solutions are unable to handle a massive number of micropayments from untrusted devices.Blockchains are promising technologies suitable for solving some of these challenges. Particularly, permissionless blockchains such as Ethereum and Bitcoin have drawn the attention of the research community. However, the increasingly large-scale deployments of blockchain reveal some of their scalability limitations. Prominent proposals to scale the payment system include off-chain protocols such as payment channels. However, the leading proposals assume powerful nodes with an always-on connection and frequent synchronization. These assumptions require in practice significant communication, memory, and computation capacity, whereas IoT devices face substantial constraints in these areas. Existing approaches also do not capture the logic and process of IoT, where applications need to process locally collected sensor data to allow for full use of IoT micro-payments.In this paper, we present TinyEVM, a novel system to generate and execute off-chain smart contracts based on sensor data. TinyEVM’s goal is to enable IoT devices to perform micro-payments and, at the same time, address the device constraints. We investigate the trade-offs of executing smart contracts on low-power IoT devices using TinyEVM. We test our system with 7,000 publicly verified smart contracts, where TinyEVM achieves to deploy 93% of them without any modification. Finally, we evaluate the execution of off-chain smart contracts in terms of run-time performance, energy, and memory requirements on IoT devices. Notably, we find that low-power devices can deploy a smart contract in 215 ms on average, and they can complete an off-chain payment in 584 ms on average.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133756643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00040
S. Issa, Miguel Viegas, Pedro Raminhas, Nuno Machado, M. Matos, P. Romano
Deterministic databases (DDs) are a promising approach for replicating data across different replicas. A fundamental component of DDs is a deterministic concurrency control algorithm that, given a set of transactions in a specific order, guarantees that their execution always results in the same serial order. State-of-the-art approaches either rely on single threaded execution or on the knowledge of read- and write-sets of transactions to achieve this goal. The former yields poor performance in multi-core machines while the latter requires either manual inputs from the user — a time-consuming and error prone task — or a reconnaissance phase that increases both the latency and abort rates of transactions.In this paper, we present Prognosticator, a novel deterministic database system. Rather than relying on manual transaction classification or an expert programmer, Prognosticator employs Symbolic Execution to build fine-grained transaction profiles (at the key-level). These profiles are then used by Prognosticator’s novel deterministic concurrency control algorithm to execute transactions with a high degree of parallelism.Our experimental evaluation, based on both TPC-C and RUBiS benchmarks, shows that Prognosticator can achieve up to 5× higher throughput with respect to state-of-the-art solutions.
{"title":"Exploiting Symbolic Execution to Accelerate Deterministic Databases","authors":"S. Issa, Miguel Viegas, Pedro Raminhas, Nuno Machado, M. Matos, P. Romano","doi":"10.1109/ICDCS47774.2020.00040","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00040","url":null,"abstract":"Deterministic databases (DDs) are a promising approach for replicating data across different replicas. A fundamental component of DDs is a deterministic concurrency control algorithm that, given a set of transactions in a specific order, guarantees that their execution always results in the same serial order. State-of-the-art approaches either rely on single threaded execution or on the knowledge of read- and write-sets of transactions to achieve this goal. The former yields poor performance in multi-core machines while the latter requires either manual inputs from the user — a time-consuming and error prone task — or a reconnaissance phase that increases both the latency and abort rates of transactions.In this paper, we present Prognosticator, a novel deterministic database system. Rather than relying on manual transaction classification or an expert programmer, Prognosticator employs Symbolic Execution to build fine-grained transaction profiles (at the key-level). These profiles are then used by Prognosticator’s novel deterministic concurrency control algorithm to execute transactions with a high degree of parallelism.Our experimental evaluation, based on both TPC-C and RUBiS benchmarks, shows that Prognosticator can achieve up to 5× higher throughput with respect to state-of-the-art solutions.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130178799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00089
Hisham Alasmary, Ahmed A. Abusnaina, Rhongho Jang, M. Abuhamad, Afsah Anwar, Daehun Nyang, David A. Mohaisen
Deep learning algorithms have been widely used for security applications, including malware detection and classification. Recent results have shown that those algorithms are vulnerable to adversarial examples, whereby a small perturbation in the input sample may result in misclassification. In this paper, we systematically tackle the problem of adversarial examples detection in the control flow graph (CFG) based classifiers for malware detection using Soteria. Unique to Soteria, we use both density-based and level-based labels for CFG labeling to yield a consistent representation, a random walk-based traversal approach for feature extraction, and n-gram based module for feature representation. End-to-end, Soteria’s representation ensures a simple yet powerful randomization property of the used classification features, making it difficult even for a powerful adversary to launch a successful attack. Soteria also employs a deep learning approach, consisting of an auto-encoder for detecting adversarial examples, and a CNN architecture for detecting and classifying malware samples. We evaluate the performance of Soteria, using a large dataset consisting of 16,814 IoT samples, and demonstrate its superiority in comparison with state-of-the-art approaches. In particular, Soteria yields an accuracy rate of 97.79% for detecting AEs, and 99.91% overall accuracy for classification malware families.
{"title":"Soteria: Detecting Adversarial Examples in Control Flow Graph-based Malware Classifiers","authors":"Hisham Alasmary, Ahmed A. Abusnaina, Rhongho Jang, M. Abuhamad, Afsah Anwar, Daehun Nyang, David A. Mohaisen","doi":"10.1109/ICDCS47774.2020.00089","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00089","url":null,"abstract":"Deep learning algorithms have been widely used for security applications, including malware detection and classification. Recent results have shown that those algorithms are vulnerable to adversarial examples, whereby a small perturbation in the input sample may result in misclassification. In this paper, we systematically tackle the problem of adversarial examples detection in the control flow graph (CFG) based classifiers for malware detection using Soteria. Unique to Soteria, we use both density-based and level-based labels for CFG labeling to yield a consistent representation, a random walk-based traversal approach for feature extraction, and n-gram based module for feature representation. End-to-end, Soteria’s representation ensures a simple yet powerful randomization property of the used classification features, making it difficult even for a powerful adversary to launch a successful attack. Soteria also employs a deep learning approach, consisting of an auto-encoder for detecting adversarial examples, and a CNN architecture for detecting and classifying malware samples. We evaluate the performance of Soteria, using a large dataset consisting of 16,814 IoT samples, and demonstrate its superiority in comparison with state-of-the-art approaches. In particular, Soteria yields an accuracy rate of 97.79% for detecting AEs, and 99.91% overall accuracy for classification malware families.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114627043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00185
Tian Huang, Tao Luo, Joey Tianyi Zhou
Learn in-situ is a growing trend for Edge AI. Training deep neural network (DNN) on edge devices is challenging because both energy and memory are constrained. Low precision training helps to reduce the energy cost of a single training iteration, but that does not necessarily translate to energy savings for the whole training process, because low precision could slows down the convergence rate. One evidence is that most works for low precision training keep an fp32 copy of the model during training, which in turn imposes memory requirements on edge devices. In this work we propose Adaptive Precision Training. It is able to save both total training energy cost and memory usage at the same time. We use model of the same precision for both forward and backward pass in order to reduce memory usage for training. Through evaluating the progress of training, APT allocates layer-wise precision dynamically so that the model learns quicker for longer time. APT provides an application specific hyper-parameter for users to play trade-off between training energy cost, memory usage and accuracy. Experiment shows that APT achieves more than 50% saving on training energy and memory usage with limited accuracy loss. 20% more savings of training energy and memory usage can be achieved in return for a 1% sacrifice in accuracy loss.
{"title":"Adaptive Precision Training for Resource Constrained Devices","authors":"Tian Huang, Tao Luo, Joey Tianyi Zhou","doi":"10.1109/ICDCS47774.2020.00185","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00185","url":null,"abstract":"Learn in-situ is a growing trend for Edge AI. Training deep neural network (DNN) on edge devices is challenging because both energy and memory are constrained. Low precision training helps to reduce the energy cost of a single training iteration, but that does not necessarily translate to energy savings for the whole training process, because low precision could slows down the convergence rate. One evidence is that most works for low precision training keep an fp32 copy of the model during training, which in turn imposes memory requirements on edge devices. In this work we propose Adaptive Precision Training. It is able to save both total training energy cost and memory usage at the same time. We use model of the same precision for both forward and backward pass in order to reduce memory usage for training. Through evaluating the progress of training, APT allocates layer-wise precision dynamically so that the model learns quicker for longer time. APT provides an application specific hyper-parameter for users to play trade-off between training energy cost, memory usage and accuracy. Experiment shows that APT achieves more than 50% saving on training energy and memory usage with limited accuracy loss. 20% more savings of training energy and memory usage can be achieved in return for a 1% sacrifice in accuracy loss.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134366924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-01DOI: 10.1109/ICDCS47774.2020.00080
Oskar Lundström, M. Raynal, E. Schiller
Fault-tolerant distributed applications require communication abstractions with provable guarantees on message deliveries. For example, Set-Constrained Delivery Broadcast (SCD-broadcast) is a communication abstraction for broadcasting messages in a manner that, if a process delivers a set of messages that includes m and later delivers a set of messages that includes m , no process delivers first a set of messages that includes m′ and later a set of messages that includes m.Imbs et al. proposed this communication abstraction and its first implementation. They have demonstrated that SCD-broadcast has the computational power of read/write registers and allows for an easy building of distributed objects such as snapshot objects and consistent counters. Imbs et al. focused on fault-tolerant implementations for asynchronous message-passing systems that are prone to process crashes. This paper aims to design an even more robust SCD-broadcast communication abstraction, namely a self-stabilizing SCD-broadcast. In addition to process and communication failures, self-stabilizing algorithms can recover after the occurrence of arbitrary transient faults; these faults represent any violation of the assumptions according to which the system was designed to operate (as long as the algorithm code stays intact).This work proposes the first self-stabilizing SCD-broadcast algorithm for asynchronous message-passing systems that are prone to process crash failures. The proposed self-stabilizing SCD-broadcast algorithm has an $mathcal{O}(1)$ stabilization time (in terms of asynchronous cycles). The communication costs of our algorithm are similar to the ones of the non-self-stabilizing state-of-the-art. The main differences are that our proposal considers repeated gossiping of $mathcal{O}(1)$ bits messages and deals with bounded space (which is a prerequisite for self-stabilization). We advance the state-of-the-art also by two new self-stabilizing applications: an atomic construction of snapshot objects and sequentially consistent counters.
{"title":"Self-Stabilizing Set-Constrained Delivery Broadcast (extended abstract)","authors":"Oskar Lundström, M. Raynal, E. Schiller","doi":"10.1109/ICDCS47774.2020.00080","DOIUrl":"https://doi.org/10.1109/ICDCS47774.2020.00080","url":null,"abstract":"Fault-tolerant distributed applications require communication abstractions with provable guarantees on message deliveries. For example, Set-Constrained Delivery Broadcast (SCD-broadcast) is a communication abstraction for broadcasting messages in a manner that, if a process delivers a set of messages that includes m and later delivers a set of messages that includes m , no process delivers first a set of messages that includes m′ and later a set of messages that includes m.Imbs et al. proposed this communication abstraction and its first implementation. They have demonstrated that SCD-broadcast has the computational power of read/write registers and allows for an easy building of distributed objects such as snapshot objects and consistent counters. Imbs et al. focused on fault-tolerant implementations for asynchronous message-passing systems that are prone to process crashes. This paper aims to design an even more robust SCD-broadcast communication abstraction, namely a self-stabilizing SCD-broadcast. In addition to process and communication failures, self-stabilizing algorithms can recover after the occurrence of arbitrary transient faults; these faults represent any violation of the assumptions according to which the system was designed to operate (as long as the algorithm code stays intact).This work proposes the first self-stabilizing SCD-broadcast algorithm for asynchronous message-passing systems that are prone to process crash failures. The proposed self-stabilizing SCD-broadcast algorithm has an $mathcal{O}(1)$ stabilization time (in terms of asynchronous cycles). The communication costs of our algorithm are similar to the ones of the non-self-stabilizing state-of-the-art. The main differences are that our proposal considers repeated gossiping of $mathcal{O}(1)$ bits messages and deals with bounded space (which is a prerequisite for self-stabilization). We advance the state-of-the-art also by two new self-stabilizing applications: an atomic construction of snapshot objects and sequentially consistent counters.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132876884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：