Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems最新文献

KingFisher: an Industrial Security Framework based on Variational Autoencoders KingFisher:一个基于变分自编码器的工业安全框架

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 2019-11-10 DOI: 10.1145/3362743.3362961

Giuseppe Bernieri, M. Conti, F. Turrin

{"title":"KingFisher: an Industrial Security Framework based on Variational Autoencoders","authors":"Giuseppe Bernieri, M. Conti, F. Turrin","doi":"10.1145/3362743.3362961","DOIUrl":"https://doi.org/10.1145/3362743.3362961","url":null,"abstract":"The recent evolution of edge computing favored the Industrial Internet of Things (IIoT) growth, opening dangerous surfaces of vulnerabilities. In this distributed sensor system scenario, due to the insecure interactions between Information Technology (IT) and Operational Technology (OT) networks, cyber-physical threats could lead to destructive consequences for environments and population safety. To deal with industrial cyber-physical security, modern anomaly detection systems implement innovative Machine Learning (ML) techniques. Unfortunately, current solutions still fail to provide an effective prevention to complex industrial threats. In this paper, we present KingFisher, an Intrusion Detection System (IDS) framework based on ML. KingFisher is, to the best of our knowledge, the first solution that looks independently at IT and OT traffic, but also from sensors deployed to capture side-channel physical processes data (e.g., vibrations, background noise). Thanks to this feature, KingFisher can detect attacks that other systems would ignore. As our tests report, the correlation of inferred physical processes status with OT-network and IT-network data can give insights into suspicious and anomalous activities targeting industrial networks. For our framework, we use the Variational Autoencoders (VAEs), an unsupervised neural network model, to categorize data without a priori knowledge of the dataset. We evaluate the detection capabilities and performances of KingFisher in a proof of concept simulated industrial scenario under cyber-physical attacks. Our preliminary results show that KingFisher identifies attacks on both network and physical layers.","PeriodicalId":425595,"journal":{"name":"Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129043155","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}

引用次数: 12

Skipping RNN State Updates without Retraining the Original Model 跳过RNN状态更新而不重新训练原始模型

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 2019-11-10 DOI: 10.1145/3362743.3362965

Jin Tao, Urmish Thakker, Ganesh S. Dasika, Jesse G. Beu

{"title":"Skipping RNN State Updates without Retraining the Original Model","authors":"Jin Tao, Urmish Thakker, Ganesh S. Dasika, Jesse G. Beu","doi":"10.1145/3362743.3362965","DOIUrl":"https://doi.org/10.1145/3362743.3362965","url":null,"abstract":"Recurrent Neural Networks (RNNs) break a time-series input (or a sentence) into multiple time-steps (or words) and process it one time-step (word) at a time. However, not all of these time-steps (words) need to be processed to determine the final output accurately. Prior work has exploited this intuition by incorporating an additional predictor in front of the RNN model to prune time-steps that are not relevant. However, they jointly train the predictor and the RNN model, allowing one to learn from the mistakes of the other. In this work we present a method to skip RNN time-steps without retraining or fine tuning the original RNN model. Using an ideal predictor, we show that even without retraining the original model, we can train a predictor to skip 45% of steps for the SST dataset and 80% of steps for the IMDB dataset without impacting the model accuracy. We show that the decision to skip is not easy by comparing against 5 different baselines based on solutions derived from domain knowledge. Finally, we present a case study about the cost and accuracy benefits of realizing such a predictor. This realistic predictor on the SST dataset is able to reduce the computation by more than 25% with at most 0.3% loss in accuracy while being 40× smaller than the original RNN model.","PeriodicalId":425595,"journal":{"name":"Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114507670","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}

引用次数: 13

Efficient Sparse Processing in Smart Home Applications 智能家居应用中的高效稀疏处理

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 2019-11-10 DOI: 10.1145/3362743.3362963

Rishikanth Chandrasekaran, Yunhui Guo, Anthony Thomas, M. Menarini, M. Ostertag, Yeseong Kim, T. Simunic

{"title":"Efficient Sparse Processing in Smart Home Applications","authors":"Rishikanth Chandrasekaran, Yunhui Guo, Anthony Thomas, M. Menarini, M. Ostertag, Yeseong Kim, T. Simunic","doi":"10.1145/3362743.3362963","DOIUrl":"https://doi.org/10.1145/3362743.3362963","url":null,"abstract":"In recent years, smart home technology has become prevalant and important for various applications. A typical smart home system consists of sensing nodes sending raw data to a cloud server which performs inference using a Machine Learning (ML) model trained offline. This approach suffers from high energy and communication costs and raises privacy concerns. To address these issues researchers proposed hierarchy aware models which distributes the inference computations across the sensor network with each node processing a part of the inference. While hierarchical models reduce these overheads significantly they are computationally intensive to run on resource constrained devices which are typical to smart home deployments. In this work we present a novel approach combining Hierarchy aware Neural Networks (HNN) with variational dropout technique to generate sparse models which have low computational overhead allowing them to be run on edge devices with limited resources. We evaluate our approach using an extensive real-world smart home deployment consisting of several edge devices. Measurements across different devices show that without significant loss of accuracy, energy consumption can be reduced by up to 35% over state-of-the-art.","PeriodicalId":425595,"journal":{"name":"Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115400802","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}

引用次数: 0

Embedded Deep Inference in Practice: Case for Model Partitioning 实践中的嵌入式深度推理:模型划分案例

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 2019-11-10 DOI: 10.1145/3362743.3362964

Swarnava Dey, Arijit Mukherjee, A. Pal, P. Balamuralidhar

{"title":"Embedded Deep Inference in Practice: Case for Model Partitioning","authors":"Swarnava Dey, Arijit Mukherjee, A. Pal, P. Balamuralidhar","doi":"10.1145/3362743.3362964","DOIUrl":"https://doi.org/10.1145/3362743.3362964","url":null,"abstract":"With increased focus on in situ analytics, artificial intelligence (AI) algorithms are getting deployed on embedded devices at the network edge. Growing popularity of Deep Learning (DL) and inference largely due to minimization of feature engineering, availability of pre-trained models and fine-tunable datasets especially in image and video analytics, have made these de-facto standard. However, the embedded systems employing these models are often resource constrained and fail to handle scenarios where arrival rate and input data volume increase over a given time period. This has a direct effect on the storage and network usage of such devices, rendering the traditional strategies of input buffering and network offloading ineffective. This paper investigates the use of dynamic layer-wise partitioning and partial execution of DL inference phase to enable inelastic embedded systems to support varying sensing rates and large data volume. The proposed partial execution scheme and partitioning algorithm perform better than standard frame-wise inference methods, when evaluated using workloads of few popular CNNs used in standard object detection models.","PeriodicalId":425595,"journal":{"name":"Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117186098","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}

引用次数: 8

Emotion Filtering at the Edge 边缘的情感过滤

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 2019-09-18 DOI: 10.1145/3362743.3362960

Ranya Aloufi, H. Haddadi, David Boyle

引用次数: 14

Finding Dory in the Crowd: Detecting Social Interactions using Multi-Modal Mobile Sensing 在人群中寻找多莉:使用多模态移动传感检测社会互动

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 2018-08-30 DOI: 10.1145/3362743.3362959

Kleomenis Katevas, Katrin Hänsel, R. Clegg, Ilias Leontiadis, H. Haddadi, L. Tokarchuk

{"title":"Finding Dory in the Crowd: Detecting Social Interactions using Multi-Modal Mobile Sensing","authors":"Kleomenis Katevas, Katrin Hänsel, R. Clegg, Ilias Leontiadis, H. Haddadi, L. Tokarchuk","doi":"10.1145/3362743.3362959","DOIUrl":"https://doi.org/10.1145/3362743.3362959","url":null,"abstract":"Remembering our day-to-day social interactions is challenging even if you aren't a blue memory challenged fish. The ability to automatically detect and remember these types of interactions is not only beneficial for individuals interested in their behavior in crowded situations, but also of interest to those who analyze crowd behavior. Currently, detecting social interactions is often performed using ethnographic studies, computer vision techniques and manual annotation-based data analysis. However, mobile phones offer easier means for data collection that is easy to analyze and can preserve the user's privacy. In this work, we present a system for detecting stationary social interactions inside crowds, leveraging multi-modal mobile sensing data such as Bluetooth Smart (BLE), accelerometer and gyroscope. To inform the development of such system we conducted a study with 24 participants where we asked them to socialize with each other for 45 minutes. We built a machine learning system based on gradient-boosted trees that predicts both 1:1 and group interactions with a 30.2% performance increase compared to a proximity-based approach. By utilizing a community detection-based method, we further detected the various group formation that exist within the crowd. Using mobile phone sensors already carried by the majority of people in a crowd makes our approach particularly well suited to real-life analysis of crowd behavior and influence strategies.","PeriodicalId":425595,"journal":{"name":"Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131344796","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}

引用次数: 18

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems 第一届传感器系统边缘机器学习研讨会论文集

Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems Pub Date : 1900-01-01 DOI: 10.1145/3362743

引用次数: 0