RESUMO
The advancements in machine learning opened a new opportunity to bring intelligence to the low-end Internet-of-Things nodes such as microcontrollers. Conventional machine learning deployment has high memory and compute footprint hindering their direct deployment on ultra resource-constrained microcontrollers. This paper highlights the unique requirements of enabling onboard machine learning for microcontroller class devices. Researchers use a specialized model development workflow for resource-limited applications to ensure the compute and latency budget is within the device limits while still maintaining the desired performance. We characterize a closed-loop widely applicable workflow of machine learning model development for microcontroller class devices and show that several classes of applications adopt a specific instance of it. We present both qualitative and numerical insights into different stages of model development by showcasing several use cases. Finally, we identify the open research challenges and unsolved questions demanding careful considerations moving forward.
RESUMO
Machine learning at the extreme edge has enabled a plethora of intelligent, time-critical, and remote applications. However, deploying interpretable artificial intelligence systems that can perform high-level symbolic reasoning and satisfy the underlying system rules and physics within the tight platform resource constraints is challenging. In this paper, we introduce TinyNS, the first platform-aware neurosymbolic architecture search framework for joint optimization of symbolic and neural operators. TinyNS provides recipes and parsers to automatically write microcontroller code for five types of neurosymbolic models, combining the context awareness and integrity of symbolic techniques with the robustness and performance of machine learning models. TinyNS uses a fast, gradient-free, black-box Bayesian optimizer over discontinuous, conditional, numeric, and categorical search spaces to find the best synergy of symbolic code and neural networks within the hardware resource budget. To guarantee deployability, TinyNS talks to the target hardware during the optimization process. We showcase the utility of TinyNS by deploying microcontroller-class neurosymbolic models through several case studies. In all use cases, TinyNS outperforms purely neural or purely symbolic approaches while guaranteeing execution on real hardware.
RESUMO
End-to-end deep learning models are increasingly applied to safety-critical human activity recognition (HAR) applications, e.g., healthcare monitoring and smart home control, to reduce developer burden and increase the performance and robustness of prediction models. However, integrating HAR models in safety-critical applications requires trust, and recent approaches have aimed to balance the performance of deep learning models with explainable decision-making for complex activity recognition. Prior works have exploited the compositionality of complex HAR (i.e., higher-level activities composed of lower-level activities) to form models with symbolic interfaces, such as concept-bottleneck architectures, that facilitate inherently interpretable models. However, feature engineering for symbolic concepts-as well as the relationship between the concepts-requires precise annotation of lower-level activities by domain experts, usually with fixed time windows, all of which induce a heavy and error-prone workload on the domain expert. In this paper, we introduce X-CHAR , an eXplainable Complex Human Activity Recognition model that doesn't require precise annotation of low-level activities, offers explanations in the form of human-understandable, high-level concepts, while maintaining the robust performance of end-to-end deep learning models for time series data. X-CHAR learns to model complex activity recognition in the form of a sequence of concepts. For each classification, X-CHAR outputs a sequence of concepts and a counterfactual example as the explanation. We show that the sequence information of the concepts can be modeled using Connectionist Temporal Classification (CTC) loss without having accurate start and end times of low-level annotations in the training dataset-significantly reducing developer burden. We evaluate our model on several complex activity datasets and demonstrate that our model offers explanations without compromising the prediction accuracy in comparison to baseline models. Finally, we conducted a mechanical Turk study to show that the explanations provided by our model are more understandable than the explanations from existing methods for complex activity recognition.
RESUMO
Inertial navigation provides a small footprint, low-power, and low-cost pathway for localization in GPS-denied environments on extremely resource-constrained Internet-of-Things (IoT) platforms. Traditionally, application-specific heuristics and physics-based kinematic models are used to mitigate the curse of drift in inertial odometry. These techniques, albeit lightweight, fail to handle domain shifts and environmental non-linearities. Recently, deep neural-inertial sequence learning has shown superior odometric resolution in capturing non-linear motion dynamics without human knowledge over heuristic-based methods. These AI-based techniques are data-hungry, suffer from excessive resource usage, and cannot guarantee following the underlying system physics. This paper highlights the unique methods, opportunities, and challenges in porting real-time AI-enhanced inertial navigation algorithms onto IoT platforms. First, we discuss how platform-aware neural architecture search coupled with ultra-lightweight model backbones can yield neural-inertial odometry models that are 31-134× smaller yet achieve or exceed the localization resolution of state-of-the-art AI-enhanced techniques. The framework can generate models suitable for locating humans, animals, underwater sensors, aerial vehicles, and precision robots. Next, we showcase how techniques from neurosymbolic AI can yield physics-informed and interpretable neural-inertial navigation models. Afterward, we present opportunities for fine-tuning pre-trained odometry models in a new domain with as little as 1 minute of labeled data, while discussing inexpensive data collection and labeling techniques. Finally, we identify several open research challenges that demand careful consideration moving forward.
RESUMO
Wearable devices like smartwatches and smart wristbands have gained substantial popularity in recent years. However, their small interfaces create inconvenience and limit computing functionality. To fill this gap, we propose ViWatch, which enables robust finger interactions under deployment variations, and relies on a single IMU sensor that is ubiquitous in COTS smartwatches. To this end, we design an unsupervised Siamese adversarial learning method. We built a real-time system on commodity smartwatches and tested it with over one hundred volunteers. Results show that the system accuracy is about 97% over a week. In addition, it is resistant to deployment variations such as different hand shapes, finger activity strengths, and smartwatch positions on the wrist. We also developed a number of mobile applications using our interactive system and conducted a user study where all participants preferred our un-supervised approach to supervised calibration. The demonstration of ViWatch is shown at https://youtu.be/N5-ggvy2qfI.
RESUMO
Human societies depend on marine ecosystems, but their degradation continues. Toward mitigating this decline, new and more effective ways to precisely measure the status and condition of marine environments are needed alongside existing rebuilding strategies. Here, we provide an overview of how sensors and wearable technology developed for humans could be adapted to improve marine monitoring. We describe barriers that have slowed the transition of this technology from land to sea, update on the developments in sensors to advance ocean observation and advocate for more widespread use of wearables on marine organisms in the wild and in aquaculture. We propose that large-scale use of wearables could facilitate the concept of an 'internet of marine life' that might contribute to a more robust and effective observation system for the oceans and commercial aquaculture operations. These observations may aid in rationalizing strategies toward conservation and restoration of marine communities and habitats.
Assuntos
Ecossistema , Dispositivos Eletrônicos Vestíveis , Humanos , Organismos Aquáticos , Oceanos e Mares , TecnologiaRESUMO
Deep inertial sequence learning has shown promising odometric resolution over model-based approaches for trajectory estimation in GPS-denied environments. However, existing neural inertial dead-reckoning frameworks are not suitable for real-time deployment on ultra-resource-constrained (URC) devices due to substantial memory, power, and compute bounds. Current deep inertial odometry techniques also suffer from gravity pollution, high-frequency inertial disturbances, varying sensor orientation, heading rate singularity, and failure in altitude estimation. In this paper, we introduce TinyOdom, a framework for training and deploying neural inertial models on URC hardware. TinyOdom exploits hardware and quantization-aware Bayesian neural architecture search (NAS) and a temporal convolutional network (TCN) backbone to train lightweight models targetted towards URC devices. In addition, we propose a magnetometer, physics, and velocity-centric sequence learning formulation robust to preceding inertial perturbations. We also expand 2D sequence learning to 3D using a model-free barometric g-h filter robust to inertial and environmental variations. We evaluate TinyOdom for a wide spectrum of inertial odometry applications and target hardware against competing methods. Specifically, we consider four applications: pedestrian, animal, aerial, and underwater vehicle dead-reckoning. Across different applications, TinyOdom reduces the size of neural inertial models by 31× to 134× with 2.5m to 12m error in 60 seconds, enabling the direct deployment of models on URC devices while still maintaining or exceeding the localization resolution over the state-of-the-art. The proposed barometric filter tracks altitude within ±0.1m and is robust to inertial disturbances and ambient dynamics. Finally, our ablation study shows that the introduced magnetometer, physics, and velocity-centric sequence learning formulation significantly improve localization performance even with notably lightweight models.
RESUMO
Smart ear-worn devices (called earables) are being equipped with various onboard sensors and algorithms, transforming earphones from simple audio transducers to multi-modal interfaces making rich inferences about human motion and vital signals. However, developing sensory applications using earables is currently quite cumbersome with several barriers in the way. First, time-series data from earable sensors incorporate information about physical phenomena in complex settings, requiring machine-learning (ML) models learned from large-scale labeled data. This is challenging in the context of earables because large-scale open-source datasets are missing. Secondly, the small size and compute constraints of earable devices make on-device integration of many existing algorithms for tasks such as human activity and head-pose estimation difficult. To address these challenges, we introduce Auritus an extendable and open-source optimization toolkit designed to enhance and replicate earable applications. Auritus serves two primary functions. Firstly, Auritus handles data collection, pre-processing, and labeling tasks for creating customized earable datasets using graphical tools. The system includes an open-source dataset with 2.43 million inertial samples related to head and full-body movements, consisting of 34 head poses and 9 activities from 45 volunteers. Secondly, Auritus provides a tightly-integrated hardware-in-the-loop (HIL) optimizer and TinyML interface to develop lightweight and real-time machine-learning (ML) models for activity detection and filters for head-pose tracking. To validate the utlity of Auritus, we showcase three sample applications, namely fall detection, spatial audio rendering, and augmented reality (AR) interfacing. Auritus recognizes activities with 91% leave 1-out test accuracy (98% test accuracy) using real-time models as small as 6-13 kB. Our models are 98-740× smaller and 3-6% more accurate over the state-of-the-art. We also estimate head pose with absolute errors as low as 5 degrees using 20kB filters, achieving up to 1.6× precision improvement over existing techniques. We make the entire system open-source so that researchers and developers can contribute to any layer of the system or rapidly prototype their applications using our dataset and algorithms.
RESUMO
Application latency requirements, privacy, and security concerns have naturally pushed computing onto smartphone and IoT devices in a decentralized manner. In response to these demands, researchers have developed micro-runtimes for WebAssembly (Wasm) on IoT devices to enable streaming applications to a runtime that can run the target binaries that are independent of the device. However, the migration of Wasm and the associated security research has neglected the urgent needs of access control on bare-metal, memory management unit (MMU)-less IoT devices that are sensing and actuating upon the physical environment. This paper presents Aerogel, an access control framework that addresses security gaps between the bare-metal IoT devices and the Wasm execution environment concerning access control for sensors, actuators, processor energy usage, and memory usage. In particular, we treat the runtime as a multi-tenant environment, where each Wasm-based application is a tenant. We leverage the inherent sandboxing mechanisms of Wasm to enforce the access control policies to sensors and actuators without trusting the bare-metal operating system. We evaluate our approach on a representative IoT development board: a cortex-M4 based development board (nRF52840). Our results show that Aerogel can effectively enforce compute resource and peripheral access control policies while introducing as little as 0.19% to 1.04% runtime overhead and consuming only 18.8% to 45.9% extra energy.
RESUMO
The increased availability of on-body sensors gives researchers access to rich time-series data, many of which are related to human health conditions. Sharing such data can allow cross-institutional collaborations that create advanced data-driven models to make inferences on human well-being. However, such data are usually considered privacy-sensitive, and publicly sharing this data may incur significant privacy concerns. In this work, we seek to protect clinical time-series data against membership inference attacks, while maximally retaining the data utility. We achieve this by adding an imperceptible noise to the raw data. Known as adversarial perturbations, the noise is specially trained to force a deep learning model to make inference mistakes (in our case, mispredicting user identities). Our preliminary results show that our solution can better protect the data from membership inference attacks than the baselines, while succeeding in all the designed data quality checks.
RESUMO
Public release of wrist-worn motion sensor data is growing. They enable and accelerate research in developing new algorithms to passively track daily activities, resulting in improved health and wellness utilities of smartwatches and activity trackers. But, when combined with sensitive attribute inference attack and linkage attack via re-identification of the same user in multiple datasets, undisclosed sensitive attributes can be revealed to unintended organizations with potentially adverse consequences for unsuspecting data contributing users. To guide both users and data collecting researchers, we characterize the re-identification risks inherent in motion sensor data collected from wrist-worn devices in users' natural environment. For this purpose, we use an open-set formulation, train a deep learning architecture with a new loss function, and apply our model to a new data set consisting of 10 weeks of daily sensor wearing by 353 users. We find that re-identification risk increases with an increase in the activity intensity. On average, such risk is 96% for a user when sharing a full day of sensor data.
RESUMO
Artificial intelligence (AI) systems hold great promise as decision-support tools, but we must be able to identify and understand their inevitable mistakes if they are to fulfill this potential. This is particularly true in domains where the decisions are high-stakes, such as law, medicine, and the military. In this Perspective, we describe the particular challenges for AI decision support posed in military coalition operations. These include having to deal with limited, low-quality data, which inevitably compromises AI performance. We suggest that these problems can be mitigated by taking steps that allow rapid trust calibration so that decision makers understand the AI system's limitations and likely failures and can calibrate their trust in its outputs appropriately. We propose that AI services can achieve this by being both interpretable and uncertainty-aware. Creating such AI systems poses various technical and human factors challenges. We review these challenges and recommend directions for future research.
RESUMO
A fully implantable wireless pressure sensor system was developed to monitor bladder pressures in vivo. The system comprises a small commercial pressure die connected via catheter to amplifying electronics, a microcontroller, wireless transmitter, battery, and a personal digital assistant (PDA) or computer to receive the wireless data. The sensor is fully implantable and transmits pressure data once every second with a pressure detection range of 1.5 psi gauge and a resolution of 0.02 psi. In vitro calibration measurements of the device showed a high degree of linearity and excellent temporal response. The implanted device performed continuously in vivo in several porcine studies lasting over 3 days. This system can be adapted for other pressure readings, as well as other vital sign measurements; it represents the first step in developing a ubiquitous sensing platform for telemedicine and remote patient monitoring.
Assuntos
Computadores , Monitorização Fisiológica/instrumentação , Monitorização Fisiológica/métodos , Pressão , Próteses e Implantes , Bexiga Urinária/fisiopatologia , Animais , Feminino , Suínos , Transdutores de PressãoRESUMO
Biologging is a scientific endeavor that studies the environment and animals within it by outfitting the latter with sensors of their dynamics as they roam freely in their natural habitats. As wearable technologies advance for the monitoring of human health, it may be instructive to reflect on the successes and failures of biologging in field biology over the past few decades. Several lessons may be of value. Physiological sensors can "encode" for a wider number of states than the one explicitly targeted, although the limits of this are debatable. The combination of orthogonal sensors turns out to be critical to delivering a high value data set. Sensor fusion and engineering for longevity are also important for success. This Perspective highlights successful strategies for biologging that hold promise for human health monitoring.
Assuntos
Monitoramento Biológico/estatística & dados numéricos , Pesquisa Interdisciplinar/estatística & dados numéricos , Dispositivos Eletrônicos Vestíveis , Tecnologia sem Fio , Animais , Monitoramento Biológico/instrumentação , Telefone Celular , Humanos , Internet das CoisasRESUMO
Mobile devices have become an essential part of our daily lives. By virtue of both their increasing computing power and the recent progress made in AI, mobile devices evolved to act as intelligent assistants in many tasks rather than a mere way of making phone calls. However, popular and commonly used tools and frameworks for machine intelligence are still lacking the ability to make proper use of the available heterogeneous computing resources on mobile devices. In this paper, we study the benefits of utilizing the heterogeneous (CPU and GPU) computing resources available on commodity android devices while running deep learning models. We leveraged the heterogeneous computing framework RenderScript to accelerate the execution of deep learning models on commodity Android devices. Our system is implemented as an extension to the popular open-source framework TensorFlow. By integrating our acceleration framework tightly into TensorFlow, machine learning engineers can now easily make benefit of the heterogeneous computing resources on mobile devices without the need of any extra tools. We evaluate our system on different android phones models to study the trade-offs of running different neural network operations on the GPU. We also compare the performance of running different models architectures such as convolutional and recurrent neural networks on CPU only vs using heterogeneous computing resources. Our result shows that although GPUs on the phones are capable of offering substantial performance gain in matrix multiplication on mobile devices. Therefore, models that involve multiplication of large matrices can run much faster (approx. 3 times faster in our experiments) due to GPU support.
RESUMO
Mobile technologies that drive just-in-time ecological momentary assessments and interventions provide an unprecedented view into user behaviors and opportunities to manage chronic conditions. The success of these methods rely on engaging the user at the appropriate moment, so as to maximize questionnaire and task completion rates. However, mobile operating systems provide little support to precisely specify the contextual conditions in which to notify and engage the user, and study designers often lack the expertise to build context-aware software themselves. To address this problem, we have developed Emu, a framework that eases the development of context-aware study applications by providing a concise and powerful interface for specifying temporal- and contextual-constraints for task notifications. In this paper we present the design of the Emu API and demonstrate its use in capturing a range of scenarios common to smartphone-based study applications.
RESUMO
Although different interaction modalities have been proposed in the field of human-computer interface (HCI), only a few of these techniques could reach the end users because of scalability and usability issues. Given the popularity and the growing number of IoT devices, selecting one out of many devices becomes a hurdle in a typical smarthome environment. Therefore, an easy-to-learn, scalable, and non-intrusive interaction modality has to be explored. In this paper, we propose a pointing approach to interact with devices, as pointing is arguably a natural way for device selection. We introduce SeleCon for device selection and control which uses an ultra-wideband (UWB) equipped smartwatch. To interact with a device in our system, people can point to the device to select it then draw a hand gesture in the air to specify a control action. To this end, SeleCon employs inertial sensors for pointing gesture detection and a UWB transceiver for identifying the selected device from ranging measurements. Furthermore, SeleCon supports an alphabet of gestures that can be used for controlling the selected devices. We performed our experiment in a 9m-by-10m lab space with eight deployed devices. The results demonstrate that SeleCon can achieve 84.5% accuracy for device selection and 97% accuracy for hand gesture recognition. We also show that SeleCon is power efficient to sustain daily use by turning off the UWB transceiver, when a user's wrist is stationary.
RESUMO
The development and validation studies of new multisensory biomarkers and sensor-triggered interventions requires collecting raw sensor data with associated labels in the natural field environment. Unlike platforms for traditional mHealth apps, a software platform for such studies needs to not only support high-rate data ingestion, but also share raw high-rate sensor data with researchers, while supporting high-rate sense-analyze-act functionality in real-time. We present mCerebrum, a realization of such a platform, which supports high-rate data collections from multiple sensors with realtime assessment of data quality. A scalable storage architecture (with near optimal performance) ensures quick response despite rapidly growing data volume. Micro-batching and efficient sharing of data among multiple source and sink apps allows reuse of computations to enable real-time computation of multiple biomarkers without saturating the CPU or memory. Finally, it has a reconfigurable scheduler which manages all prompts to participants that is burden- and context-aware. With a modular design currently spanning 23+ apps, mCerebrum provides a comprehensive ecosystem of system services and utility apps. The design of mCerebrum has evolved during its concurrent use in scientific field studies at ten sites spanning 106,806 person days. Evaluations show that compared with other platforms, mCerebrum's architecture and design choices support 1.5 times higher data rates and 4.3 times higher storage throughput, while causing 8.4 times lower CPU usage. CCS CONCEPTS: ⢠Human-centered computing â Ubiquitous and mobile computing; Ubiquitous and mobile computing systems and tools; ⢠Computer systems organization â Embedded and cyber-physical systems. ACM REFERENCE FORMAT: Syed Monowar Hossain, Timothy Hnat, Nazir Saleheen, Nusrat Jahan Nasrin, Joseph Noor, Bo-Jhang Ho, Tyson Condie, Mani Srivastava, and Santosh Kumar. 2017. mCerebrum: A Mobile Sensing Software Platform for Development and Validation of Digital Biomarkers and Interventions. In Proceedings of SenSys '17, Delft, Netherlands, November 6-8, 2017, 14 pages.
RESUMO
Differential privacy concepts have been successfully used to protect anonymity of individuals in population-scale analysis. Sharing of mobile sensor data, especially physiological data, raise different privacy challenges, that of protecting private behaviors that can be revealed from time series of sensor data. Existing privacy mechanisms rely on noise addition and data perturbation. But the accuracy requirement on inferences drawn from physiological data, together with well-established limits within which these data values occur, render traditional privacy mechanisms inapplicable. In this work, we define a new behavioral privacy metric based on differential privacy and propose a novel data substitution mechanism to protect behavioral privacy. We evaluate the efficacy of our scheme using 660 hours of ECG, respiration, and activity data collected from 43 participants and demonstrate that it is possible to retain meaningful utility, in terms of inference accuracy (90%), while simultaneously preserving the privacy of sensitive behaviors.