Internet of Things (IoT) devices and applications are being deployed in our homes and workplaces and in our daily lives. These devices often rely on continuous data collection and machine learning models for analytics and actuations. However, this approach introduces a number of privacy and efficiency challenges, as the service operator can perform arbitrary inferences on the available data. Recently, advances in edge processing have paved the way for more efficient, and private, data processing at the source for simple tasks and lighter models, though they remain a challenge for larger, and more complicated models. In this paper, we present a hybrid approach for breaking down large, complex deep neural networks for cooperative, privacy-preserving analytics. To this end, instead of performing the whole operation on the cloud, we let an IoT device to run the initial layers of the neural network, and then send the output to the cloud to feed the remaining layers and produce the final result. We manipulate the model with Siamese fine-tuning and propose a noise addition mechanism to ensure that the output of the user's device contains no extra information except what is necessary for the main task, preventing any secondary inference on the data. We then evaluate the privacy benefits of this approach based on the information exposed to the cloud service. We also asses the local inference cost of different layers on a modern handset. Our evaluations show that by using Siamese fine-tuning and at a small processing cost, we can greatly reduce the level of unnecessary, potentially sensitive information in the personal data, and thus achieving the desired trade-off between utility, privacy and performance.

Abstract-Todays smartphones come equipped with a range of advanced sensors capable of sensing motion, orientation, audio as well as environmental data with high accuracy. With the existence of application distribution channels such as the Apple App Store and the Google Play Store, researchers can distribute applications and collect large scale data in ways that previously were not possible. Motivated by the lack of a universal, multiplatform sensing library, in this work we present the design and implementation of SensingKit, an open-source continuous sensing system that supports both iOS and Android mobile devices. One of the unique features of SensingKit is the support of the latest beacon technologies based on Bluetooth Smart (BLE), such as iBeacon TM and Eddystone TM . We evaluate and compare the power consumption of each supported sensor individually, using an iPhone 5S device running on iOS 9. We believe that this platform will be beneficial to all researchers and developers who plan to use mobile sensing technology in large-scale experiments.

The success of live comedy depends on a performer's ability to “work” an audience. Ethnographic studies suggest that this involves the co-ordinated use of subtle social signals such as body orientation, gesture, gaze by both performers and audience members. Robots provide a unique opportunity to test the effects of these signals experimentally. Using a life-size humanoid robot, programmed to perform a stand-up comedy routine, we manipulated the robot's patterns of gesture and gaze and examined their effects on the real-time responses of a live audience. The strength and type of responses were captured using SHORE™computer vision analytics. The results highlight the complex, reciprocal social dynamics of performer and audience behavior. People respond more positively when the robot looks at them, negatively when it looks away and performative gestures also contribute to different patterns of audience response. This demonstrates how the responses of individual audience members depend on the specific interaction they're having with the performer. This work provides insights into how to design more effective, more socially engaging forms of robot interaction that can be used in a variety of service contexts.

et al. Effective patient-clinician interaction to improve treatment outcomes for patients with psychosis: a mixed-methods design. Programme Grants Appl Res 2017;5(6). Programme Grants for Applied ResearchISSN 2050-4322 (Print) ISSN 2050-4330 (Online) This journal is a member of and subscribes to the principles of the Committee on Publication Ethics (COPE) (www.publicationethics.org/).Editorial contact: nihredit@southampton.ac.ukThe full PGfAR archive is freely available to view online at www.journalslibrary.nihr.ac.uk/pgfar. Print-on-demand copies can be purchased from the report pages of the NIHR Journals Library website: www.journalslibrary.nihr.ac.uk Criteria for inclusion in the Programme Grants for Applied Research journalReports are published in Programme Grants for Applied Research (PGfAR) if (1) they have resulted from work for the PGfAR programme, and (2) they are of a sufficiently high scientific quality as assessed by the reviewers and editors. Programme Grants for Applied Research programmeThe Programme Grants for Applied Research (PGfAR) programme, part of the National Institute for Health Research (NIHR), was set up in 2006 to produce independent research findings that will have practical application for the benefit of patients and the NHS in the relatively near future. The Programme is managed by the NIHR Central Commissioning Facility (CCF) with strategic input from the Programme Director.The programme is a national response mode funding scheme that aims to provide evidence to improve health outcomes in England through promotion of health, prevention of ill health, and optimal disease management (including safety and quality), with particular emphasis on conditions causing significant disease burden.For more information about the PGfAR programme please visit the website: http://www.nihr.ac.uk/funding/programme-grants-forapplied-research.htm This reportThe research reported in this issue of the journal was funded by PGfAR as project number RP-PG-0108-10023. The contractual start date was in April 2010. The final report began editorial review in October 2015 and was accepted for publication in August 2016. As the funder, the PGfAR programme agreed the research questions and study designs in advance with the investigators. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The PGfAR editors and production house have tried to ensure the accuracy of the authors' report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, CCF, NETSCC, PGfAR or the Department of Health. If there are verbatim quotations included in this publicatio...

We present and evaluate Deep Private-Feature Extractor (DPFE), a deep model which is trained and evaluated based on information theoretic constraints. Using the selective exchange of information between a user's device and a service provider, DPFE enables the user to prevent certain sensitive information from being shared with a service provider, while allowing them to extract approved information using their model. We introduce and utilize the log-rank privacy, a novel measure to assess the effectiveness of DPFE in removing sensitive information and compare different models based on their accuracy-privacy tradeoff. We then implement and evaluate the performance of DPFE on smartphones to understand its complexity, resource demands, and efficiency tradeoffs. Our results on benchmark image datasets demonstrate that under moderate resource utilization, DPFE can achieve high accuracy for primary tasks while preserving the privacy of sensitive information.