{"type":"ScholarlyArticle","title":"Privacy Issues Regarding the Application of DNNs to Activity-Recognition using Wearables and Its Countermeasures by Use of Adversarial Training","authors":[{"name":"Yusuke Iwasawa","slug":"yusuke-iwasawa","given":"Yusuke","family":"Iwasawa","affiliation":"The University of Tokyo , Japan"},{"name":"Kotaro Nakayama","slug":"kotaro-nakayama","given":"Kotaro","family":"Nakayama","affiliation":"The University of Tokyo , Japan"},{"name":"Ikuko Yairi","slug":"ikuko-yairi","given":"Ikuko","family":"Yairi","affiliation":"Sophia University , Japan"},{"name":"Yutaka Matsuo","slug":"yutaka-matsuo","given":"Yutaka","family":"Matsuo","affiliation":"The University of Tokyo , Japan"}],"abstract":"Deep neural networks have been successfully applied to activity recognition with wearables in terms of recognition performance. However, the black-box nature of neural networks could lead to privacy concerns. Namely, generally it is hard to expect what neural networks learn from data, and so they possibly learn features that highly discriminate user-information unintentionally, which increases the risk of information-disclosure. In this study, we analyzed the features learned by conventional deep neural networks when applied to data of wearables to confirm this phenomenon.Based on the results of our analysis, we propose the use of an adversarial training framework to suppress the risk of sensitive/unintended information disclosure. Our proposed model considers both an adversarial user classifier and a regular activity-classifier during training, which allows the model to learn representations that help the classifier to distinguish the activities but which, at the same time, prevents it from accessing user-discriminative information. This paper provides an empirical validation of the privacy issue and efficacy of the proposed method using three activity recognition tasks based on data of wearables. The empirical validation shows that our proposed method suppresses the concerns without any significant performance degradation, compared to conventional deep nets on all three tasks.","doi":"10.24963/ijcai.2017/268","url":"https://ijcai.org/proceedings/2017/268","pdf":"https://ijcai.org/proceedings/2017/0268.pdf","markdown":"https://ijcai.org/proceedings/2017/268.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/268","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"Machine Learning A-R","track":"Main track","number":268,"pages":{"first":1930,"last":1936},"keywords":["Machine Learning: Neural Networks","Multidisciplinary Topics and Applications: AI and Ubiquitous Computing Systems","Machine Learning: Deep Learning"],"videos":[],"references":[{"position":1,"raw":"[Ajakan et al., 2014] Hana Ajakan, Pascal Germain, Hugo Larochelle, Franc ¸ois Laviolette, and Mario Marchand. Domain-adversarial neural networks. In NIPS Workshop on Transfer and Multi-Task Learning: Theory meets Prac- tice, 2014.","title":"Domain-adversarial neural networks","authors":"Hana Ajakan, Pascal Germain, Hugo Larochelle, Mario Franc ¸ois Laviolette, Marchand","year":"2014","venue":"NIPS Workshop on Transfer and Multi-Task Learning: Theory meets Practice","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chakraborty et al., 2014] Supriyo Chakraborty, Chenguang Shen, Kasturi Rangan Raghavan, Yasser Shoukry, Matt Millar, and Mani Srivastava. ipShield: A Framework for Enforcing Context-Aware Privacy. In Workshop on Secure Data Management, pages 85-100, 2014.","title":"ipShield: A Framework for Enforcing Context-Aware Privacy","authors":"Supriyo Chakraborty, Chenguang Shen, Kasturi Rangan Raghavan, Yasser Shoukry, Matt Millar, Mani Srivastava","year":"2014","venue":"Workshop on Secure Data Management","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Consolvo et al., 2008] Sunny Consolvo, David W McDon- ald, Tammy Toscos, Mike Y Chen, Jon Froehlich, Bev- erly Harrison, Predrag Klasnja, Anthony LaMarca, Louis LeGrand, Ryan Libby, et al. Activity sensing in the wild: a field trial of ubifit garden. In Proc. of CHI, pages 1797- 1806. ACM, 2008.","title":"Activity sensing in the wild: a field trial of ubifit garden","authors":"Sunny Consolvo, David W Mcdonald, Tammy Toscos, Mike Y Chen, Jon Froehlich, Beverly Harrison, Predrag Klasnja, Anthony Lamarca, Louis Legrand, Ryan Libby","year":"2008","venue":"Proc. of CHI","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Edwards and Storkey, 2016] Harrison Edwards and Amos Storkey. Censoring representations with an adversary. In Proc. of ICLR, 2016.","title":"Censoring representations with an adversary","authors":"Storkey Edwards, Harrison Edwards, Amos Storkey","year":"2016","venue":"Proc. of ICLR","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Goodfellow et al., 2014] Ian Goodfellow, Jean Pouget- Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Gen- erative adversarial nets. In Proc. of NIPS, pages 2672- 2680, 2014.","title":"Generative adversarial nets","authors":"Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio","year":"2014","venue":"Proc. of NIPS","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Hammerla et al., 2013] Nils Y Hammerla, Reuben Kirkham, Peter Andras, and Thomas Ploetz. On preserving statistical characteristics of accelerometry data using their empirical cumulative distribution. In Proc. of International Symposium on Wearable Computers (ISWC), pages 65-68. ACM, 2013.","title":"On preserving statistical characteristics of accelerometry data using their empirical cumulative distribution","authors":"Y Nils, Reuben Hammerla, Peter Kirkham, Thomas Andras, Ploetz","year":"2013","venue":"Proc. of International Symposium on Wearable Computers (ISWC)","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Hammerla et al., 2015] Nils Yannick Hammerla, James Fisher, Peter Andras, Lynn Rochester, Richard Walker, and Thomas Plötz. PD Disease State Assessment in Naturalis- tic Environments Using Deep Learning. In Proc. of AAAI, pages 1742-1748. AAAI Press, 2015.","title":"PD Disease State Assessment in Naturalistic Environments Using Deep Learning","authors":"Nils Yannick Hammerla, James Fisher, Peter Andras, Lynn Rochester, Richard Walker, Thomas Plötz","year":"2015","venue":"Proc. of AAAI","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Hammerla et al., 2016] Nils Y Hammerla, Shane Halloran, and Thomas Ploetz. Deep, convolutional, and recurrent models for human activity recognition using wearables. In Proc. of IJCAI, pages 1533-1540. Citeseer, 2016.","title":"Deep, convolutional, and recurrent models for human activity recognition using wearables","authors":"Y Nils, Shane Hammerla, Thomas Halloran, Ploetz","year":"2016","venue":"Proc. of IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2016/220"},{"position":9,"raw":"[Kingma and Ba, 2015] Diederik Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In Proc. of ICLR, 2015.","title":"Adam: A method for stochastic optimization","authors":"Diederik Kingma, Jimmy Ba","year":"2015","venue":"Proc. of ICLR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Lee et al., 2012] Youngki Lee, SS Iyengar, Chulhong Min, Younghyun Ju, Seungwoo Kang, Taiwoo Park, Jinwon Lee, Yunseok Rhee, and Junehwa Song. Mobicon: a mo- bile context-monitoring platform. Communications of the ACM, 55(3):54-65, 2012.","title":"Mobicon: a mobile context-monitoring platform","authors":"Lee","year":"2012","venue":"Communications of the ACM","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Ordóñez and Roggen, 2016] Francisco Javier Ordóñez and Daniel Roggen. Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recog- nition. Sensors, 16(1):115, 2016.","title":"Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recognition","authors":"Roggen ; Ordóñez, Javier Francisco, Daniel Ordóñez, Roggen","year":"2016","venue":"Sensors","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Plötz et al., 2011] Thomas Plötz, Nils Y Hammerla, and Patrick Olivier. Feature learning for activity recognition in ubiquitous computing. In Proc. of IJCAI, number 1, pages 1729-1734, 2011.","title":"Feature learning for activity recognition in ubiquitous computing","authors":"Thomas Plötz, Nils Y Hammerla, Patrick Olivier","year":"2011","venue":"Proc. of IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2011/290"},{"position":13,"raw":"[Rachuri et al., 2010] Kiran K Rachuri, Mirco Musolesi, Ce- cilia Mascolo, Peter J Rentfrow, Chris Longworth, and Andrius Aucinas. Emotionsense: a mobile phones based adaptive platform for experimental social psychology re- search. In Proc. of UbiComp, pages 281-290. ACM, 2010.","title":"Emotionsense: a mobile phones based adaptive platform for experimental social psychology research","authors":"Mirco Kiran K Rachuri, Cecilia Musolesi, Mascolo, J Peter, Chris Rentfrow, Andrius Longworth, Aucinas","year":"2010","venue":"Proc. of UbiComp","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Raghavan et al., 2012] Kasturi Rangan Raghavan, Supriyo Chakraborty, Mani Srivastava, and Harris Teague. OVER- RIDE: A Mobile Privacy Framework for Context-Driven Perturbation and Synthesis of Sensor Data Streams. In In- ternational Workshop on Sensing Applications on Mobile Phones -PhoneSense, pages 1-5, New York, New York, USA, 2012. ACM Press.","title":"OVER-RIDE: A Mobile Privacy Framework for Context-Driven Perturbation and Synthesis of Sensor Data Streams","authors":"Supriyo Kasturi Rangan Raghavan, Mani Chakraborty, Harris Srivastava, Teague","year":"2012","venue":"International Workshop on Sensing Applications on Mobile Phones -PhoneSense","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Raij et al., 2011] Andrew Raij, Animikh Ghosh, Santosh Kumar, and Mani Srivastava. Privacy Risks Emerging From the Adoption of Innocuous Wearable Sensors in the Mobile Environment. In Annual conference on Human factors in computing systems -CHI '11, page 11, New York, New York, USA, 2011. ACM Press.","title":"Privacy Risks Emerging From the Adoption of Innocuous Wearable Sensors in the Mobile Environment","authors":"Andrew Raij, Animikh Ghosh, Santosh Kumar, Mani Srivastava","year":"2011","venue":"Annual conference on Human factors in computing systems -CHI '11","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Sagha et al., 2011] Hesam Sagha, Sundara Tejaswi Digu- marti, José del R Millán, Ricardo Chavarriaga, Alberto Ca- latroni, Daniel Roggen, and Gerhard Tröster. Benchmark- ing classification techniques using the opportunity human activity dataset. In Proc. of INSS, pages 36-40. IEEE, 2011.","title":"Benchmarking classification techniques using the opportunity human activity dataset","authors":"Hesam Sagha, Tejaswi Sundara, José Digumarti, Ricardo Del R Millán, Alberto Chavarriaga, Daniel Calatroni, Gerhard Roggen, Tröster","year":"2011","venue":"Proc. of INSS","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Schmidhuber, 1992] Jürgen Schmidhuber. Learning facto- rial codes by predictability minimization. Neural Compu- tation, 4(6):863-879, 1992.","title":"Learning factorial codes by predictability minimization","authors":"Jürgen Schmidhuber","year":"1992","venue":"Neural Computation","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Supriyo et al., 2012] Chakraborty Supriyo, Raghavan Kas- turi Rangan, Srivastava Mani B., and Bisdikian Catschik. Balancing Value and Risk in Information Sharing Through Obfuscation. In International Conference on Information Fusion (FUSION), page 2624. IEEE, 2012.","title":"Balancing Value and Risk in Information Sharing Through Obfuscation","authors":"Chakraborty Supriyo, Raghavan Kasturi Rangan, Srivastava Mani, B, Bisdikian Catschik","year":"2012","venue":"International Conference on Information Fusion (FUSION)","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Yang et al., 2015] Jian Bo Yang, Minh Nhut Nguyen, Phyo Phyo San, Xiao Li Li, and Shonali Krishnaswamy. Deep convolutional neural networks on multichannel time series for human activity recognition. In Proc. of IJCAI, pages 3995-4001, 2015.","title":"Deep convolutional neural networks on multichannel time series for human activity recognition","authors":"Bo Jian, Minh Nhut Yang, Phyo Nguyen, Xiao Li Phyo San, Shonali Li, Krishnaswamy","year":"2015","venue":"Proc. of IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2015/561"},{"position":20,"raw":"[Zhang and Sawchuk, 2012] Mi Zhang and Alexander A Sawchuk. Usc-had: a daily activity dataset for ubiquitous activity recognition using wearable sensors. In Proc. of UbiComp, pages 1036-1043. ACM, 2012.","title":"Usc-had: a daily activity dataset for ubiquitous activity recognition using wearable sensors","authors":"Mi Zhang, Alexander A Sawchuk","year":"2012","venue":"Proc. of UbiComp","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}