Context-awareness is the ability of software systems to sense and adapt their physical environment. Many contemporary mobile applications changing locations, connectivity states, available computational energy resources, proximity other users devices. Nevertheless, there little systematic support for context-awareness in operating systems. Because this, application developers must build own adaptation engines, dealing directly with sensors polluting code complex decisions. In this paper, we...

Reinforcement learning (RL) presents numerous benefits compared to rule-based approaches in various applications. Privacy concerns have grown with the widespread use of RL trained privacy-sensitive data IoT devices, especially for human-in-the-loop systems. On one hand, methods enhance user experience by trying adapt highly dynamic nature humans. other policies can leak user's private information. Recent attention has been drawn designing privacy-aware algorithms while maintaining an...

Modern social media platforms, such as TikTok, Facebook, and YouTube, rely on recommendation systems to personalize content for users based user interactions with endless streams of content, "For You" pages. However, these complex algorithms can inadvertently deliver problematic related self-harm, mental health, eating disorders. We introduce AutoLike, a framework audit in platforms topics interest their sentiments. To automate the process, we formulate problem reinforcement learning...

Thanks to the rapid growth in wearable technologies, monitoring complex human context becomes feasible, paving way develop human-in-the-loop IoT systems that naturally evolve adapt and environment state autonomously. Nevertheless, a central challenge designing such personalized applications arises from variability. Such variability stems fact different humans exhibit behaviors when interacting with (intra-human variability), same may change behavior over time application (inter-human be...

Modern integrated circuits, fabricated in nanometer technologies, suffer from significant power/performance variation across-chip, chip-to-chip and over time due to aging ambient fluctuations. Furthermore, several existing emerging reliability loss mechanisms have caused increased transient, intermittent permanent failure rates. While this variability has been typically addressed by process, device circuit designers, there a recent push towards sensing adapting the various layers of...

Thanks to the adoption of more sensors in automotive industry, context-aware Advanced Driver Assistance Systems (ADAS) become possible. On one side, a common thread ADAS applications is focus entirely on context vehicle and its surrounding vehicles leaving human (driver) out consideration. other due increasing sensing capabilities mobile phones wearable technologies, monitoring complex becomes feasible which paves way develop driver-in-the-loop that provide personalized driving experience....

Modern integrated circuits, fabricated in nanometer technologies, suffer from significant power/performance variation across-chip, chip-to-chip and over time due to aging ambient fluctuations. Furthermore, several existing emerging reliability loss mechanisms have caused increased transient, intermittent permanent failure rates. While this variability has been typically addressed by process, device circuit designers, there a recent push towards sensing adapting the various layers of...

This paper outlines the prevalent challenges for emerging wearable neurotechnology in modern IoT systems. We underline recent insights neuroscience and ability to decode brain circuitry with high confidence. address imminent of translating advanced high-cost medical setup neural activity recording a commodity system. Emphasis is placed on non-invasive sensing technology advances data analytic neurosignals. In particular, we focus human-in-the-loop systems where privacy concerns leaking...

Thanks to the rapid growth in wearable technologies and advancements machine learning, monitoring complex human contexts becomes feasible, paving way develop human-in-the-loop IoT systems that naturally evolve adapt environment state autonomously. Nevertheless, a central challenge designing many of these arises from requirement infer mental state, such as intention, stress, cognition load, or learning ability. While different can be inferred fusion sensor modalities correlate particular...

Personalized IoT adapt their behavior based on contextual information, such as user and location. Unfortunately, the fact that personalized to context opens a side-channel leaks private information about user. To end, we start by studying extent which malicious eavesdropper can monitor actions taken an system extract user's information. In particular, show two concrete instantiations (in of mobile phones smart homes) new category spyware refer Context-Aware Adaptation Based Spyware (SpyCon)....

Adblocking relies on filter lists, which are manually curated and maintained by a community of list authors. Filter curation is laborious process that does not scale well to large number sites or over time. In this paper, we introduce AutoFR, reinforcement learning framework fully automate the rule creation evaluation for interest. We design an algorithm based multi-arm bandits generate rules block ads while controlling trade-off between blocking avoiding visual breakage. test AutoFR...

Personalized IoT adapts their behavior based on contextual information, such as user and location. Unfortunately, the fact that personalized to context opens a side-channel leaks private information about user. To end, we start by studying extent which malicious eavesdropper can monitor actions taken an system extract users' information. In particular, show two concrete instantiations (in of mobile phones smart homes) new category spyware refer Context-Aware Adaptation Based Spyware...

Multiple sensory modalities are fast becoming a key instrument in the future of automotive industry. Collision avoidance, lane departure warning, and self-parking examples Advanced Driver Assistance Systems (ADAS) that possible with adoption more sensors. Moreover, thanks to recent advances mobile computing wearable devices, driver is now equipped advanced systems. This rich environment paves way integrate human factor into loop computation ADAS provide personalized experience. In this...

Interdisciplinary research among engineering, computer science, and neuroscience to understand utilize the human brain signals resulted in advances widespread applicability of wearable neurotechnology adaptive human-in-the-loop smart systems. Considering these advances, we envision that future education will exploit move toward more personalized classrooms where instructions interactions are tailored towards. students' individual strengths needs. In this paper, discuss how neuroscience,...

Adblocking relies on filter lists, which are manually curated and maintained by a community of list authors. Filter curation is laborious process that does not scale well to large number sites or over time. In this paper, we introduce AutoFR, reinforcement learning framework fully automate the rule creation evaluation for interest. We design an algorithm based multi-arm bandits generate rules block ads while controlling trade-off between blocking avoiding visual breakage. test AutoFR...

Automotive is becoming more and sensor-equipped. Collision avoidance, lane departure warning, self-parking are examples of applications possible with the adoption sensors in automotive industry. Moreover, driver now equipped sensory systems like wearables mobile phones. This rich environment real-time streaming contextual data from vehicle make human factor integral loop computation. By integrating human's behavior reaction into advanced driver-assistance (ADAS), vehicles become a...

Excerpted from "CAreDroid: Adaptation Framework for Android Context-Aware Applications," Proceedings of the 21st Annual International Conference on Mobile Computing and Networking. http://dl.acm.org/citation.cfm?id=2790108 © ACM 2015.

Achieving fairness in sequential-decision making systems within Human-in-the-Loop (HITL) environments is a critical concern, especially when multiple humans with different behavior and expectations are affected by the same adaptation decisions system. This human variability factor adds more complexity since policies deemed fair at one point time may become discriminatory over due to variations preferences resulting from inter- intra-human variability. paper addresses problem an equity lens,...

Virtual reality (VR) platforms enable a wide range of applications, however pose unique privacy risks. In particular, VR devices are equipped with rich set sensors that collect personal and sensitive information (e.g., body motion, eye gaze, hand joints, facial expression), which can be used to uniquely identify user, even without explicit identifiers. this paper, we interested in understanding the extent user identified based on data collected by different sensors. We consider adversaries...

Instances of casualties resulting from large crowds persist, highlighting the existing limitations current crowd management practices. One notable drawback is insufficient provision for disadvantaged individuals who may require additional time to evacuate due their slower running speed. Moreover, escape strategies fall short ensuring safety all during a surge. To address these pressing concerns, this paper proposes two methodologies. Firstly, we advocate implementation fair evacuation...