Search for: All records

The integration of the Internet of Things (IoT) and modern Artificial Intelligence (AI) has given rise to a new paradigm known as the Artificial Intelligence of Things (AIoT). In this survey, we provide a systematic and comprehensive review of AIoT research. We examine AIoT literature related to sensing, computing, and networking & communication, which form the three key components of AIoT. In addition to advancements in these areas, we review domain-specific AIoT systems that are designed for various important application domains. We have also created an accompanying GitHub repository, where we compile the papers included in this survey: https://github.com/AIoT-MLSys-Lab/AIoT-Survey. This repository will be actively maintained and updated with new research as it becomes available. As both IoT and AI become increasingly critical to our society, we believe that AIoT is emerging as an essential research field at the intersection of IoT and modern AI. It is our hope that this survey will serve as a valuable resource for those engaged in AIoT research and act as a catalyst for future explorations to bridge gaps and drive advancements in this exciting field. 

Despite the increase in frequency and intensity of wildfires around the world, little research has examined households’ expectations of evacuation logistics and evacuation time estimate (ETE) components during such rapid-onset disasters. To address this gap, this study analyzes data from 152 household responses affected by the devastating 2018 wildfire in Mati, Greece where the second-deadliest wildfire of the 21st century took place. The questionnaire measured residents’ expectations of how they would respond to a future wildfire. This includes the number of vehicles they would take, their evacuation destination and route choices, and their expected evacuation preparation and travel times. Explanatory variables include risk perceptions, wildfire preparedness, wildfire experience, and demographic characteristics. The univariate results reveal some similarities to, but also some differences from, expected evacuation logistics and ETE components in other natural hazards. Moreover, correlation and regression analyses show that expected evacuation logistics and ETE components are primarily related to wildfire preparedness actions. Comparison of this study’s results with other rapid onset events such as tsunamis and hazardous material incidents, as well as longer onset events such as hurricanes, sheds light on household responses to wildfires. Emergency managers can use the similarities in results across studies to better prepare for wildfire evacuations. 

Binary neural network (BNN) delivers increased compute intensity and reduces memory/data requirements for computation. Scalable BNN enables inference in a limited time due to different constraints. This paper explores the application of Scalable BNN in oblivious inference, a service provided by a server to mistrusting clients. Using this service, a client can obtain the inference result on his/her data by a trained model held by the server without disclosing the data or learning the model parameters. Two contributions of this paper are: 1) we devise lightweight cryptographic protocols explicitly designed to exploit the unique characteristics of BNNs. 2) we present an advanced dynamic exploration of the runtime-accuracy tradeoff of scalable BNNs in a single-shot training process. While previous works trained multiple BNNs with different computational complexities (which is cumbersome due to the slow convergence of BNNs), we train a single BNN that can perform inference under various computational budgets. Compared to CryptFlow2, the state-of-the-art technique in the oblivious inference of non-binary DNNs, our approach reaches 3 × faster inference while keeping the same accuracy. Compared to XONN, the state-of-the-art technique in the oblivious inference of binary networks, we achieve 2 × to 12 × faster inference while obtaining higher accuracy. 

This study analyzes 488 household residents’ responses to the 2018 Indonesia M7.5 earthquake and tsunami. Comparison of this event with past earthquake and tsunami events, such as the 2009 Samoa (M8.1), 2011 Christchurch (M6.3), and 2011 Tohoku (M9.0) events, identifies commonalities and differences among people’s responses to these events. The results show that many Palu respondents failed to recognize strong earthquake ground motion as an environmental cue to a tsunami, but this was partially offset by an informal peer warning network. Most of the warnings only mentioned one of the six recommended message elements—the tsunami hazard. However, this brief message might have been adequate for many people if they could infer the certainty, severity, and immediacy of the threat, and appropriate evacuation modes, routes, and destinations. Unlike two comparison cases, some Palu respondents actually began their evacuation later than they expected the tsunami to strike. This might be due to spending too much time milling (seeking additional information, relaying warnings, reuniting families, and preparing to evacuate)—given the tsunami’s extremely rapid onset. This finding underscores the need for coastal emergency managers to promote evacuation preparedness for near-field tsunamis in which households pack Grab and Go kits in advance, warn others while evacuating, and plan in advance where to reunite household members who must evacuate separately.