Mobile Crowd Sensing (MCS) is a promising paradigm for collecting large-scale data by using distributed sensors on smartphones and wearables. However, building comprehensive spatiotemporal fields of measurements with continuous sensing at all locations and time cycles is often impractical and costly. To address this, we leverage our proposed Learned Gumbel-Based Active Sparse MCS framework, utilizing an Encoder-Decoder Time Series Transformer that reconstructs the full spatiotemporal field for a given sensing cycle using the data collected by a small subset of available sensors. The framework incorporates a selector that actively employs the Gumbel distribution to dynamically pick the most effective sensors for the next sensing cycle. Rather than applying a single selector layer across all sensors, we introduce a patched forecasting approach to account for temporal patterns inherent in sensing cycles. Each patch forecasts its own subsequence of the time series while simultaneously selecting a subset of sensors to reuse in the next sensing cycle.

Mobile Crowd Sensing (MCS) has revolutionized the way we collect environmental and urban data by leveraging a tremendous number of distributed sensors equipped on wearable devices like smartphones, smart watches, and so on. Gathering continuous data from these sensors allows us to generate a complete spatiotemporal field for the desired measurements. However, obtaining continuous sensing data at all locations and times is often infeasible due to resource limits, cost constraints, and practicality. To address this problem, we proposed a Learned Gumbel-Based Active Sparse MCS framework utilizing an Encoder-Decoder Time Series Transformer that reconstructs the full spatiotemporal field for a given time cycle and a small subset of available sensors. Notably, an active sensor selection layer is included, which leverages the Gumbel distribution to create noise from learned weights, helping to dynamically select the most effective sensors for the next sensing cycle. We verified the effectiveness of our proposed model by comparing it to random selection, attention-scoring, and statistical methods. Our Gumbel-Based selector achieves higher performance in both firstand second-order norm metrics.

Calibrating blackbox machine learning models to achieve risk control is crucial to ensure reliable decision-making. A rich line of literature has been studying how to calibrate a model so that its predictions satisfy explicit finite-sample statistical guarantees under a _fixed_, _static_, and unknown data-generating distribution. However, prediction-supported decisions may influence the outcome they aim to predict, a phenomenon named _performativity_ of predictions, which is commonly seen in social science and economics. In this paper, we introduce _Performative Risk Control_, a framework to calibrate models to achieve risk control under performativity with provable theoretical guarantees. Specifically, we provide an iteratively refined calibration process, where we ensure the predictions are improved and risk-controlled throughout the process. We also study different types of risk measures and choices of tail bounds. Lastly, we demonstrate the effectiveness of our framework by numerical experiments on the task of predicting credit default risk. To the best of our knowledge, this work is the first one to study statistically rigorous risk control under performativity, which will serve as an important safeguard against a wide range of strategic manipulation in decision-making processes.

The lack of standardized benchmarks for reinforcement learning (RL) in sustainability applications has made it difficult to both track progress on specific domains and identify bottlenecks for researchers to focus their efforts. In this paper, we present SustainGym, a suite of five environments designed to test the performance of RL algorithms on realistic sustainable energy system tasks, ranging from electric vehicle charging to carbon-aware data center job scheduling. The environments test RL algorithms under realistic distribution shifts as well as in multi-agent settings. We show that standard off-the-shelf RL algorithms leave significant room for improving performance and highlight the challenges ahead for introducing RL to real-world sustainability tasks.

Conversational agents (CAs) are increasingly ubiquitous and are now commonly used to access medical information. However, we lack systematic data about the quality of advice such agents provide. This paper evaluates CA advice for mental health (MH) questions, a pressing issue given that we are undergoing a mental health crisis. Building on prior work, we define a new method to systematically evaluate mental health responses from CAs. We develop multi-utterance conversational probes derived from two widely used mental health diagnostic surveys, the PHQ-9 (Depression) and the GAD-7 (Anxiety). We evaluate the responses of two text-based chatbots and four voice assistants to determine whether CAs provide relevant responses and treatments. Evaluations were conducted both by clinicians and immersively by trained raters, yielding consistent results across all raters. Although advice and recommendations were generally low quality, they were better for Crisis probes and for probes concerning symptoms of Anxiety rather than Depression. Responses were slightly improved for text versus speech-based agents, and when CAs had access to extended dialogue context. Design implications include suggestions for improved responses through clarification sub-dialogues. Responses may also be improved by the incorporation of empathy although this needs to be combined with effective treatments or advice.

Many people struggle with their weight and are turning to online communities for social and informational support. The aim of this study is to understand the issues commonly discussed in online weight loss communities. Through observation and content analysis of threads in one specific weight loss community, we identified 17 distinct categories discussed by the participants. We detail four categories specifically: Personal Experience, Consumption Choices, Dieting Strategy, and Exercise. Our analysis describes some key user roles and states that often relate to different phases of a person's weight loss journey. We identify a set of transient states, which are not proper roles but are significant in an online community where individuals are attempting to change their own behaviors. We close with design suggestions for encouraging and maintaining participant engagement in an online health community

This paper presents a model for analyzing the performance of transmission strategies in a multihop packet radio network where each station has adjustable transmission radius. A larger transmission radius will increase the probability of finding a receiver in the desired direction and contribute bigger progress if the transmission is successful, but it also has a higher probability of collision with other transmissions. The converse is true for shorter transmission range. We illustrate our model by comparing three transmission strategies. Our results show that the network can achieve better performance by suitably controlling the transmission range. One of the transmission strategies, namely transmitting to the nearest forward neighbor by using adjustable transmission power, has desirable features in a high terminal density environment.