Text classifiers are an indispensable tool for machine learning practitioners, but adapting them to new classes is expensive.To reduce the cost of new classes, previous work exploits class descriptions and/or labels from existing classes.However, these approaches leave a gap in the model development cycle as they support either zero-or few-shot learning but not both.Existing classifiers either do not work on zeroshot problems, or fail to improve much with few-shot labels.Further, prior work is aimed at concise class descriptions, which may be insufficient for complex classes.We overcome these shortcomings by casting text classification as a matching problem, where a model matches examples with relevant class descriptions.This formulation lets us leverage labels and complex class descriptions to perform zero-and fewshot learning on new classes.We compare this approach with numerous baselines on text classification tasks with complex class descriptions and find that it achieves strong zero-shot performance and scales well with few-shot samples, beating strong baselines by 22.48% (average precision) in the 10-shot setting.Furthermore, we extend the popular Model-Agnostic Meta-Learning algorithm to the zero-shot matching setting and show it improves zero-shot performance by 4.29%.Our results show that expressing text classification as a matching problem is a cost-effective way to address new classes.This strategy enables zero-shot learning for cold-start scenarios and few-shot learning so the model can improve until it is capable enough to deploy.

Management of daily stress can be greatly improved by delivering sensor-triggered just-in-time interventions (JITIs) on mobile devices. The success of such JITIs critically depends on being able to mine the time series of noisy sensor data to find the most opportune moments. In this paper, we propose a time series pattern mining method to detect significant stress episodes in a time series of discontinuous and rapidly varying stress data. We apply our model to 4 weeks of physiological, GPS, and activity data collected from 38 users in their natural environment to discover patterns of stress in real-life. We find that the duration of a prior stress episode predicts the duration of the next stress episode and stress in mornings and evenings is lower than during the day. We then analyze the relationship between stress and objectively rated disorder in the surrounding neighborhood and develop a model to predict stressful episodes.

Craving usually precedes a lapse for impulsive behaviors such as overeating, drinking, smoking, and drug use. Passive estimation of craving from sensor data in the natural environment can be used to assist users in coping with craving. In this paper, we take the first steps towards developing a computational model to estimate cigarette craving (during smoking abstinence) at the minute-level using mobile sensor data. We use 2,012 hours of sensor data and 1,812 craving self-reports from 61 participants in a smoking cessation study. To estimate craving, we first obtain a continuous measure of stress from sensor data. We find that during hours of day when craving is high, stress associated with self-reported high craving is greater than stress associated with low craving. We use this and other insights to develop feature functions, and encode them as pattern detectors in a Conditional Random Field (CRF) based model to infer craving probabilities.

Recent advances in mobile health have produced several new models for inferring stress from wearable sensors. But, the lack of a gold standard is a major hurdle in making clinical use of continuous stress measurements derived from wearable sensors. In this paper, we present a stress model (called cStress) that has been carefully developed with attention to every step of computational modeling including data collection, screening, cleaning, filtering, feature computation, normalization, and model training. More importantly, cStress was trained using data collected from a rigorous lab study with 21 participants and validated on two independently collected data sets --- in a lab study on 26 participants and in a week-long field study with 20 participants. In testing, the model obtains a recall of 89% and a false positive rate of 5% on lab data. On field data, the model is able to predict each instantaneous self-report with an accuracy of 72%.

Driving is known to be a daily stressor. Measurement of driver's stress in real-time can enable better stress management by increasing self-awareness. Recent advances in sensing technology has made it feasible to continuously assess driver's stress in real-time, but it requires equipping the driver with these sensors and/or instrumenting the car. In this paper, we present "GStress", a model to estimate driver's stress using only smartphone GPS traces. The GStress model is developed and evaluated from data collected in a mobile health user study where 10 participants wore physiological sensors for 7 days ( for an average of 10.45 hours/day) in their natural environment. Each participant engaged in 10 or more driving episodes, resulting in a total of 37 hours of driving data. We find that major driving events such as stops, turns, and braking increase stress of the driver. We quantify their impact on stress and thus construct our GStress model by training a Generalized Linear Mixed Model (GLMM) on our data. We evaluate the applicability of GStress in predicting stress from GPS traces, and obtain a correlation of 0.72. By obviating any burden on the driver or the car, we believe, GStress can make driver's stress assessment ubiquitous.

Smoking has been conclusively proved to be the leading cause of mortality that accounts for one in five deaths in the United States. Extensive research is conducted on developing effective smoking cessation programs. Most smoking cessation programs achieve low success rate because they are unable to intervene at the right moment. Identification of high-risk situations that may lead an abstinent smoker to relapse involve discovering the associations among various contexts that precede a smoking session or a smoking lapse. In the absence of an automated method, detection of smoking events still relies on subject self-report that is prone to failure to report and involves subject burden. Automated detection of smoking events in the natural environment can revolutionize smoking research and lead to effective intervention.

We present a classification algorithm built on our adaptation of the Generalized Lotka-Volterra model, well-known in mathematical ecology. The training algorithm itself consists only of computing several scalars, per each training vector, using a single global user parameter and then solving a linear system of equations. Construction of the system matrix is driven by our model and based on kernel functions. The model allows an interesting point of view of kernels' role in the inductive learning process. We describe the model through axiomatic postulates. Finally, we present the results of the preliminary validation experiments.