Search for: All records

India is largely devoid of high‐quality and reliable on‐the‐ground measurements of fine particulate matter (PM_2.5). Ground‐level PM_2.5concentrations are estimated from publicly available satellite Aerosol Optical Depth (AOD) products combined with other information. Prior research has largely overlooked the possibility of gaining additional accuracy and insights into the sources of PM using satellite retrievals of tropospheric trace gas columns. We evaluate the information content of tropospheric trace gas columns for PM_2.5estimates over India within a modeling testbed using an Automated Machine Learning (AutoML) approach, which selects from a menu of different machine learning tools based on the data set. We then quantify the relative information content of tropospheric trace gas columns, AOD, meteorological fields, and emissions for estimating PM_2.5over four Indian sub‐regions on daily and monthly time scales. Our findings suggest that, regardless of the specific machine learning model assumptions, incorporating trace gas modeled columns improves PM_2.5estimates. We use the ranking scores produced from the AutoML algorithm and Spearman’s rank correlation to infer or link the possible relative importance of primary versus secondary sources of PM_2.5as a first step toward estimating particle composition. Our comparison of AutoML‐derived models to selected baseline machine learning models demonstrates that AutoML is at least as good as user‐chosen models. The idealized pseudo‐observations (chemical‐transport model simulations) used in this work lay the groundwork for applying satellite retrievals of tropospheric trace gases to estimate fine particle concentrations in India and serve to illustrate the promise of AutoML applications in atmospheric and environmental research.

 

Collaborative bandit learning, i.e., bandit algorithms that utilize collaborative filtering techniques to improve sample efficiency in online interactive recommendation, has attracted much research attention as it enjoys the best of both worlds. However, all existing collaborative bandit learning solutions impose a stationary assumption about the environment, i.e., both user preferences and the dependency among users are assumed static over time. Unfortunately, this assumption hardly holds in practice due to users' ever-changing interests and dependency relations, which inevitably costs a recommender system sub-optimal performance in practice. In this work, we develop a collaborative dynamic bandit solution to handle a changing environment for recommendation. We explicitly model the underlying changes in both user preferences and their dependency relation as a stochastic process. Individual user's preference is modeled by a mixture of globally shared contextual bandit models with a Dirichlet process prior. Collaboration among users is thus achieved via Bayesian inference over the global bandit models. To balance exploitation and exploration during the interactions, Thompson sampling is used for both model selection and arm selection. Our solution is proved to maintain a standard $\tilde O(\sqrt{T})$ Bayesian regret in this challenging environment. Extensive empirical evaluations on both synthetic and real-world datasets further confirmed the necessity of modeling a changing environment and our algorithm's practical advantages against several state-of-the-art online learning solutions. 

Non-stationary bandits and clustered bandits lift the restrictive assumptions in contextual bandits and provide solutions to many important real-world scenarios. Though they have been studied independently so far, we point out the essence in solving these two problems overlaps considerably. In this work, we connect these two strands of bandit research under the notion of test of homogeneity, which seamlessly addresses change detection for non-stationary bandit and cluster identification for clustered bandit in a unified solution framework. Rigorous regret analysis and extensive empirical evaluations demonstrate the value of our proposed solution, especially its flexibility in handling various environment assumptions, e.g., a clustered non-stationary environment. 

Multi-armed bandit algorithms have become a reference solution for handling the explore/exploit dilemma in recommender systems, and many other important real-world problems, such as display advertisement. However, such algorithms usually assume a stationary reward distribution, which hardly holds in practice as users' preferences are dynamic. This inevitably costs a recommender system consistent suboptimal performance. In this paper, we consider the situation where the underlying distribution of reward remains unchanged over (possibly short) epochs and shifts at unknown time instants. In accordance, we propose a contextual bandit algorithm that detects possible changes of environment based on its reward estimation confidence and updates its arm selection strategy respectively. Rigorous upper regret bound analysis of the proposed algorithm demonstrates its learning effectiveness in such a non-trivial environment. Extensive empirical evaluations on both synthetic and real-world datasets for recommendation confirm its practical utility in a changing environment. 

In this work, we propose to improve long-term user engagement in a recommender system from the perspective of sequential decision optimization, where users' click and return behaviors are directly modeled for online optimization. A bandit-based solution is formulated to balance three competing factors during online learning, including exploitation for immediate click, exploitation for expected future clicks, and exploration of unknowns for model estimation. We rigorously prove that with a high probability our proposed solution achieves a sublinear upper regret bound in maximizing cumulative clicks from a population of users in a given period of time, while a linear regret is inevitable if a user's temporal return behavior is not considered when making the recommendations. Extensive experimentation on both simulations and a large-scale real-world dataset collected from Yahoo frontpage news recommendation log verified the effectiveness and significant improvement of our proposed algorithm compared with several state-of-the-art online learning baselines for recommendation.