In this paper, we analyze a novel information-sharing policy to reduce stockouts and improve performance on online grocery retail platforms. Stockouts are particularly relevant to our setting, as online platforms lack control over retailers' inventory management, and consequently when/whether items are in stock. Our policy involves leveraging the digital interface of online platforms to disseminate information about delivery windows when retailer stock levels are at their peak. Ex-ante, it is unclear if such a policy would positively influence customers' purchasing patterns, particularly in terms of its impact on platforms' fundamentals. On the one hand, this information might not be informative to customers, having already been factored into their optimal delivery window decision. Alternatively, customers may be limited in their cognitive abilities and ignore stock information when choosing delivery windows. In addition, sharing information about high-stock delivery windows might raise customer awareness (or aversion) regarding stockouts, potentially decreasing the conversion rate. We conducted a large-scale field experiment on over 1M users on Instacart, an online grocery delivery platform. In this experiment, customers in the treatment group were prompted "Higher stock at this time" under delivery windows corresponding to peak supermarket stock levels. We find our cost-free information-sharing policy significantly affected customer behavior, as well as order fulfillment rates. Our main results can be summarized as follows:

Hyper-parameter optimization is crucial for pushing the accuracy of a deep learning model to its limits. However, a hyper-parameter optimization job, referred to as a study, involves numerous trials of training a model using different training knobs, and therefore is very computation-heavy, typically taking hours and days to finish. We observe that trials issued from hyper-parameter optimization algorithms often share common hyper-parameter sequence prefixes. Based on this observation, we propose Hippo, a hyper-parameter optimization system that reuses computation across trials to reduce the overall amount of computation significantly. Instead of treating each trial independently as in existing hyper-parameter optimization systems, Hippo breaks down the hyper-parameter sequences into stages and merges common stages to form a tree of stages (a stage tree). Hippo maintains an internal data structure, search plan, to manage the current status and history of a study, and employs a critical path based scheduler to minimize the overall study completion time. Hippo applies to not only single studies but multi-study scenarios as well. Evaluations show that Hippo's stage-based execution strategy outperforms trial-based methods for several models and hyper-parameter optimization algorithms, reducing end-to-end training time by up to 2.76X (3.53x) and GPU-hours by up to 4.81X (6.77x), for single (multiple) studies.