Customer-obsessed science
Research areas
-
November 28, 20254 min readLarge language models are increasing the accuracy, reliability, and consistency of the product catalogue at scale.
-
November 20, 20254 min read
-
October 20, 20254 min read
-
October 14, 20257 min read
-
October 2, 20253 min read
Featured news
-
Transportation Science2022The 2021 Amazon Last Mile Routing Research Challenge, hosted by Amazon.com’s Last Mile Research team, and scientifically supported by the Massachusetts Institute of Technology’s Center for Transportation and Logistics, prompted participants to leverage real operational data to find new and better ways to solve a real-world routing problem. In this article, we describe the data set released for the research
-
ColdGuess: A general and effective relational graph convolutional network to tackle cold start casesKDD 2022 Workshop on Mining and Learning with Graphs2022Low-quality listings and bad actor behavior in online retail websites threatens e-commerce business as these result in sub-optimal buying experience and erode customer trust. When a new listing is created, how to tell it has good quality? Is the method effective, fast, and scalable? Previous approaches often face three limitations/challenges: (1) unable to handle cold start problems where new sellers/listings
-
RecSys 2022 Workshop on CONSEQUENCES – Causality, Counterfactuals and Sequential Decision-Making2022Adaptive experimental design methods are increasingly being used in industry as a tool to boost testing throughput or reduce experimentation cost relative to traditional A/B/N testing methods. This paper shares lessons learned regarding the challenges and pitfalls of naively using adaptive experimentation systems in industrial settings where non-stationarity is prevalent, while also providing perspectives
-
RecSys 2022 Workshop on CONSEQUENCES – Causality, Counterfactuals and Sequential Decision-Making2022We propose diagnostics, based on control variates, to detect data quality issues in logged bandit feedback data, which is of critical importance for accurate offline evaluation and training of recommendation policies. Our diagnostics can provably detect two common types of data issues: (1) when the policy that logged the data was insufficiently randomized; (2) when the logged propensity values are incorrect
-
CONSEQUENCES+REVEAL 20222022A critical need for industrial recommender systems is the ability to evaluate recommendation policies offline, before deploying them to production. Unfortunately, widely used off-policy evaluation methods either make strong assumptions about how users behave that can lead to excessive bias, or they make fewer assumptions and suffer from large variance. We tackle this problem by developing a new estimator
Collaborations
View allWhether you're a faculty member or student, there are number of ways you can engage with Amazon.
View all