-
Customers search for movie and series titles released across the world on streaming services like primevideo.com (PV), netflix.com (Netflix). In non-English speaking countries like India, Nepal and many others, the regional titles are transliterated from native language to English and are being searched in English. Given that there can be multiple transliterations possible for almost all the titles, searching
-
2023A locality-sensitive hash (or LSH) is a function that can efficiently map dataset points into a latent space while preserving pairwise distances. Such LSH functions have been used in approximate nearest-neighbor search (ANNS) in the following classic way, which we call classic hash clustering (CHC): first, the dataset points are hashed into a low-dimensional binary space using the LSH function; then, the
-
PAKDD 20232023Dense embedding-based semantic matching is widely used in e-commerce product search to address the shortcomings of lexical matching such as sensitivity to spelling variants. The recent advances in BERT-like language model encoders, have however, not found their way to realtime search due to the strict inference latency requirement imposed on e-commerce websites. While bi-encoder BERT architectures enable
-
2023Commercial search engines use different semantic models to augment lexical matches. These models provide candidate items for a user’s query from a target space of millions to billions of items. Models with different inductive biases provide relatively different predictions, making it desirable to launch multiple semantic models in production. However, latency and resource constraints make simultaneously
-
Online testing is indispensable in decision making for information retrieval systems. Interleaving emerges as an online testing method with orders of magnitude higher sensitivity than the pervading A/B testing. It merges the compared results into a single interleaved result to show to users, and attributes user actions back to the systems being tested. However, its pairwise design also brings practical
Related content
-
May 17, 2023The Amazon senior principal scientist coauthored a 2010 paper that introduced a new way to develop algorithms that make personalized recommendations for website users.
-
April 11, 2023The collaboration supports education, community outreach, and the application of academic research to video streaming and robotics.
-
March 21, 2023Tailoring neighborhood sizes and sampling probability to nodes’ degree of connectivity improves the utility of graph-neural-network embeddings by as much as 230%.
-
March 14, 2023Ren Zhang and her team tackle the interesting science challenges behind surfacing the most relevant offerings.
-
March 10, 2023Augmenting query-product graphs with hypergraphs describing product-product relationships improves recall score by more than 48%.
-
March 07, 2023Using reinforcement learning improves candidate selection and ranking for search, ad platforms, and recommender systems.
-
October 11, 2022Dual embeddings of each node, as both source and target, and a novel loss function enable 30% to 160% improvements over predecessors.
-
October 05, 2022Dataset that requires question-answering models to look up multiple facts and perform comparisons bridges a significant gap in the field.
-
September 20, 2022Adapting natural-language-processing techniques to recommendation systems and algorithmic fairness are two central topics at this year’s conference.
-
September 02, 2022Method would enable customers to evaluate supporting evidence for tip reliability.
-
August 25, 2022Launched under the auspices of the KDD Cup at KDD 2022, the competition included the release of a new product query dataset.
-
July 15, 2022New method optimizes the twin demands of retrieving relevant content and filtering out bad content.
-
July 08, 2022New model sets new standard in accuracy while enabling 60-fold speedups.
-
June 23, 2022Two-day RecSys workshop that extends the popular REVEAL to include CONSEQUENCES features Amazon organizers, speakers.
-
May 06, 2022Locality-sensitive hashing enables cache to hold more than three times as many query results.
-
April 21, 2022Amazon Scholar Eugene Agichtein on incorporating knowledge into natural-language-processing models, multimodal interactions, and more.
-
February 28, 2022Novel pretraining method enables increases of 5% to 14% on five different evaluation metrics.
-
February 18, 2022Method using hyperboloid embeddings improves on methods that use vector embeddings by up to 33%.
-
January 19, 2022The scientist's work is driving practical outcomes within an exploding machine learning research field.
-
January 14, 2022A new metric-learning loss function groups together superclasses and learns commonalities within them.
-
December 14, 2021Two NeurIPS papers examine the assignment of the same label to multiple categories, fast training of Transformer-based models.