Latest news
View all-
Glynis CondonJune 24, 2021As board members, Seokhwan Kim and Alexandros Papangelis will further SIGDIAL’s objectives on a variety of fronts. -
June 23, 2021A new method based on Transformers and trained with self-supervised learning achieves state-of-the-art performance.
-
June 22, 2021Scientists describe the use of privacy-preserving machine learning to address privacy challenges in XGBoost training and prediction.
-
-
June 20 - 25, 2021
-
June 19 - 25, 2021
-
July 11 - 15, 2021
-
-
Glynis CondonJune 21, 2021Özer’s paper published in INFORMS’ Management Science 2021 explores the dynamics behind “cheap-talk” communications.
-
DSci/AdobeJune 20, 2021Amazon Scholar works within Amazon’s Tübingen, Germany research center.
-
June 18, 2021Efficient learning and the capacity for abstraction are attributes that will probably require new insights — but self-supervised learning could help.
-
June 17, 2021Combining classic signal processing with deep learning makes method efficient enough to run on a phone.
Recent publications
View all-
ACL-IJCNLP 2021 Workshop on e-Commerce and NLP (ECNLP)2021Large-scale unsupervised abstractive summarization is sorely needed to automatically scan millions of customer reviews in today’s fast-paced e-commerce landscape. We address a key challenge in unsupervised abstractive summarization – reducing generic and uninformative content and producing useful information that relates to specific product aspects. To do so, we propose to model reviews in the context of
-
2021In deployed real-world spoken language understanding (SLU) applications, data continuously flows into the system. This leads to distributional differences between training and application data that can deteriorate model performance. While regularly retraining the deployed model with new data helps mitigating this problem, it implies significant computational and human costs. In this paper, we develop a
-
The Web Conference 2021 Workshop on Knowledge Management in E-Commerce2021Online stores often utilize product relationships such as bundles and substitutes to improve their catalog quality and guide customers through myriad choices. Entity resolution using pairwise product matching models offers a means of inferring relationships between products. In mature data repositories, the relationships may be mostly correct but require incremental improvements owing to errors in the original
-
2021Language modeling (LM) for automatic speech recognition (ASR) does not usually incorporate utterance level contextual information. For some domains like voice assistants, however, additional context, such as time at which an utterance was spoken, provides a rich input signal. We introduce an attention mechanism for training neural speech recognition language models on both text and nonlinguistic contextual
-
2021Speaker identification in the household scenario (e.g., for smart speakers) is typically based on only a few enrollment utterances but a much larger set of unlabeled data, suggesting semi-supervised learning to improve speaker profiles. We propose a graph-based semi-supervised learning approach for speaker identification in the household scenario, to leverage the unlabeled speech samples. In contrast to
Research areas
Hear from Amazon scientists, watch interviews and panels, and more.
Up next
On now
Meet our scientists
On now
Yoelle Maarek, Alexa Shopping vice president of research and science
On now
CVPR Panel - Computer vision: A look at the past, present, and future
On now
Women in Tech coffee talk at VLDB 2020 | Amazon Science
On now
Swami Sivasubramanian keynote outtake from Amazon AWS re:Invent 2020
Meet our scientists
Five scientists from across the company describe what they love about working at Amazon, and the type of customer-obsessed science and innovation they're responsible for in their role.
Working at Amazon
View all-
Courtesy of Daliana LiuMay 28, 2021Amazon's Daliana Liu helps others in the field chart their own paths.
-
Courtesy of Giovanni PaoliniMay 10, 2021Early on, Giovanni Paolini knew little about machine learning — now he’s leading new science on artificial intelligence that could inform AWS products.
Academic engagements
Learn more-
JHVEPhoto/Getty ImagesJune 08, 2021Proposal submissions for the third round of fairness in AI research are due August 3.
-
Credit: Glynis CondonMay 26, 2021Teams from three continents will compete to develop agents that assist customers in completing multi-step tasks.
Resources
-
-
Amazon
