Latest news
View All-
June 16, 2020Black, a distinguished Amazon scholar, has been awarded the Longuet-Higgins Prize.
-
June 11, 2020Amazon’s Larry Davis on the past and future of computer vision research.
-
June 09, 2020New end-to-end approach to zero-shot video classification dramatically outperforms predecessors.
-
-
June 14 - 19, 2020
-
June 14 - 19, 2020
-
June 16 - 19, 2020
-
Recent publications
View All-
2020
One of the core components of voice assistants is the Natural Language Understanding (NLU) model. Its ability to accurately classify the user’s request (or “intent”) and recognize named entities in an utterance is pivotal to the success of these assistants. NLU models can be challenged in some languages by code-switching or morphological and orthographic variations. This work explores the possibility of -
2020
Channel pruning is one of the predominant approaches for accelerating deep neural networks. Most existing pruning methods either train from scratch with a sparsity inducing term such as group lasso, or prune redundant channels in a pretrained network and then fine tune the network. Both strategies suffer from some limitations: the use of group lasso is computationally expensive, difficult to converge and -
2020
Non-autoregressive (NAR) neural machine translation is usually done via knowledge distillation from an autoregressive (AR) model. Under this framework, we leverage large monolingual corpora to improve the NAR model’s performance, with the goal of transferring the AR model’s generalization ability while preventing overfitting. On top of a strong NAR baseline, our experimental results on the WMT14 En-De and -
2020
Enterprise companies use spatial data for decision optimization and gain new insights regarding the locality of their business and services. Industries rely on efficiently combining spatial and business data from different sources, such as data warehouses, geospatial information systems, transactional systems, and data lakes, where spatial data can be found in structured or unstructured form. In this demonstration -
2020
Most prior work on task-oriented dialogue systems are restricted to a limited coverage of domain APIs, while users oftentimes have domain related requests that are not covered by the APIs. In this paper, we propose to expand coverage of task-oriented dialogue systems by incorporating external unstructured knowledge sources. We define three sub-tasks: knowledge-seeking turn detection, knowledge selection
Research areas
Working at Amazon
Learn More-
June 05, 2020Ping Xu, forecasting science director within Amazon’s Supply Chain Optimization Technologies (SCOT) organization, talks about the importance of using science to forecast the future.
Work with us
See More JobsAcademic engagements
Learn More-
June 17, 2020Recipients represent 39 universities in 10 countries.
-
May 04, 2020Award-winning scientist joined Amazon team in Tübingen earlier this year.
Videos and webinars
View All-
June 10, 2020Tune in for a webcast discussion with Gerard Medioni, Amazon distinguished scientist, Pietro Perona, Amazon Fellow, and Larry Davis, Amazon senior principal scientist.
-
June 04, 2020Tune into Natarajan's LIVE interview on Thursday, June 11th
Resources
-
Discover more customer-obsessed science and innovation from around Amazon, including the AWS News Blog and the official Amazon Blog: Day One.
-
Check out the latest news and innovations from Amazon’s scientific community, which have been featured in the media around the world.

