Amazon Developer Blogs

Amazon Developer Blogs

Showing posts tagged with Alexa science

October 17, 2019

Christos Christodoulopoulos

The open challenge for the Fact Extraction and Verification (FEVER) workshop at EMNLP involved devising adversarial examples that would stump fact verification systems trained on the FEVER data set.

[Read More]

October 11, 2019

Janet Slifka

Synthetic-data generators provided initial training data for natural-language-understanding models in Hindi, U.S. Spanish, and Brazilian Portuguese.

[Read More]

October 01, 2019

Zaid Ahmed

Recorded in the lab during simulated dinner parties, a new data set should aid the development of systems for separating speech signals in reverberant rooms with multiple speakers.

[Read More]

September 23, 2019

Adrian de Wynter

Theoretical analysis shows how to efficiently search a large space of possible neural architectures, to identify the one that offers optimal performance.

[Read More]

September 17, 2019

Dilek Hakkani-Tur

Data set includes more than 230,000 dialogue turns, most of which are annotated to indicate the sources of their factual assertions.

[Read More]

September 16, 2019

Shuyang Gao

Treating a conversation as a text, and dialogue state tracking as answering questions about the text, enables an 11.75% improvement in accuracy over the best-performing prior system.

[Read More]

September 10, 2019

Larry Hardesty

Research spans the five core areas of Alexa functionality, as well as more-general questions in machine learning.

[Read More]

September 05, 2019

Pranav Ladkat

By combining two state-of-the-art techniques for parallelizing machine learning — one that prioritizes accuracy, one that prioritizes efficiency — Alexa researchers improve on both.

[Read More]

September 04, 2019

Dilek Hakkani-Tur

Universal dialogue-act tagging scheme, hybrid slot-tracking system promise to improve dialogue state tracking.

[Read More]

August 29, 2019

Anirudh Raju

Techniques include weighting training samples from out-of-domain data sets and noise contrastive estimation, which turns the calculation of massive probability distributions into simple binary decisions.

[Read More]