Amazon Science homepage
Learn about Amazon's approach to customer-obsessed science. Amazon Science provides the latest news, research and innovation in artificial intelligence (AI) and machine learning (ML), including jobs, publications, conferences and events.
Search in www.amazon.science
Domain Posts www.amazon.science
“Calibeating”: Beating forecasters at their own game - Amazon Science
In order to identify expertise, forecasters should not be tested by their calibration score, which can always be made arbitrarily small, but rather by their Brier score. The Brier score is the sum of the calibration score and the refinement score; the latter measures how good the sorting into bins…
Language-informed transfer learning for embodied household activities - Amazon Science
For service robots to become general-purpose in everyday household environments, they need not only a large library of primitive skills, but also the ability to quickly learn novel tasks specified by users. Fine-tuning neural networks on a variety of downstream tasks has been successful in many…
Gaurav S. Sukhatme - Amazon Science
Gaurav S. Sukhatme is the Fletcher Jones Foundation Endowed Chair in Computer Science and a professor of computer science and electrical and computer engineering at the University of Southern California.
Domain data trumps teacher knowledge for distilling NLU models - Amazon Science
On natural-language-understanding tasks, student models trained only on task-specific data outperform those trained on a mix that includes generic data.
Improve explainability of ML models to meet regulatory requirements - Amazon Science
How an AWS customer uses Lookout for Vision to build custom computer vision models to automate quality inspection and detect defects.
Popular deep-learning book from Amazon authors gets update - Amazon Science
Google JAX Python library implementation and new topics added; volume 1 of book to be published by Cambridge University Press.
Image quality assessment using semi-supervised representation learning - Amazon Science
In this paper, we propose a framework for learning feature representations for Image Quality Assessment (IQA) using contrastive learning. To account for the absence of large-scale IQA dataset, we pretrain an image encoder to cluster images based on the image quality using synthetically distorted…
An automated and scalable ML solution for mapping invasive species: The case of the Australian tree fern in Hawaiian forests - Amazon Science
Biodiversity loss and ecosystem degradation are global challenges demanding creative and scalable solutions. Recent increases in data collection coupled with machine learning have the potential to expand landscape monitoring capabilities. We present a computer vision solution to the problem of…
High precision sound event detection based on transfer learning using transposed convolutions and feature pyramid network - Amazon Science
We introduce two models for high precision sound event detection leveraging transfer learning. The sound events we detect include “speech”, “music”, and “chime”. Both models consist of a CNN backbone pre-trained using AudioSet for audio classification. To get high precision detection results, the…
Shifting left for early detection of machine-learning bugs - Amazon Science
Computational notebooks are widely used for machine learning (ML). However, notebooks raise new correctness concerns beyond those found in traditional programming environments. ML library APIs are easy to misuse, and the notebook execution model raises entirely new problems concerning…