Customer-obsessed science
Research areas
-
November 28, 20254 min readLarge language models are increasing the accuracy, reliability, and consistency of the product catalogue at scale.
-
November 20, 20254 min read
-
October 20, 20254 min read
-
October 14, 20257 min read
-
October 2, 20253 min read
Featured news
-
KDD 2022 Workshop on First Content Understanding and Generation for e-Commerce2022This work explores the usage of Fourier Transform in conjunction with Triplet loss applied on image styles, for reduction of the domain gap between the Source (e.g. Product Images in natural setting) and Target domain (e.g. Product Images on Ecommerce store pages) towards solving the Domain Adaptation problem. Most Unsupervised Domain Adaptation (UDA) algorithms reduce the domain gap between labelled Source
-
ECCV 20222022With the shift towards on-device deep learning, ensuring a consistent behavior of an AI service across diverse compute platforms becomes tremendously important. Our work tackles the emergent problem of reducing predictive inconsistencies arising as negative flips: test samples that are correctly predicted by a less accurate model, but incorrectly by a more accurate one. We introduce REGression constrained
-
ECCV 20222022We introduce Semi-supervised Performance Evaluation for Face Recognition (SPE-FR). SPE-FR is a statistical method for evaluating the performance and algorithmic bias of face verification systems when identity labels are unavailable or incomplete. The method is based on parametric Bayesian modeling of the face embedding similarity scores. SPE-FR produces point estimates, performance curves, and confidence
-
Rayleigh EigenDirections (REDs): Nonlinear GAN latent space traversals for multidimensional featuresECCV 20222022We present a method for finding paths in a deep generative model’s latent space that can maximally vary one set of image features while holding others constant. Crucially, unlike past traversal approaches, ours can manipulate arbitrary multidimensional features of an image such as facial identity and pixels within a specified region. Our method is principled and conceptually simple: optimal traversal directions
-
ECCV 20222022Fashion designs are rich in visual details associated with various visual attributes at both global and local levels. As a result, effective modeling and analyzing fashion requires fine-grained representations for individual attributes. In this work, we present a deep learning based online clustering method to jointly learn fine-grained fashion representations for all attributes at both instance and cluster
Collaborations
View allWhether you're a faculty member or student, there are number of ways you can engage with Amazon.
View all