Home /

Research

Showing 151 - 156 / 897

Soft Contrastive Learning for Visual Localization


Authors:  JanineThoma, DandaPaniPaudel, LucV.Gool....
Published date-12/01/2020
Tasks:  ContrastiveLearning, ImageRetrieval, VisualLocalization

Abstract: Localization by image retrieval is inexpensive and scalable due to simple mapping and matching techniques. Such localization, however, depends upon the quality of image features often obtained using Contrastive learning …

Rethinking Learnable Tree Filter for Generic Feature Transform


Authors:  LinSong, YanweiLi, ZhengkaiJiang....
Published date-12/01/2020
Tasks:  InstanceSegmentation, ObjectDetection, SemanticSegmentation

Abstract: The Learnable Tree Filter presents a remarkable approach to model structure-preserving relations for semantic segmentation. Nevertheless, the intrinsic geometric constraint forces it to focus on the regions with close spatial …

Evaluating Attribution for Graph Neural Networks


Authors:  BenjaminSanchez-Lengeling, JenniferWei, BrianLee....
Published date-12/01/2020

Abstract: Interpretability of machine learning models is critical to scientific understanding, AI safety, as well as debugging. Attribution is one approach to interpretability, which highlights input dimensions that are influential to …

VIME: Extending the Success of Self- and Semi-supervised Learning to Tabular Domain


Authors:  JinsungYoon, YaoZhang, JamesJordon....
Published date-12/01/2020
Tasks:  DataAugmentation, Imputation, Self-SupervisedLearning

Abstract: Self- and semi-supervised learning frameworks have made significant progress in training machine learning models with limited labeled data in image and language domains. These methods heavily rely on the unique …

Learning to Adapt to Evolving Domains


Authors:  HongLiu, MingshengLong, JianminWang....
Published date-12/01/2020
Tasks:  DomainAdaptation, Meta-Learning, TransferLearning, UnsupervisedDomainAdaptation

Abstract: Domain adaptation aims at knowledge transfer from a labeled source domain to an unlabeled target domain. Current domain adaptation methods have made substantial advances in adapting discrete domains. However, this …

Mixed Information Flow for Cross-domain Sequential Recommendations


Authors:  MuyangMa, PengjieRen, ZhuminChen....
Published date-12/01/2020
Tasks:  TransferLearning

Abstract: Cross-domain sequential recommendation (CDSR) is the task of predict the next item that the user is most likely to interact with based on past sequential behavior from multiple domains. One …

Filter by

Categories

Tags