Home /

Research

Showing 409 - 414 / 897

Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learning


Authors:  ZhendaXie, YutongLin, ZhengZhang....
Published date-11/19/2020
Tasks:  ContrastiveLearning, ObjectDetection, RepresentationLearning, SemanticSegmentation

Abstract: Contrastive learning methods for unsupervised visual representation learning have reached remarkable levels of transfer performance. We argue that the power of contrastive learning has yet to be fully unleashed, as …

Finding the Homology of Decision Boundaries with Active Learning


Authors:  WeizhiLi, GautamDasarathy, KarthikeyanNatesanRamamurthy....
Published date-11/19/2020
Tasks:  ActiveLearning, Meta-Learning, ModelSelection, TopologicalDataAnalysis

Abstract: Accurately and efficiently characterizing the decision boundary of classifiers is important for problems related to model selection and meta-learning. Inspired by topological data analysis, the characterization of decision boundaries using …

FedEval: A Benchmark System with a Comprehensive Evaluation Model for Federated Learning


Authors:  DiChai, LeyeWang, KaiChen....
Published date-11/19/2020
Tasks:  FederatedLearning

Abstract: As an innovative solution for privacy-preserving machine learning (ML), federated learning (FL) is attracting much attention from research and industry areas. While new technologies proposed in the past few years …

Unmixing Convolutional Features for Crisp Edge Detection


Authors:  LinxiHuan, XianweiZheng, NanXue....
Published date-11/19/2020
Tasks:  EdgeDetection

Abstract: This paper presents a context-aware tracing strategy (CATS) for crisp edge detection with deep edge detectors, based on an observation that the localization ambiguity of deep edge detectors is mainly …

Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation


Authors:  XingeZhu, HuiZhou, TaiWang....
Published date-11/19/2020
Tasks:  3DSemanticSegmentation, PanopticSegmentation

Abstract: State-of-the-art methods for large-scale driving-scene LiDAR segmentation often project the point clouds to 2D space and then process them via 2D convolution. Although this corporation shows the competitiveness in the …

Node Similarity Preserving Graph Convolutional Networks


Authors:  WeiJin, TylerDerr, YiqiWang....
Published date-11/19/2020
Tasks:  GraphRepresentationLearning, RepresentationLearning, Self-SupervisedLearning

Abstract: Graph Neural Networks (GNNs) have achieved tremendous success in various real-world applications due to their strong ability in graph representation learning. GNNs explore the graph structure and node features by …

Filter by

Categories

Tags