排列(音乐)
基数(数据建模)
深度学习
人工智能
不可见的
计算机科学
人工神经网络
集合(抽象数据类型)
对象(语法)
过程(计算)
深层神经网络
模式识别(心理学)
算法
数学
数据挖掘
计量经济学
物理
操作系统
程序设计语言
声学
作者
S. Hamid Rezatofighi,Roman Kaskman,Farbod Motlagh,Qinfeng Shi,Daniel Cremers,Laura Leal-Taixé,Ian Reid
出处
期刊:Cornell University - arXiv
日期:2018-05-02
被引量:20
摘要
Many real-world problems, e.g. object detection, have outputs that are naturally expressed as sets of entities. This creates a challenge for traditional deep neural networks which naturally deal with structured outputs such as vectors, matrices or tensors. We present a novel approach for learning to predict sets with unknown permutation and cardinality using deep neural networks. Specifically, in our formulation we incorporate the permutation as unobservable variable and estimate its distribution during the learning process using alternating optimization. We demonstrate the validity of this new formulation on two relevant vision problems: object detection, for which our formulation outperforms state-of-the-art detectors such as Faster R-CNN and YOLO, and a complex CAPTCHA test, where we observe that, surprisingly, our set based network acquired the ability of mimicking arithmetics without any rules being coded.
科研通智能强力驱动
Strongly Powered by AbleSci AI