...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Knowledge-Based Systems >Regularizing deep networks with prior knowledge: A constraint-based approach
【24h】

Regularizing deep networks with prior knowledge: A constraint-based approach

机译：将深度网络与先前知识进行规范：基于约束的方法

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Deep Learning architectures can develop feature representations and classification models in an integrated way during training. This joint learning process requires large networks with many parameters, and it is successful when a large amount of training data is available. Instead of making the learner develop its entire understanding of the world from scratch from the input examples, the injection of prior knowledge into the learner seems to be a principled way to reduce the amount of require training data, as the learner does not need to induce the rules from the data. This paper presents a general framework to integrate arbitrary prior knowledge into learning. The domain knowledge is provided as a collection of first-order logic (FOL) clauses, where each task to be learned corresponds to a predicate in the knowledge base. The logic statements are translated into a set of differentiable constraints, which can be integrated into the learning process to distill the knowledge into the network, or used during inference to enforce the consistency of the predictions with the prior knowledge. The experimental results have been carried out on multiple image datasets and show that the integration of the prior knowledge boosts the accuracy of several state-of-the-art deep architectures on image classification tasks. (C) 2021 The Authors. Published by Elsevier B.V.
机译：深度学习架构可以在培训期间以综合方式开发特征表示和分类模型。此联合学习过程需要具有许多参数的大型网络，并且当大量培训数据可用时，它是成功的。从输入示例的划痕中，将学习者培养其对世界的全部了解，而是将先验知识注入学习者似乎是减少需要培训数据量的原则方式，因为学习者不需要诱导来自数据的规则。本文介绍了一般框架，可以将任意事先知识整合到学习中。域知识被提供为一阶逻辑（FOL）条款的集合，其中要学习的每个任务对应于知识库中的谓词。逻辑语句被翻译成一组可差的约束，可以集成到学习过程中以将知识蒸馏到网络，或者在推理期间使用以强制使用先前知识来实施预测的一致性。实验结果已经在多个图像数据集上进行，并表明了先前知识的集成促进了在图像分类任务上进行了几种最先进的深度架构的准确性。 （c）2021作者。 elsevier b.v出版。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号