迁移学习

PrefaceChapter 1 Introduction 1．1 Background and Motivation 1．2 COntributiong 1．2．1 Extended MDLP for Transfer Learning 1．2．2 Compact Coding for Hyperplane Classifiers in Transfer Learning 1．2．3 Transfer Active Learning 1．2．4 Gaussian Process for Transfer Learning 1．3 Book OverviewChapter 2 Literature Review and Preliminaries for MDLP 2．1 Transfer Learning 2．2 Active Learning and Transfer Active Learning 2．3 Preljminaries for MD[．P Chapter 3 Extended MDL Principle for Feature-based Transfer Learning 3．1 IntroductiOn 3．2 Problem Statement 3．3 Preliminaries for Encoding 3．3．1 Theoretical Foundation of the EMDLP 3．3．2 Adaptation of the EMDLP to Our Problem 3．4 Supervised Inductive Transfer Learning Algorithm 3．4．1 EMDLP with Incremental Search 3．4．2 EMDLP with Hill Climbing 3．5 Experiments 3．5．1 Experimental Settings 3．5．2 Experimental Results on Synthetic Data Sets 3．5．3 Experimental Results on Real Data Sets 3．6 SummaryChapter 4 Compact Coding for Hyperplane Classifiers in a Heterogeneous Environment 4．1 Introduction 4．2 Problem Setting 4．3 Compact Coding for Hyperplane Classifiers in Heterogeneous Environment 4．3．1 Macro Level：Arrange Related Tasks 4．3．2 Micro Level Evaluation 4．3．3 The Transfer Learning Algorithm 4．4 Experiments 4．4．1 Experimental Setting 4．4．2 Experimental Results 4．5 SummaryChapter 5 Adaptive Transfer Learning with Query by Committee 5．1 IntroductiOn 5．2 Problem Setting and Preliminaries 5．3 Probabilistic Framework for ALTL 5．4 The ALTL Algorithm and Analysis 5．4．1 The Procedure of ALTL 5．4．2 Termination Condition and Analysis 5．5 Experiments 5．5．1 Experimental Setting 5．5．2 Results on Synthetic Data Sets 5．5．3 Results on Real Data Sets 5．6 SummaryChapter 6 Gaussian Process for Transfer Learning through Minimum Encoding 6．1 IntrOduction 6．2 Gaussian Process for Classification 6．3 The GPTL Algorithm 6．3．1 Arrange Related Tasks 6．3．2 The Instance Level Similarities 6．4 Experiments 6．5 SummaryChapter 7 Concluding CommentsAppendix A Target Concepts in Chapter 3Bibliography

《迁移学习--理论与实践》(作者邵浩)着重介绍了应用最为广泛的分类学习，将最前沿的研究进行了归纳总结，并通过实际算法分析，将领域内的最新进展提供给读者，使读者能够使用迁移学习的工具构建模型并应用到实际问题，本书适合从事相关研究工作的人员参考阅读。 《迁移学习--理论与实践》(作者邵浩)着眼于管理实际中的资源再利用，对数据挖掘领域最前沿的迁移学习进行了详细阐述，并着重介绍了应用最为广泛的分类学习，将最前沿的研究进行了归纳总结，并通过实际算法分析，将领域内的最新进展提供给读者，使读者能够使用迁移学习的工具构建模型并应用到实际问题。《迁移学习--理论与实践》主要读者对象为具有管理和计算机背景并在数据挖掘领域有初步研究的学者。【作者简介】 邵浩，上海对外经贸大学WTO研究教育学院讲师，日本国立九州大学工学博士，曾就读于中国科学技术大学管理学院硕博连读课程。研究方向为数据挖掘、管理科学与工程。