Skip to content
Knowledge Amalgamation Engine
Python
Branch: master
Clone or download
Ssssseason
Ssssseason Update sbm
Latest commit afd0557 Oct 24, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
docs add vipa logo Jul 6, 2019
examples Update sbm Oct 24, 2019
kamal Update sbm Oct 24, 2019
.gitignore v0.1 Jul 4, 2019
README.md add paper authors Sep 25, 2019

README.md

icon

KAmalEngine

KAmalEngine is a software system that implements Knowledge Amalgamation algorithms. In knowledge amalgamation, we use trained deep network models available online to do a model-reusing task. Given multiple pretrained teacher networks, which specializes in different vision problems, the goal of knowledge amalgamation is to learn a lightweight student model capable of handling the comprehensive tasks, without human-labeled annotations. It is written in Python and powered by the Pytorch deep learning framework.

Table of contents

Introduction

The goal of KamalEngine is to provide a high-quality, high-performance code framework for Knowledge Amalgamation research. For more extensive discussions about Knowledge Amalgamation, please see the AAAI 2019 paper. This respository focuses on designed a flexible code framework in order to support rapid implementation and evaluation of novel research. KamalEngine includes implementations of the following Knowledge Amalgamation algorithms:

Algorithms

This repo provides some algorithms of KA.

Student Becoming the Master

Knowledge amalgamation for multiple teachers by feature projection.
Student Becoming the Master: Knowledge Amalgamation for Joint Scene Parsing, Depth Estimation, and More -- CVPR 2019
sbm-demo

Common Feature Learning

Extract common features from multiple teacher models.
Knowledge Amalgamation from Heterogeneous Networks by Common Feature Learning -- IJCAI 2019

Feature Space Common Space
cfl-feature-space cfl-feature-space

Amalgamating Knowledge towards Comprehensive Classification

Layer-wise amalgamation
Amalgamating Knowledge towards Comprehensive Classification -- AAAI 2019
layerwise-ka-framework

Customizing student networks from heterogeneous teachers via adaptive knowledge amalgamation (In Progress)

adaptive-ka-framework

Recombination

Build a new multi-task model by combining&pruning weight matrixs from distinct-task teachers. recombination-framework

Amalgamating Filtered Knowledge

Amalgamate knowledge from a pool of multi- or single-task teachers working on different tasks. The TargetNet after training, at a reasonably compact size, handles multiple customized tasks.

icon

Help

For more information, see docs and examples

Authors

Gongfan Fang fgfvain97@zju.edu.cn
Yixin Ji jiyixin@zju.edu.cn
Yanling Yin yanlingyin@zju.edu.cn
Jingwen Ye yejingwen@zju.edu.cn
Sihui Luo sihuiluo829@zju.edu.cn
Chengchao Shen chengchaoshen@zju.edu.cn
Mengqi Xue mqxue@zju.edu.cn

Homepage of VIPA Group, Zhejiang University, China

icon

Citation

@inproceedings{shen2019amalgamating,
  author={Shen, Chengchao and Wang, Xinchao and Song, Jie and Sun, Li and Song, Mingli},
  title={Amalgamating Knowledge towards Comprehensive Classification},
  booktitle={AAAI Conference on Artificial Intelligence (AAAI)},
  pages={3068--3075},
  year={2019}
}
@inproceedings{ye2019student,
  title={Student Becoming the Master: Knowledge Amalgamation for Joint Scene Parsing, Depth Estimation, and More},
  author={Ye, Jingwen and Ji, Yixin and Wang, Xinchao and Ou, Kairi and Tao, Dapeng and Song, Mingli},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={2829--2838},
  year={2019}
}
@inproceedings{luo2019knowledge,
  title={Knowledge Amalgamation from Heterogeneous Networks by Common Feature Learning},
  author={Luo, Sihui and Wang, Xinchao and Fang, Gongfan and Hu, Yao and Tao, Dapeng and Song, Mingli},
  booktitle={Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI)},
  year={2019},
}
@inproceedings{shen2019customizing,
  author={Shen, Chengchao and Xue, Mengqi and Wang, Xinchao and Song, Jie and Sun, Li and Song, Mingli},
  title={Customizing student networks from heterogeneous teachers via adaptive knowledge amalgamation},
  booktitle={The IEEE International Conference on Computer Vision (ICCV)},
  year={2019}
}
@inproceedings{Ye_Amalgamating_2019,
  year={2019},
  author={Ye, Jingwen and Wang, Xinchao and Ji, Yixin and Ou, Kairi and Song, Mingli},
  title={Amalgamating Filtered Knowledge: Learning Task-customized Student from Multi-task Teachers}
  booktitle={Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI)},
  year={2019},
}
You can’t perform that action at this time.