https://github.com/adamdad/knowledgefactor

[ECCV2022] Factorizing Knowledge in Neural Networks
https://github.com/adamdad/knowledgefactor

deep-learning distillation eccv2022 kd knowldge-distillation knowledge-transfer multitask-learning pytorch

Last synced: about 19 hours ago
JSON representation

[ECCV2022] Factorizing Knowledge in Neural Networks

• Host: GitHub
• URL: https://github.com/adamdad/knowledgefactor
• Owner: Adamdad
• License: apache-2.0
• Created: 2022-06-27T06:58:04.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2022-09-12T15:42:10.000Z (about 2 years ago)
• Last Synced: 2024-05-14T00:05:32.182Z (6 months ago)
• Topics: deep-learning, distillation, eccv2022, kd, knowldge-distillation, knowledge-transfer, multitask-learning, pytorch
• Language: Python
• Homepage:
• Size: 781 KB
• Stars: 79
• Watchers: 2
• Forks: 5
• Open Issues: 2
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Factorizing Knowledge in  Neural Networks

This is a PyTorch implementation of the paper

[**Factorizing Knowledge in  Neural Networks(ECCV 2022)**](https://arxiv.org/abs/2207.03337)

[Supplementary Material](https://drive.google.com/file/d/1BkXgJHJ-64NAADZOLHZAWdgK0VcLoZXQ/view?usp=sharing)

*Xingyi Yang, Jingwen Ye, Xinchao Wang* 

![kf](assets/KnowledgeFactior_full-1.png)

In this paper, we explore a novel and ambitious knowledge-transfer task, termed Knowledge Factorization~(KF). The core idea of KF lies in the modularization 

and assemblability of knowledge: given a pretrained network model as input, KF aims to decompose it into several factor networks, each of which handles only a dedicated task and maintains task-specific knowledge factorized from the source network.

## Licence

This project is released under the [Apache 2.0 license](LICENCE).

## Citation

If you find this project useful in your research, please consider cite:

        @Article{yang2022knowledgefactor,

        author  = {Xingyi Yang, Jingwen Ye, Xinchao Wang},

        title   = {Factorizing Knowledge in  Neural Networks},

        journal = {European Conference on Computer Vision},

        year    = {2022},

        }