Eeeros
/
beimingwu
forked from beimingwu/beimingwu
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

perf(docs): polish contents in README

This commit is contained in:
Gene 2024-01-25 01:05:40 +08:00
parent a320b582b6
commit b003ed1059
2 changed files with 12 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
README.md
View File

@ -27,7 +27,7 @@
<div>
<h3 align="center">Beimingwu: The First Learnware Dock System</h3>
<h3 align="center">A Research Platform for Learnware Paradigm — Systematic Implementation</h3>
<h3 align="center">A Research Platform for Learnware — Systematic Implementation</h3>
<h3 align="center">
<a href="README_zh.md">中文</a> |
<b>English</b>
@ -36,16 +36,18 @@
# Introduction
The _learnware_ paradigm was proposed by Professor Zhi-Hua Zhou in 2016 [1, 2]. In this paradigm, developers worldwide can share models with the _learnware dock system_, which effectively searches for and reuse learnware(s) to help users solve machine learning tasks efficiently without starting from scratch.
_Learnware_ was proposed by Professor Zhi-Hua Zhou in 2016 [1, 2]. In the _learnware paradigm_, developers worldwide can share models with the _learnware dock system_, which effectively searches for and reuse learnware(s) to help users solve machine learning tasks efficiently without starting from scratch.
Beimingwu is the first systematic implementation of the _learnware_ paradigm, offering a solid research platform for learnware-related studies. A learnware is a well-performed trained model with a specification that describes its capabilities, enabling it to be readily identified and reused in the future based on user requirements. The specification includes a semantic specification in text and a statistical specification sketching the model's statistical information.
Beimingwu is the first systematic open-source implementation of learnware dock system, providing a preliminary research platform for learnware studies. Developers worldwide can submit their models freely to the learnware dock. They can generate specifications for the model with the help of Beimingwu without disclosing their raw data, and then the model and specification can be assembled into a learnware, which will be accommodated in the learnware dock. Future users can solve their tasks by submitting their requirements and reusing helpful learnwares returned by Beimingwu, while also not disclosing their own data. It is anticipated that after Beimingwu accumulates millions of learnwares, an "emergent" behavior may occur: machine learning tasks that have never been specifically tackled may be solved by assembling and reusing some existing learnwares.
A learnware is a well-performed trained model with a specification that describes its capabilities, enabling it to be readily identified and reused in the future based on user requirements. The specification includes a semantic specification in text and a statistical specification sketching the model's statistical information.
[1] Zhi-Hua Zhou. Learnware: on the future of machine learning. _Frontiers of Computer Science_, 2016, 10(4): 589–590 <br/>
[2] Zhi-Hua Zhou. Machine Learning: Development and Future. _Communications of CCF_, 2017, vol.13, no.1 (2016 CNCC keynote)
## What features does Beimingwu have?
As shown in the diagram below, the Beimingwu learnware dock system, serving as a research platform for the learnware paradigm, systematically implements the core processes of the paradigm for the first time:
As shown in the diagram below, the Beimingwu learnware dock system, serving as a preliminary research platform for learnware, systematically implements the core processes of the learnware paradigm for the first time:
- **Submitting Stage**: The system includes multiple detection mechanisms to ensure the quality of uploaded learnwares. Additionally, the system trains a heterogeneous engine based on existing learnware specifications in the system to merge different specification islands and assign new specifications to learnwares. With more learnwares are submitted, the heterogeneous engine will continue to update, achieving continuous iteration of learnware specifications and building a more precise specification world.
- **Deploying Stage**: After users upload task requirements, the system automatically selects whether to recommend a single learnware or multiple learnware combinations and provides efficient deployment methods. Whether it's a single learnware or a combination of multiple learnwares, the system offers convenient learnware reuse tools.

10
README_zh.md
View File

@ -27,7 +27,7 @@
<div>
<h3 align="center">北冥坞：学件基座系统</h3>
<h3 align="center">学件范式科研平台 — 系统性实现</h3>
<h3 align="center">学件科研平台 — 系统性实现</h3>
<h3 align="center">
<b>中文</b> |
<a href="README.md">English</a>
@ -36,16 +36,18 @@
# 简介
学件范式由周志华教授在 2016 年提出 [1, 2]。在学件范式下，世界各地的开发者可分享模型至学件基座系统，系统通过有效查搜和复用学件帮助用户高效解决机器学习任务，而无需从零开始构建机器学习模型。
学件由周志华教授在 2016 年提出 [1, 2]。在学件范式下，世界各地的开发者可分享模型至学件基座系统，系统通过有效查搜和复用学件帮助用户高效解决机器学习任务，而无需从零开始构建机器学习模型。
北冥坞是学件范式的首次系统性实现，为学件相关研究提供了坚实的科研平台。学件由性能优良的机器学习模型和描述模型的规约组成。规约刻画了模型的能力，使得模型在未来能够根据用户需求被充分识别和复用。规约由两部分构成：语义规约通过文本描述模型的功能，而统计规约刻画模型所蕴含的统计信息。
北冥坞是学件的第一个系统性开源实现，为学件相关研究提供了一个初步科研平台。有分享意愿的开发者可免费提交模型，学件坞协助产生规约形成学件存放在学件坞中，开发者在这个过程中无需向学件坞泄露自己的训练数据。未来的用户可以通过向学件坞提交需求，在学件坞协助下查搜复用学件来完成自己的机器学习任务，且用户可以不向学件坞泄露自有数据。预计在学件坞拥有数以百万计的学件后，将可能出现“涌现”行为：以往没有专门开发过模型的机器学习任务，可能通过复用若干个现有学件而解决。
学件由性能优良的机器学习模型和描述模型的规约组成。规约刻画了模型的能力，使得模型在未来能够根据用户需求被充分识别和复用。规约由两部分构成：语义规约通过文本描述模型的功能，而统计规约刻画模型所蕴含的统计信息。
[1] Zhi-Hua Zhou. Learnware: on the future of machine learning. Frontiers of Computer Science, 2016, 10(4): 589–590 <br/>
[2] 周志华. 机器学习: 发展与未来. 中国计算机学会通讯, 2017, vol.13, no.1 (2016 中国计算机大会 keynote)
## 北冥坞系统有哪些特性？
如下图所示，北冥坞学件基座系统作为学件范式的科研平台，首次系统性地实现了学件范式中的核心流程：
如下图所示，北冥坞学件基座系统作为学件的初步科研平台，首次系统性地实现了学件范式中的核心流程：
- **提交阶段**：系统内置了多重检测机制，以确保上传学件的质量。另外，系统会根据已有的学件规约，训练一个异构引擎，用于合并不同的规约岛屿，以及为学件赋予新规约。随着更多学件的上传，异构引擎将持续更新，实现学件规约的持续迭代，构建更精准的规约世界。
- **部署阶段**：用户上传任务需求后，系统会自动选择是推荐单学件还是多学件组合，并提供高效的部署方式。无论是单个学件还是多学件组合，系统均提供了便捷的学件复用接口。