关于深度学习

深度学习革命

★ 世界范围内主要的人工智能玩家有哪些？ ★ 哪家科技公司在这场脑力与毅力的角逐中脱颖而出？ ★ 深度学习革命如何颠覆了整个人工智能领域？ ★ 人工智能还会经历第三次寒冬吗？ 长期以来，人工智能一直被...

https://book.douban.com/subject/36171345/

2007 年——杰夫·辛顿创造了术语“深度学习”，一种描述神经网络的方式。

60 岁生日那天，辛顿在温哥华举行的年度NIPS大会上发表演讲，“深度学习”这个词第一次出现在标题中。这是一次巧妙的品牌重塑。提到多层神经网络，讲深度学习并不是什么新鲜事。但“深度学习”是一个令人回味的术语，旨在激励研究人员在一个再次失宠的领域进行研究。当他在演讲中说其他人都在做“浅薄学习”时，听众们发出了笑声，他知道这是一个好名字。从长远来看，这将被证明是一个高明的选择，它立刻提高了在学术界边缘工作的这一小部分研究人员的声誉。

History

人工智能发展简史

人工智能的起源可以追溯到阿兰·图灵(Alan Turing)1936年发表的《论可计算数及其在判定问题中的应用》

https://www.aminer.cn/ai-history

AI Canon | Andreessen Horowitz

A curated list of resources we’ve relied on to get smarter about modern AI, including generative AI, LLMs, and transformer models.

https://a16z.com/ai-canon/

So, in this post, we’re sharing a curated list of resources we’ve relied on to get smarter about modern AI. We call it the “AI Canon” because these papers, blog posts, courses, and guides have had an outsized impact on the field over the past several years.

Tutorials

For Beginners

Datawhale-学用 AI,从此开始

Datawhale 是国内最大的 AI 开源学习社区，覆盖海内外2000多所院校，吸引具有影响力的顶尖学者、知名教授、企业家、青年贡献者等在此分享知识和远见，加快高校和产业的链接，持续推动AI人才培养。

https://www.datawhale.cn/

Dive into Deep Learning — Dive into Deep Learning 1.0.3 documentation

Interactive deep learning book with code, math, and discussions Implemented with PyTorch, NumPy/MXNet, JAX, and TensorFlow Adopted at 500 universities from 70 countries Star

https://d2l.ai/

Welcome to the UvA Deep Learning Tutorials! — UvA DL Notebooks v1.2 documentation

Interested in learning JAX? We have recently started translating the notebooks from PyTorch to JAX+Flax. Check out our new notebooks in the tab Deep Learning 1 (JAX+Flax) to learn how you can speedup the model training with JAX!

https://uvadlc-notebooks.readthedocs.io/en/latest/index.html

深度学习入门教程 | 人工智能安全研究实验室·上海交通大学

新手教程内含四个板块，分别为网络课程，常用文档，技术博客，常用网上资源。四个板块中的链接都是主流的深度学习中强有力的学习资料。

https://magiclabsjtu.github.io/guide/2018-10-27-deep-learning-tutorial/

Technique Blogs

Lil'Log

Document my learning notes.

https://lilianweng.github.io/

DaNing的博客

DaNing的个人博客.

https://adaning.github.io/

OpenFace入门心得 - 余弦G的博客

前言因为毕设使用到了OpenFace，趁着学习和使用的机会记录下入门的心得。简介OpenFace —— 一种旨在用于计算机视觉和机器学习研究人员，情感计算社区以及对基于面部行为分析构建交互式应用...

https://www.cosineg.com/archives/11/

Playground

Transformer Explanation

mp.weixin.qq.com

https://mp.weixin.qq.com/mp/appmsgalbum?__biz=MzI5MDUyMDIxNA==&action=getalbum&album_id=2090301627206303744&scene=173&subscene=&sessionid=svr_103c5be1a52&enterid=1717657689&from_msgid=2247604726&from_itemidx=1&count=3&nolastread=1#wechat_redirect

Generative AI exists because of the transformer

The technology has resulted in a host of cutting-edge AI applications — but its real power lies beyond text generation

https://ig.ft.com/generative-ai/

LLM Explanation

How LLMs Work, Explained Without Math

I'm sure you agree that it has become impossible to ignore Generative AI (GenAI), as we are constantly bombarded with mainstream news about Large Language Models (LLMs). Very likely you have tried…

https://blog.miguelgrinberg.com/post/how-llms-work-explained-without-math

LLM Visualization

A 3D animated visualization of an LLM with a walkthrough.

https://bbycroft.net/llm

What Is ChatGPT Doing … and Why Does It Work?

Stephen Wolfram explores the broader picture of what's going on inside ChatGPT and why it produces meaningful text. Discusses models, training neural nets, embeddings, tokens, transformers, language syntax.

https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/

mp.weixin.qq.com

https://mp.weixin.qq.com/mp/appmsgalbum?action=getalbum&__biz=MzIzNzI3ODUyNw==&scene=1&album_id=2810936357031542785&count=3#wechat_redirect

评论

特德·姜：ChatGPT Is a Blurry JPEG of the Web

原文地址： https://www.newyorker.com/tech/annals-of-technology/chatgpt-is-a-blurry-jpeg-of-the-web Deepl 机翻，未做校对。 ChatGPT Is a Blurry JPEG of the Web ChatGPT是网络上的一个模糊的JPEG文件 OpenAI’s chatbot offers paraphr...

https://www.douban.com/note/844770085

Simply explained: how does GPT work?

By now, you have probably heard of OpenAI’s ChatGPT, or any of the alternatives GPT-3, GPT-4, Microsoft’s Bing Chat, Facebook’s LLaMa or even Google’s Bard. They are artificial intelligence programs that can participate in a conversation. Impressively smart, they can easily be mistaken for humans, and are skilled in a variety of tasks, from writing a dissertation to the creation of a website. How can a computer hold such a conversation?

https://confusedbit.dev/posts/how_does_gpt_work/