集智因果读书会项目说明

这里需要简单说明读书会的愿景,日程表等内容

Causal AI

Judea Pearl 是 Causal AI 的奠基人,Bernhard Scholkopf 推进了 Causality for Machine Learning,Yoshua Bengio 提出了 System 2 deep learning 作为 Causal AI 的一个范式。沉醉于 life and Intelligence 之美,尤其是人类社会系统的群体智能。众多工具中(包括数学,计算机,物理,复杂系统等等),偏好用信息论视角研究如何教会机器因果思维,希望创造具备 free will 的 AI,使之成为我们的良师益友,一起探索解密生命和智能的终极奥秘。

该项目包含的内容有:

  • 因果推断一个人,一篇论文,一个视频,一个slide, 一个会议,一个教程,一句话,说明因果推断。

  • 从 Judea Pearl 到 Bernhard Scholkopf,再到 Youshua Bengio 和相关综述来把握因果理论前沿研究。

  • Causal Inference and Data-Fusionin Econometrics Paul Hunermund 和 Elias Bareinboim(Judea Pearl 学生)是在披着经济学的皮讲解着 Causal AI 如何解决 confounding bias, selection bias and 迁移学习这些难题的因果理论框架。

  • 序列数据因果推断:1)经典教材 Elements of Causal Inference: Foundations and Learning Algorithms Ch10 Time Series 和全面综述博客 Inferring causality in time series data A concise review of the major approaches.

常见问题集

  1. 什么是小图灵测试?

How can machines represent causal knowledge in a way that would enable them to access the necessary information swiftly, answer questions correctly, and do it with ease, as a human can?

  1. How causal graphs will change the future?

Pearl: it will enable us to understand ourselves. To implement agency, to implement free will in a machine so that we can communicate better with machine. And by communicating better we gain a greater understanding of ourselves.