# Using Machine Learning to understand our society I'm Ram Rachum, and this is the knowledge base for my research. It has many notes of thoughts that I have that I hope will lead me to interesting results. Feel free to dive in. **I want to use Machine Learning to understand our society.** Specifically, I'd like to show that selfish agents can show [[emergent reciprocity]], and from that, social behavior like forming groups. I hope that this can help us understand how humans can cooperate better, and how we can build safe Artificial General Intelligence. The field of my research is called [Multi-Agent Reinforcement Learning](https://en.wikipedia.org/wiki/Multi-agent_reinforcement_learning). I'll be happy to get any comments and feedback. Email me at [email protected] ## Intro to my research - [[About Ram Rachum]] - [[The goals of my research]] - [Talk video](http://r.rachum.com/talk-video) A 50-minute talk in which I explain what I'm trying to achieve with my research. No background in Machine Learning is needed. This is a newer version of the talk I've given internally to an audience of 2,300 Googlers. - [Talk video (Hebrew)](http://r.rachum.com/talk-video-hebrew) - [Slides](http://r.rachum.com/talk-slides) ## Going deeper - [[How my approach is different]] A list of the different ways in which my research effort is different than other researchers'. - [[Why I'm developing Marley]] ## Sign up to get updates Sign up to my [research mailing list](http://r.rachum.com/announce) to get monthly updates about my research. Every month I outline the goals for that month, and evaluate my progress on last month's goals. If you sign up you could also look at the most recent updates. ## 🌟 Latest updates * 2022-08-01: I'm flying out to Prague to participate in [HAAISS 2022](http://humanaligned.ai/). * 2022-07-13: I gave a [lightning talk](https://www.youtube.com/watch?v=tRtxCCRdZOs&t=33904s) about my research at [EuroPython 2022](https://ep2022.europython.eu/). * 2022-07-02: I've prepared a talk about the [[Fruit Slots]] experiments. In these experiments I set up an environment where the agents learn to communicate with each other implicitly, without being directly rewarded for it. I'm going to give that talk at the [Workshop on Ad Hoc Teamwork at IJCAI 22](https://sites.google.com/view/ad-hoc-teamwork/home).