Learning-by-doing AI Safety Research workshop
- [Open in Google Maps] [Open in local app]
- 16 August 2019, 10:00 am—19 August 2019, 7:00 pm
- EA Hotel, York Street, Blackpool, UK
- Contact: linda.linsefors@gmail.com
Edit: This workshop is now full, but due to the enthusiasm I have received I am going to organize a second Learning-by-doing AI Safety workshop some time in October/November this year. If you want to influence when it will be you can fill in our doodle: https://doodle.com/poll/haxdy8iup4hes9xy
I am leaving the application form open. You can fill it in to show interest in the second Learning-by-doing AI Safety workshop and future similar events.
Start: Friday, August 16, 10am
End: Monday, August 19, 7pm
Location: EA Hotel, 36 York Street, Blackpool
The main activity during the workshop will be trying to solve AI Safety. Maybe we will discover something useful, maybe not. This is an experimental workshop so I don’t know what the outcome will be. But you will for sure learn things about the AI Safety along the way.
This event is beginner friendly. You don’t need to know anything about AI Safety to attend. You will need some understanding of Machine Learning, but we’ll teach you that too if you want.
On Friday August 16, I will teach a very basic overview of Machine Learning. At the end of the day you will hopefully have a decent understanding of what ML can do and how and why it works. This day is optional. If you already know some ML, or promise to teach yourself before the event, you can skip this day.
Preliminary Schedule
Friday (optional): Machine Learning Speed Learning
Saturday: We try to solve AI Safety + Lectures on AI Safety
Sunday: Lectures / Discussions / Guided self study / Free time
Monday: We try to solve AI Safety again + Write down good ideas
The AI Safety lectures will be me and Davide Zagami sharing our knowledge. I have studied ML and AI Safety independently for about two years and I have interned at MIRI. Davide has studied AI Safety independently for about one year and he is currently working for RAISE. There are still lots of things we don’t know. If you are looking for something more advanced, I recommend that you apply for Human-aligned AI Summer School in Prague.
You are welcome to stay longer at EA Hotel before or after the workshop. However, be aware that there is another event, Technical AI Safety Unconference, happening the following week. If you want to participate in the unconference you should apply for that separately.
If you are staying more than a few days extra, we ask you to book your stay though the EA Hotel booking system.
Price: This workshop is sponsored by MIRI and will therefore be free :)
Food and Lodging: Food and lodging is included. All food will be vegan.
If you want to join: Sign up here
- The Case for Impact Purchase | Part 1 by (EA Forum; 14 Apr 2020 13:08 UTC; 51 points)
- EA Hotel Fundraiser 7: Pitch focusing on case studies with counterfactuals by (EA Forum; 22 Nov 2019 11:25 UTC; 44 points)
- [AN #57] Why we should focus on robustness in AI safety, and the analogous problems in programming by (5 Jun 2019 23:20 UTC; 26 points)
- Two AI Safety events at EA Hotel in August by (EA Forum; 21 May 2019 18:57 UTC; 24 points)
- AIXSU—AI and X-risk Strategy Unconference by (3 Sep 2019 11:35 UTC; 24 points)
- TAISU—Technical AI Safety Unconference by (21 May 2019 18:34 UTC; 15 points)
- AI and X-risk Strategy Unconference at EA Hotel in November by (EA Forum; 24 Oct 2019 10:52 UTC; 5 points)
This workshop is now full, but due to the enthusiasm I have received I am going to organize a second Learning-by-doing AI Safety workshop some time in October/November this year. If you want to influence when it will be you can fill in our doodle: https://doodle.com/poll/haxdy8iup4hes9xy
I am leaving the application form open. You can fill it in to show interest in the second Learning-by-doing AI Safety workshop and future similar events.