Ever now and then I talk to someone who tell me that they can not get good feedback on their research (e.g. the they don't get much responses on their alignment forum post), and been thinking about how to solve this? Also, right now is a good time to try out various online solutions.

I am offering to run webinars if enough people are interested.

The format I suggest is that each webinar is 1h 40min and centers around one person presenting their idea or result.

  • 5 min introduction
  • 30 min presentation
  • 30 min of questions and discussion centered around trying to understand and improve idea.
  • 30 min of attacking the idea to find out all the ways it can fail.
  • 5 min rounding off but the presenter summering what their take away is.

The reason I suggest this structure is

  1. I think a webinar needs a structure to keep on track
  2. It is important to look for weaknesses in an idea, but it is even more important to fist understand it and to make sure you attack the strongest version of that idea.

I am very much open to different format, and would like to experiment a bit if there is enough interest.

Edit: Not sure if I should interpret up votes as interest. If this is something you would like to participate in, pleas leave a comment.

AICommunity
Frontpage
New Answer
New Comment

1 Answers sorted by

Chris_Leong

20

I would be keen run a webinar on Logical Counterfactuals

Let's do it!

If you pick a time and date and write up an abstract, then I will sort out the logistic. Worst case it's just you and me having a conversation, but most likely some more people will show up.