@BenPrystawski transmitting knowledge between generations, and over decades, by human memory, with noise

Ben Prystawski @BenPrystawski I’ll be presenting on “Cultural reinforcement learning: a framework for modeling cumulative culture on a limited channel” at 1:45 #cogsci2023 this afternoon! My talk will be in the “Culture and Language II” session.

The paper is here if you’re interested: https://psyarxiv.com/q4tz8/
Replying to @BenPrystawski

Ben, When I was a teenager about 60 years ago, I found that I could send messages to my future self by forming strong memories with triggers. “In this very specific situation, ALWAYS remember ___”. I did not send many notes or warnings, but it works over at least 50 years.

Ben, As a teenager 60 years ago, I found how to send messages to my future self forming memories with strong triggers. “In this VERY SPECIFIC situation, ALWAYS remember ___”. I did not send many notes or warnings, but it works losslessly over 20, 40, 50 years. Try global issues.


xAI @xai What best describes your view for a positive future on a long time horizon?

Solar system colonization, atomic fuels, 3D interiors of suns planets, gravitational engineering, rewrite the Internet, lifetime human-AI partners, simplify DNA, full brain storage, all global and systemic issues and opportunities mapped, all knowledge open and accessible to all.


Nature,

I cannot read all things, but I might check what Nature publishes, if it is in open database or index format.
I want to set topics of interest, view the abstracts titles and authors.  I used to read 55 journals each month, and tried to read most of the articles in depth.  I might be able to teach my AI how to read and filter that now, but human eyeballs and human processing has limits.

An AI-assisted topic and group monitoring service?

You might want to rethink “more print makes smarter humans”.

Richard Collins, The Internet Foundation


29 Jul 2023: Dinghuai Zhang @ ICML 张鼎怀 @zdhnarsil We are holding a panel discussion about probabilistic methods and generative modeling with @xuyilun2 @YSongStanford @_bunnech @ArashVahdat @RickyTQChen moderated by @chenlin_meng at our SPIGM workshop at room 323  https://spigmworkshop.github.io https://pic.twitter.com/O7oqic9tF3
Replying to @zdhnarsil @chenlin_meng and 5 others

This looks interesting, but seems to be “pay per view”, only available to paid conference attendees. Much AI “discussion” on the Internet is not clear organized accessible open for global collaborative goals. 5 Billion humans, some access to the Internet – write for ALL of them.

Richard K Collins

About: Richard K Collins

The Internet Foundation Internet policies, global issues, global open lossless data, global open collaboration


Leave a Reply

Your email address will not be published. Required fields are marked *