Rational Feed

Rationalist:

Dimensional Decoupling by Brian Lui – Its often useful to split concepts into component parts. For example someone might split people into ‘good friends’ (nice and generous) and ‘bad friends’ (mean and selfish). However ‘niceness’ and ‘generosity’ are fairly orthogonal concepts and should not be compressed into a single axis.

Science Under High Modernism by sam[]zdat – Meditations on Thomas Kuhn’s theory of scientific paradigm shifts and the difficulty of crafting explicit definitions.

The Second Circle by Zvi Moshowitz – Zvi discusses a rationalist ‘circle’ in NYC that did not go well. (Circling is way of having group conversations about your experiences and emotions)

AI:

Challenges To Christiano S Capability Amplification Proposal by Eliezer – Paul Christiano (long term rationalist who works at Open AI) has a theory of how to successfully align an advanced AI. Eliezer responds with a long and detailed criticism. Probably only worth reading if you are familiar with Paul’s work already.

Erik Brynjolfsson Interviews Daniel Kahneman by Marginal Revolution – Daniel Kahneman text interview about AI and various problems with bias. Kahneman think AI handles bias much better than humans.

Advertisements

Author: deluks917

Software Engineer in New York City. Admin of the SlateStarcodex discord. Math PHD drop out.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s