Curated Posts

Curated Posts
starCurated - Recent, high quality posts selected by the LessWrong moderation team.
rss_feed Create an RSS Feed
Frontpage Posts
Posts meeting our frontpage guidelines: • interesting, insightful, useful • aim to explain, not to persuade • avoid meta discussion • relevant to people whether or not they are involved with the LessWrong community.
(includes curated content and frontpage posts)
rss_feed Create an RSS Feed
...
See more options

Adult Neurogenesis – A Pointed Review

SSC
15d
10 min read
Show Highlightsubdirectory_arrow_left
12

A Sketch of Good Communication

19d
2 min read
Show Highlightsubdirectory_arrow_left
17

Can corrigibility be learned safely?

18d
3 min read
Show Highlightsubdirectory_arrow_left
82

Reframing misaligned AGI's: well-intentioned non-neurotypical assistants

19d
1 min read
Show Highlightsubdirectory_arrow_left
14

Explicit and Implicit Communication

1mo
13 min read
Show Highlightsubdirectory_arrow_left
18

Argument, intuition, and recursion

2mo
8 min read
Show Highlightsubdirectory_arrow_left
9

An Untrollable Mathematician Illustrated

1mo
1 min read
Show Highlightsubdirectory_arrow_left
30

Hammertime Final Exam

Radimentary
1mo
2 min read
Show Highlightsubdirectory_arrow_left
14

The Costly Coordination Mechanism of Common Knowledge

1mo
66 min read
Show Highlightsubdirectory_arrow_left
23

[Event] AI Summer Fellows Program

06/27/2018 Bodega Bay, CA, USA
Show Highlightsubdirectory_arrow_left
0

[Event] Less Wrong Warsaw April Meetup

Thursday at 6:00 PM Warsaw, Poland
Show Highlightsubdirectory_arrow_left
0

[Event] Portland SSC Meetup 4/29/18

04/30/2018 Behind The Museum Cafe, Southwest 10th Avenue, Portland, OR, USA
Show Highlightsubdirectory_arrow_left
0
Meetup : Ann Arbor Meetup, 2/19/16
Bound_up2y1 min readShow Highlightsubdirectory_arrow_left

Discussion article for the meetup : Ann Arbor Meetup, 2/19/16

WHEN: 19 February 2016 07:00:00PM (-0500)

WHERE: 611 Church St, Ann Arbo... (Read More)

It is easy to understand, detailed and meticulous! I have had a lot of harvest after watching this article from you! I feel it interesting, your post gave me a new perspective! I have read many other articles about the same topic, but your article convinced me! I hope you continue to have high quali...(read more)

Weird question: could we see distant aliens?
paulfchristiano2h3 min readShow Highlightsubdirectory_arrow_left

tl;dr: Paul has a weird question about aliens that is probably easy. He is offering a prize that starts at $100 and grows to $10,000 over the next 7 weeks.

Suppose there was a large alien civilization halfway across the observable univer... (Read More)

Clarifying question - how much can these aliens move? You talked about visual signals, but is that necessary? If they're allowed to move as much as the want, what's wrong with a plain old von Neumann probe? Too slow? Too expensive? But if they're not allowed to move from their galaxy, then I'm afrai...(read more)

Minor issue - for us to see a signal from "far away", the signal needs to have been sent "long time ago" (naively you'd say that a signal from 7 billion light years away needs to be sent 7 billion years ago, but with expansion that's not quite true, so I'll just stick with "far away" and "long ago")...(read more)

Funeral Ritual
Raemon7h7 min readShow Highlightsubdirectory_arrow_left

Followup to: Solstice: What Memes May Come (Part II)

_Content note: My own opinionated thoughts on funeral ritual, which have evolve... (Read More)

Couple thoughts on that:

**1\. An important principle in ritual is "be as oddly specific as you can get away with" (but no more).**

The more a ritual communicates "the people in this room are connected by something special unique and important", the more moving an experience it'll be. It doesn't d...(read more)

I don't think Beyond the Reach of God is an ideal text for an event like this as it's not inclusive to everyone. With the last Solstice we stopped using it because we had a rationalist Christian with us. Inclusivity seems to me even more important for an event like a funeral.

Fortunately, I haven't...(read more)

:(

Understanding Iterated Distillation and Amplification: Claims and Oversight
William_S2d9 min readShow Highlightsubdirectory_arrow_left

[Background: Intended for an audience that has some familiarity with Paul Christiano’s approach to AI Alignment. Understanding Iterated Distillation and Amplification

> [Note: I probably like explicit reasoning a lot more than most people, so keep that in mind.] Great! We need more people like you to help drive this forward. For example I think we desperately need explicit, worked out examples of meta-execution (see my request to Paul [here](https://www.lesswr...(read more)

\[Note: I probably like explicit reasoning a lot more than most people, so keep that in mind.\]

"our naive intuitions about what a large group of humans can do become less and less applicable, so we need to use explicit reasoning and/or develop new intuitions about what HBO-based and especially LBO...(read more)

When you say "better", do you mean in an empirical way that will be apparent with testing? If so, would it result in answers that are rated worse by users, or in answers that are less corrigible/safe?

My personal guess is that explicit reasoning with very large budgets can outperform reasonably-bou...(read more)

Can corrigibility be learned safely?
Wei_Dai18d3 min readShow Highlightsubdirectory_arrow_left

EDIT: Please note that the way I use the word "corrigibility" in this post isn't quite how Paul uses it. See this thread for clarific... (Read More)

> But in meta-execution the type signature is a giant tree of messages (which can be compressed by an approval-directed encoder); I don’t see how to use that type of “value” with any value-learning approach not based on amplification (and I don’t see what other type of “value” is plausible). In t...(read more)

> I also take your point that it's 'a potential approach for learning a "comprehensible" model of the world', however I don't have a good understanding of how this is really supposed to work (e.g., how does the comprehensibility property survive the distillation steps)

Models and facts and so on ar...(read more)

Could we approximate a naive function that agents would be attempting to maximize (for the sake of understanding)? I imagine it would include:

1\. If the user were to rate this answer, where supplemental & explanatory information is allowed, what would be their expected rating?

2\. How much did th...(read more)

Humans need places
Benquo13h3 min readShow Highlightsubdirectory_arrow_left

If, when you try to improve the world, you think about people but not about communities, you will tend to favor unsustainable net outflows of resources from your community. I wrote about this in Why I am not a Quaker

I agree that community space is something very important.

When it comes to running events having a good space in which to run the events is very valuable. Even through I live in Berlin with much lower costs of space, how to get a good space to run an event has often be a constraining factor.

When ...(read more)

Looking forward to seeing your perspective :) Added a link to Melting Gold.

I want to do a big shout-out to the awesome Eric Rogstad, Elizabeth Garrett and (more recently) Claire Wang who have also created a large amount of slack in the Berkeley community by (amongst other things) maintaining the Event Horizon Rationalist House, where people are able to visit and stay. Sara...(read more)