text stringlengths 300 320k | source stringlengths 52 154 |
|---|---|
# Combat vs Nurture: Cultural Genesis
_In my post [Conversational Cultures: Combat vs Nurture](https://www.lesswrong.com/posts/ExssKjAaXEEYcnzPd/conversational-cultures-combat-vs-nurture), I described two different sets of norms and assumptions norms used in discussion. In this follow-up post, I add some important cla... | https://www.lesswrong.com/posts/Y47hCzDMKmiTzBeHi/combat-vs-nurture-cultural-genesis |
# Aligned AI, The Scientist
The problem with constructing an aligned AI is that any active utility function or attempt at world optimization is likely to succumb to the Goodhart's law in one of its many forms, as discussed here and elsewhere by the good people of MIRI. I wonder if a more passive approach is worth cons... | https://www.lesswrong.com/posts/6TmoCLNrTm8iep9G6/aligned-ai-the-scientist |
# AI development incentive gradients are not uniformly terrible
Much of the work for this post was done together with Nuño Sempere
Perhaps you think that your values will be best served if the AGI you (or your team, company or nation) are developing is deployed first. Would you decide that it's worth cutting a few ... | https://www.lesswrong.com/posts/bkG4qj9BFEkNva3EX/ai-development-incentive-gradients-are-not-uniformly |
# Alignment Newsletter #32
Remember, treat all of the "sequence" posts as though I had highlighted them!
Highlights
----------
**[Spinning Up in Deep RL](https://blog.openai.com/spinning-up-in-deep-rl/)** _(Joshua Achiam)_: OpenAI has released an educational resource aimed to help software engineers become skilled a... | https://www.lesswrong.com/posts/uHE9b4HWQjuqnFYqP/alignment-newsletter-32 |
# Kelly bettors
### Updated Introduction
I wrote this post roughly two years ago, and somehow it randomly got crossposted to LessWrong today. Nevertheless, despite being poorly written, I think it's neat enough that I've polished it off and left it up. It goes through a needless amount of needlessly general algebra t... | https://www.lesswrong.com/posts/iWXQgwpksstozSDeA/kelly-bettors |
# Laughing Away the Little Miseries
_Note: This falls under the category of self-help advice, the usefulness of which tends to be variable between individuals. This technique has noticeably improved my day to day feelings of well being, but people are often very different and your mileage may vary._
It is notorio... | https://www.lesswrong.com/posts/66b9TuDnfBeGcSzTb/laughing-away-the-little-miseries |
# The Steering Problem
Most AI research focuses on reproducing human abilities: to learn, infer, and reason; to perceive, plan, and predict. There is a complementary problem which (understandably) receives much less attention: if you _had_ these abilities, what would you do with them?
**The steering problem:** Using ... | https://www.lesswrong.com/posts/4iPBctHSeHx8AkS6Z/the-steering-problem |
# Acknowledging Human Preference Types to Support Value Learning
We analyze the usefulness of the framework of preference types \[Berridge et al. 2009\] to value learning by an artificial intelligence. In the context of AI the purpose of value learning is giving an AI goals aligned with humanity. We will lay the g... | https://www.lesswrong.com/posts/mSPsyEwaymS74unND/acknowledging-human-preference-types-to-support-value |
# Mandatory Obsessions
Cross-posted from [Putanumonit](https://putanumonit.com/2018/11/07/mandatory-obsessions/).
The Worst is to Care a Little
-----------------------------
At a party the other day, a woman expressed her displeasure at men who “pretend to be allies of gay rights but don’t really care about it”. Did... | https://www.lesswrong.com/posts/cLcAXd8jNBWbZ9sQA/mandatory-obsessions |
# Stoicism: Cautionary Advice
Stoicism, or life strategies in that sphere, are something that often gets positive attention. I’m referring to the wide variety of advice that is along the lines of “look for the silver lining” and “laugh off the small things”. At the core, the stoic mindset is the recognition that we ha... | https://www.lesswrong.com/posts/JCoD9n3cBiXxgyRZK/stoicism-cautionary-advice |
# Explore/Exploit for Conversations
_Something status: I've started to try on an explore/exploit lens for thinking about conversation, and I like it. Here's some rough ideas. On the editing pass of this, a feel like I haven't properly separated the important ideas explicitly enough, but it's fine._
You are in explore... | https://www.lesswrong.com/posts/vgd38SNS8M43t2sqf/explore-exploit-for-conversations |
# Clarifying "AI Alignment"
When I say an AI A is _aligned with_ an operator H, I mean:
> _A is trying to do what H wants it to do._
The “alignment problem” is the problem of building powerful AI systems that are aligned with their operators.
This is significantly narrower than some other definitions of the alignme... | https://www.lesswrong.com/posts/ZeE7EKHTFMBs8eMxn/clarifying-ai-alignment |
# Switching hosting providers today, there probably will be some hiccups
We are switching hosting providers today from Meteor Galaxy to Amazon AWS (mostly because they have better autoscaling and are cheaper). This will involve a bunch of DNS server changes, and will be the first time we hook up our new autoscaling sy... | https://www.lesswrong.com/posts/amGDvs2ztj35mBF4Y/switching-hosting-providers-today-there-probably-will-be |
# Embedded Agency (full-text version)
Suppose you want to build a robot to achieve some real-world goal for you—a goal that requires the robot to learn for itself and figure out a lot of things that you don't already know.
There's a complicated engineering problem here. But there's also a problem of figuring out what... | https://www.lesswrong.com/posts/i3BTagvt3HbPMx6PN/embedded-agency-full-text-version |
# Sam Harris and the Is–Ought Gap
Many secular materialist are puzzled by Sam Harris's frequent assertion that [science can bridge Hume's is–ought gap](https://www.amazon.com/Moral-Landscape-Science-Determine-Values/dp/143917122X/ref=sr_1_1?ie=UTF8&qid=1542325209&sr=8-1&keywords=the+moral+landscape&dpID=51O-%252BvFIVB... | https://www.lesswrong.com/posts/HLJGabZ6siFHoC6Nh/sam-harris-and-the-is-ought-gap |
# History of LessWrong: Some Data Graphics
Some graphs showing posting activity on LessWrong through the years.
*NOTE: If you’re reading this post on GreaterWrong, you can click on the images to enlarge, zoom in, and click through them all as a slideshow.*
Comments per post:
}
\newcommand... | https://www.lesswrong.com/posts/Qa5jG9z9dC6E4s9JH/dimensional-regret-without-resets |
# Is Clickbait Destroying Our General Intelligence?
(Cross-posted from Facebook.)
* * *
Now and then people have asked me if I think that other people should also avoid high school or college if they want to develop new ideas. This always felt to me like a wrong way to look at the question, but I didn't know a right... | https://www.lesswrong.com/posts/YicoiQurNBxSp7a65/is-clickbait-destroying-our-general-intelligence |
# Fixed Point Exercises
Sometimes people ask me what math they should study in order to get into agent foundations. My first answer is that I have found the introductory class in every subfield to be helpful, but I have found the later classes to be much less helpful. My second answer is to learn enough math to unders... | https://www.lesswrong.com/posts/mojJ6Hpri8rfzY78b/fixed-point-exercises |
# Topological Fixed Point Exercises
This is one of three sets of fixed point exercises. The first post in this sequence is [here](https://www.lesswrong.com/posts/mojJ6Hpri8rfzY78b/fixed-point-theorems), giving context.
**1.** (1-D Sperner's lemma) Consider a path built out of $n$ edges as shown. Color each vertex blu... | https://www.lesswrong.com/posts/svE3S6NKdPYoGepzq/topological-fixed-point-exercises |
# Act of Charity
_(Cross-posted from [my blog](https://unstableontology.com/2018/11/17/act-of-charity/))_
> The stories and information posted here are artistic works of fiction and falsehood. Only a fool would take anything posted here as fact.
>
> —Anonymous
# Act I.
Carl walked through the downtown. He came ac... | https://www.lesswrong.com/posts/rwjv8bZfSuE9ZAigH/act-of-charity |
# On Rigorous Error Handling
In my long career as a programmer I've seen a lot of projects with a lot of error handling practices and, sadly, almost none that actually worked.
We are facing facing few high-level problems here:
1. For many applications error handling is not critical.
2. Error handling code is rarel... | https://www.lesswrong.com/posts/MJiHEhES5SfEZLE3E/on-rigorous-error-handling |
# An unaligned benchmark
My goal is to design AI systems that are aligned with human interests and [competitive](https://ai-alignment.com/directions-and-desiderata-for-ai-control-b60fca0da8f4) with unaligned AI.
I find it useful to have a particular AI algorithm in mind. Then I can think about how that algorithm coul... | https://www.lesswrong.com/posts/ZHXutm7KpoWEj9G2s/an-unaligned-benchmark |
# Ia! Ia! Extradimensional Cephalopod Nafl'fhtagn!
Hello, I'm ExCeph (short for Extradimensional Cephalopod). I've been mostly lurking here for the past decade and occasionally commenting within the past year or so. However, I never got around to actually introducing myself until now. It's good to meet you.
I was adv... | https://www.lesswrong.com/posts/K9TbuFyrxqEdByD6a/ia-ia-extradimensional-cephalopod-nafl-fhtagn |
# Diagonalization Fixed Point Exercises
This is the second of three sets of fixed point exercises. The first post in this sequence is [here](https://www.lesswrong.com/posts/mojJ6Hpri8rfzY78b/fixed-point-theorems), giving context.
1) Recall Cantor’s diagonal argument for the uncountability of the real numbers. Apply t... | https://www.lesswrong.com/posts/FZkLa3GRLW97fpknG/diagonalization-fixed-point-exercises |
# Collaboration-by-Design versus Emergent Collaboration
Introduction
------------
This seems to be a non-trivial problem for even current narrow AI, which is much more problematic for strong NAI, which I haven't seen called out or named explicitly. I provide a quick literature review to explain why I think it's ignor... | https://www.lesswrong.com/posts/ts3CKRDHgkBasAy4J/collaboration-by-design-versus-emergent-collaboration |
# [Link] "They go together: Freedom, Prosperity, and Big Government"
[Link to article](https://evonomics.com/freedom-prosperity-big-government/?utm_source=newsletter_campaign=organic)
> Our statistical investigations lead to two substantive conclusions:
>
> First, the data appear to support notion that economic fre... | https://www.lesswrong.com/posts/nwvnZhDoGg6GLF4jE/link-they-go-together-freedom-prosperity-and-big-government |
# Clickbait might not be destroying our general Intelligence
Epistemic status This post is a "plausible conjecture" alternative to Eliezer's take on clickbait. My arrogant epistimology thinks it's substantially closer to the truth than Eliezer's version.
https://www.lesswrong.com/posts/YicoiQurNBxSp7a65/is-clickbait-... | https://www.lesswrong.com/posts/HSjQSSw94NHhmq8cQ/clickbait-might-not-be-destroying-our-general-intelligence |
# Letting Others Be Vulnerable
_\[Because people are predisposed to give others good impressions, when our relationships deepen, it’s more likely that we’ll have to revise our initial “perfect” models of the other person. For the person doing the updating, it can internally feel uncomfortable. I outline this dynamic a... | https://www.lesswrong.com/posts/RB6qAS8YoWSjb6oRy/letting-others-be-vulnerable |
# Alignment Newsletter #33
Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through the [database](https://docs.google.com/spreadsheets/d/1PwWbWZ6FPqAgZWOoOcXM8N_tUCuxpEyMbN1NYYC02aM/edit?usp=sharing) of all ... | https://www.lesswrong.com/posts/KJPLW9XTaF3WxKxmq/alignment-newsletter-33 |
# [Insert clever intro here]
Hello, everyone. I'm not entirely sure where to start, or exactly how to say it, but here goes.
I only stumbled across Less Wrong this year, after reading HPMOR (I'm sure you get this all the time). Since then, I have gone from an evangelical Christian to a passionate rationalist with an ... | https://www.lesswrong.com/posts/rad4gei5EYgXhHo5s/insert-clever-intro-here |
# Prosaic AI alignment
(_Related:_ _[a possible stance for AI control](https://medium.com/ai-control/a-possible-stance-for-ai-control-research-fe9cf717fc1b)._)
It’s conceivable that we will build “prosaic” AGI, which doesn’t reveal any fundamentally new ideas about the nature of intelligence or turn up any “unknown u... | https://www.lesswrong.com/posts/YTq4X6inEudiHkHDF/prosaic-ai-alignment |
# New safety research agenda: scalable agent alignment via reward modeling
Jan Leike and others from the DeepMind safety team have released a new [](https://medium.com/@deepmindsafetyresearch/scalable-agent-alignment-via-reward-modeling-bf4ab06dfd84) [research agenda](https://arxiv.org/abs/1811.07871) on reward learni... | https://www.lesswrong.com/posts/HBGd34LKvXM9TxvNf/new-safety-research-agenda-scalable-agent-alignment-via |
# Preschool: Much Less Than You Wanted To Know
Response to (Scott Alexander): [Preschool: Much More Than You Wanted to Know](http://slatestarcodex.com/2018/11/13/preschool-much-more-than-you-wanted-to-know/)
Previously (Here): [The Case Against Education](https://thezvi.wordpress.com/2018/04/15/the-case-against-educa... | https://www.lesswrong.com/posts/mYk4uTrMFqf6R5a3W/preschool-much-less-than-you-wanted-to-know |
# Incorrect hypotheses point to correct observations
**1\. The Consciousness Researcher and Out-Of-Body Experiences**
In his book _[Consciousness and the Brain](https://smile.amazon.com/Consciousness-Brain-Deciphering-Codes-Thoughts-ebook/dp/B00DMCVXO0/)_, cognitive neuroscientist Stansilas Dehaene writes about scien... | https://www.lesswrong.com/posts/MPj7t2w3nk4s9EYYh/incorrect-hypotheses-point-to-correct-observations |
# EA Funds: Long-Term Future fund is open to applications until November 24th (this Saturday)
I am one of the Fund Managers for the Long-Term Future fund and we are doing a very short round of applications to help donors get a sense of what projects we are likely to fund before giving season:
> Have an x-risk or far-... | https://www.lesswrong.com/posts/Re2EDpXTam7c8GmZM/ea-funds-long-term-future-fund-is-open-to-applications-until |
# Suggestion: New material shouldn't be released too fast
Three sequences on AI Safety were recently released. I'd suggest that in general it is probably better when new material isn't released that fast. Those sequences are quite technical so each post requires a reasonable amount of time to be invested. This means:
... | https://www.lesswrong.com/posts/JZSakuAZZo46vu3Sn/suggestion-new-material-shouldn-t-be-released-too-fast |
# Iteration Fixed Point Exercises
This is the third of three sets of fixed point exercises. The first post in this sequence is [here](https://www.lesswrong.com/posts/mojJ6Hpri8rfzY78b/fixed-point-theorems), giving context.
Note: Questions 1-5 form a coherent sequence and questions 6-10 form a separate coherent sequen... | https://www.lesswrong.com/posts/9a2asxypuNjCmga3p/iteration-fixed-point-exercises |
# The Semantic Man
Irving Lee discusses the goals of General Semantics, one of the historical movements that underlies intellectual currents expressed here now on LessWrong.
>I want to speak very briefly to you. There is one small notion that I should like to talk about rather briefly. It was in 1946, I remember al... | https://www.lesswrong.com/posts/HSYGRnTnvW3ieGnpr/the-semantic-man |
# Review: Artifact
Epistemic Status: Alpha tester
Bottom Line: If you are willing to devote the time and attention to a deep strategic game, Artifact will reward you handsomely. I highly recommended those who like such experiences to make the time. If you are not willing to devote the time and attention, you will lik... | https://www.lesswrong.com/posts/DTravCjZCWbbQC8YG/review-artifact |
# If You Want to Win, Stop Conceding
_Author's note: This is the first in what I suppose might be a series of posts with respect to things I learned from playing traditional games competitively that I think might have broader applications. I write this as a private individual, not on behalf of CFAR or any other organi... | https://www.lesswrong.com/posts/gvnBkE5eFWnLQYA9f/if-you-want-to-win-stop-conceding |
# Speculative Evopsych, Ep. 1
(cw death, religion, suicide, evolutionary psychology, shameless tongue-in-cheek meta-contrarianism)
I have a passing interest in biology, so I recently bought some fruit flies to run experiments on. I did two things to them. First, I bred them for intelligence. The details are kinda bor... | https://www.lesswrong.com/posts/mooQ5QkBsLJBGgfbh/speculative-evopsych-ep-1 |
# Approval-directed agents
_Note: This is the first post from **part two: basic intuitions** of [the sequence on iterated amplification](https://www.alignmentforum.org/s/EmDuGeRw749sD3GKd). The second part of the sequence outlines the basic intuitions that motivate iterated amplification. I think that these intuitions... | https://www.lesswrong.com/posts/7Hr8t6xwuuxBTqADK/approval-directed-agents-1 |
# LW Update 2018-11-22 – Abridged Comments
Haven't posted an update in a while. We've been working on a mixture of bug fixes, small features to support the new AI Alignment Forum, as well as some longer-term, larger features that aren't quite ready yet.
Abridged Comments
-----------------
The biggest change that sho... | https://www.lesswrong.com/posts/ckWgCF8CX32JhJHvu/lw-update-2018-11-22-abridged-comments |
# On MIRI's new research directions
Nate Soares has written up a post that discusses MIRI's new research directions: a mix of reasons why we're pursuing them, reasons why we're pursuing them the way we are, and high-level comparisons to Agent Foundations.
Read the full (long!) post [here](https://intelligence.org/201... | https://www.lesswrong.com/posts/SZ2h2Yte9pDi95Hnf/on-miri-s-new-research-directions |
# Oversight of Unsafe Systems via Dynamic Safety Envelopes
Idea
----
I had an idea for short-term, non-superhuman AI safety that I recently wrote up and ̶w̶i̶l̶l̶ ̶b̶e̶ ̶p̶o̶s̶t̶i̶n̶g̶ [have now posted](https://arxiv.org/abs/1811.09246) on Arxiv. This post serves to introduce the idea, and request feedback from a mo... | https://www.lesswrong.com/posts/frMdaZGtpRmEe26Wu/oversight-of-unsafe-systems-via-dynamic-safety-envelopes |
# What if people simply forecasted your future choices?
tldr: If you could have a team of smart forecasters predicting your future decisions & actions, they would likely improve them in accordance with your epistemology. This is a very broad method that's less ideal than more reductionist approaches for specific thing... | https://www.lesswrong.com/posts/DCkHSrD53Methoxu6/what-if-people-simply-forecasted-your-future-choices |
# Upcoming: Open Questions
_(note: everything in this post is still subject to change)_
I wanted to give people some sense of our next priorities (and why you've seen fewer public-facing features lately). The LessWrong team's current project is Open Questions (or "Q&A" for short)
This will be a new format for posts,... | https://www.lesswrong.com/posts/8XXjxjED6poddys9y/upcoming-open-questions |
# deluks917 on Online Weirdos
Cross-posted from [Putanumonit](https://putanumonit.com/2018/11/20/deluks917-on-online-communities/).
This is part 1 of an interview with [deluks917](https://www.lesswrong.com/users/deluks917). Stay tuned for part 2 in which we talk about Buddhism and peanut butter, and part 3 which ... | https://www.lesswrong.com/posts/4FRGmHgdsP9ds9SBf/deluks917-on-online-weirdos |
# Four factors that moderate the intensity of emotions
_Epistemic status: influenced by my study of emotions over the years, but primarily based on personal observation. Presented in the spirit of "here's a model, judge for yourself whether you think it's any good."_
_If you’re short on time or just skimming, I sugge... | https://www.lesswrong.com/posts/esJWfrGY89QrhGQbo/four-factors-that-moderate-the-intensity-of-emotions |
# Fixed Point Discussion
**Warning:** This post contains some important spoilers for [Topological Fixed Point Exercises](https://www.lesswrong.com/s/5WF3wmwvxX9TEbFXf/p/svE3S6NKdPYoGepzq), [Diagonalization Fixed Point Exercises](https://www.lesswrong.com/s/5WF3wmwvxX9TEbFXf/p/FZkLa3GRLW97fpknG), and [Iteration Fixed P... | https://www.lesswrong.com/posts/mvqmY9MQ3qf88xRuM/fixed-point-discussion |
# Values Weren't Complex, Once.
The central argument of this post is that human values are only complex because all the obvious constraints and goals are easily fulfilled. The resulting post-optimization world is deeply confusing, and leads to noise as the primary driver of human values. This has worrying implications... | https://www.lesswrong.com/posts/AMF5cRBjmMeeHB6Rf/values-weren-t-complex-once |
# Approval-directed bootstrapping
Approval-directed behavior works best when the overseer is very smart. Where can we find a smart overseer?
One approach is _bootstrapping_. By thinking for a long time, a weak agent can oversee an agent (slightly) smarter than itself. Now we have a slightly smarter agent, who can ove... | https://www.lesswrong.com/posts/6x7oExXi32ot6HjJv/approval-directed-bootstrapping |
# Humans Consulting HCH
_(See also:_ _[strong HCH](https://medium.com/ai-control/strong-hch-bedb0dc08d4e).)_
Consider a human Hugh who has access to a question-answering machine. Suppose the machine answers question Q by perfectly imitating how Hugh would answer question Q, _if_ _Hugh had access to the question-answe... | https://www.lesswrong.com/posts/NXqs4nYXaq8q6dTTx/humans-consulting-hch |
# Status model
\[Epistemic status: My best guess.\]
Following [a conversation](https://www.lesswrong.com/posts/BktRFP2jtYLJvFLEg/in-praise-of-heuristics#y7tJsFAgKj3CMwvLZ) on a previous post, I decided to do some research into the community’s thoughts on status. The community talks about status a lot and I spent a fe... | https://www.lesswrong.com/posts/v6hh8E8Qq8WCJMGcG/status-model |
# Alignment Newsletter #34
Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through the [database](https://docs.google.com/spreadsheets/d/1PwWbWZ6FPqAgZWOoOcXM8N_tUCuxpEyMbN1NYYC02aM/edit?usp=sharing) of all ... | https://www.lesswrong.com/posts/ZFT78ezD2yxLjo6QM/alignment-newsletter-34 |
# Bodega Bay: workshop
What was the workshop like? Well there was really amazing carpet. I don’t know how relevant this is, but a lot of people commented on it.
Another thing that was salient to me—and many of the participants—was these things called ‘back jacks’. They are like adjustable chairs, without arms or legs... | https://www.lesswrong.com/posts/w2iRiGw9wDrPNFv6Q/bodega-bay-workshop |
# Is Science Slowing Down?
*\[This post was up a few weeks ago before getting taken down for complicated reasons. They have been sorted out and I’m trying again.\]*
Is scientific progress slowing down? I recently got a chance to attend a conference on this topic, centered around a paper by [Bloom, Jones, Reenen & Web... | https://www.lesswrong.com/posts/v7c47vjta3mavY3QC/is-science-slowing-down |
# MIRI’s 2018 Fundraiser
(Crossposted from the [MIRI blog](https://intelligence.org/2018/11/26/miris-2018-fundraiser/))
**Update Nov. 28**: We’ve updated this post to announce that Haskell developer Edward Kmett [has joined the MIRI team](https://intelligence.org/2018/11/28/miris-newest-recruit-edward-kmett/)! Additi... | https://www.lesswrong.com/posts/4ktkt6LYryNuwAc9b/miri-s-2018-fundraiser |
# How democracy ends: a review and reevaluation
Last month I attended a talk by David Runciman, the author of a recent book called _How Democracy Ends_. I was prepared for outrage-stirring and pearl-clutching, but was pleasantly surprised by the quality of his arguments, which I’ve summarised below, along with my own ... | https://www.lesswrong.com/posts/HAYvEBdqK3bzXvDBo/how-democracy-ends-a-review-and-reevaluation |
# CFAR reunion Europe
Tl;dr: there will be a CFAR reunion in Europe in spring (in addition to the usual one in August in Cali), March 29 - April 1, 2019, in [Czech mountains](https://en.wikipedia.org/wiki/Jizera_Mountains). Tickets are 150 EUR, first come, first serve. [Register here!](https://docs.google.com/forms/d/... | https://www.lesswrong.com/events/DCSrLcYcbpifjGPtc/cfar-reunion-europe |
# Summary: Surreal Decisions
This post summarises a [paper](https://www.eddykemingchen.net/uploads/4/6/1/3/46137503/surreal_decisions_ppr.pdf) by Eddy Chen and Daniel Rubio on using Surreal numbers to resolve problems of [Infinite Ethics](https://nickbostrom.com/ethics/infinite.pdf). Future posts will argue that surre... | https://www.lesswrong.com/posts/MtYDr7zyicq4J7rWe/summary-surreal-decisions |
# Corrigibility
(_Warning: rambling._)
I would like to build AI systems which help me:
* Figure out whether I built the right AI and correct any mistakes I made
* Remain informed about the AI’s behavior and avoid unpleasant surprises
* Make better decisions and clarify my preferences
* Acquire resources and ... | https://www.lesswrong.com/posts/fkLYhTQteAu5SinAc/corrigibility |
# Winter Solstice 2018 Roundup
Many LessWrong communities are holding Winter Solstice holiday events. What exactly that entails varies from community-to-community, but often includes:
* A ceremony commemorating humanity's progress over the millenia (often involving speeches and singalong songs)
* Meetups and part... | https://www.lesswrong.com/posts/bFn4YYSrKFoAk3sZY/winter-solstice-2018-roundup |
# Double-Dipping in Dunning--Kruger
Originally posted at [sandymaguire.me](http://sandymaguire.me/blog/double-dipping/).
> I have to get you to drop modesty and say to yourself, "Yes, I would like to
> do first-class work." Our society frowns on people who set out to do really
> good work. You're not supposed to; luc... | https://www.lesswrong.com/posts/qS2dnpeQyppetgXqv/double-dipping-in-dunning-kruger |
# Bounded Oracle Induction
**Introduction:**
-----------------
Because [logical induction](https://intelligence.org/files/LogicalInduction.pdf) relies on the Brouwer fixed-point theorem, and [reflective oracles](https://arxiv.org/pdf/1508.04145.pdf) rely on the Kakutani fixed-point theorem which Brouwer is a spec... | https://www.lesswrong.com/posts/MgLeAWSeLbzx8mkZ2/bounded-oracle-induction |
# Genetically Modified Humans Born (Allegedly)
I realize normally we don't talk about the news or hot-button issues, but this is of sufficiently high importance I am posting anyway.
There is a link from Nature News on Monday [here](https://www.nature.com/articles/d41586-018-07545-0). There is a link from MIT Technolo... | https://www.lesswrong.com/posts/3Smcn7mNoTNkQFpNb/genetically-modified-humans-born-allegedly |
# Stabilize-Reflect-Execute
You've recently joined a major organization in a senior management role. How can you organize your plans?
One simple way to think about them is with what can be called the "Stabilize-Reflect-Execute" cycle.
**Stabilize**
You first check if there are any urgent issues and address them imm... | https://www.lesswrong.com/posts/3bcjPhwCnnMonc4EP/stabilize-reflect-execute |
# Counterintuitive Comparative Advantage
This has been sitting in my drafts folder since 2011. Decided to post it today given the [recent post about Dunning—Kruger](https://www.lesswrong.com/posts/qS2dnpeQyppetgXqv/double-dipping-in-dunning-kruger) and related discussions.
The standard rationalist answer when someo... | https://www.lesswrong.com/posts/DN6q3SmgTrJgnRYWb/counterintuitive-comparative-advantage |
# Madison Solstice Gathering
Some far-flung rationalists will be in Madison for the holidays and want to bring secular solstice to Wisconsin! It will be a potluck, so bring something to share if you can. We'll eat food and sing some 'traditional' songs (lyrics will be sent out beforehand) and maybe go outside for a bi... | https://www.lesswrong.com/events/CXdQtazqT6CuNv7pj/madison-solstice-gathering |
# Hyperreal Brouwer
_(This post was originally published on Oct 6th 2017, and has been temporarily brought forwarded as part of the AI Alignment Forum launch sequence on fixed points.)_
This post explains how to view Kakutani's fixed point theorem as a special case of Brouwer's fixed point theorem with hyperreal numb... | https://www.lesswrong.com/posts/5bd75cc58225bf06703754e4/hyperreal-brouwer |
# The Ubiquitous Converse Lawvere Problem
_(This post was originally published on Oct 20th 2017, and is 1 of 4 posts brought forwarded today as part of the AI Alignment Forum launch sequence on fixed points.)_
In this post, I give a stronger version of the open question presented [here](https://agentfoundations.org/i... | https://www.lesswrong.com/posts/5bd75cc58225bf06703753b9/the-ubiquitous-converse-lawvere-problem |
# Reflective oracles as a solution to the converse Lawvere problem
_(This post was originally published on Nov 30th 2017, and is 1 of 4 posts brought forwarded today as part of the AI Alignment Forum launch sequence on fixed points.)_
**1 Introduction**
Before the work of Turing, one could justifiably be skeptical o... | https://www.lesswrong.com/posts/5bd75cc58225bf067037550d/reflective-oracles-as-a-solution-to-the-converse-lawvere |
# Formal Open Problem in Decision Theory
_(This post was originally published on March 31st 2017, and has been brought forwarded as part of the AI Alignment Forum launch sequence on fixed points.)_
In this post, I present a new formal open problem. A positive answer would be valuable for decision theory research. A n... | https://www.lesswrong.com/posts/5bd75cc58225bf06703753a9/formal-open-problem-in-decision-theory |
# Moving Factward
In the legendarium of J.R.R. Tolkien, the land of the gods is known as "The Uttermost West." For the world was originally created flat, and the gods took the westernmost region of this flat world for their dwelling place.
On our globe, of course, there is no westernmost point. And yet it is stil... | https://www.lesswrong.com/posts/wX26kvBTRNATwaYCY/moving-factward |
# Iterated Distillation and Amplification
This is a guest post summarizing Paul Christiano’s proposed scheme for training machine learning systems that can be robustly aligned to complex and fuzzy values, which I call Iterated Distillation and Amplification (IDA) here. IDA is [notably similar](https://ai-alignment.com... | https://www.lesswrong.com/posts/HqLxuZ4LhaFhmAHWk/iterated-distillation-and-amplification-1 |
# Tentatively considering emotional stories (IFS and “getting into Self”)
I’ve recently been getting a lot out of the psychotherapy model of [Internal Family Systems](https://selfleadership.org/about-internal-family-systems.html), as described in [this book](https://www.amazon.com/dp/B00452V8EG/). I just wrote [a comm... | https://www.lesswrong.com/posts/8ZrDp7d5t4fwKXfj3/tentatively-considering-emotional-stories-ifs-and-getting |
# Overconfident talking down, humble or hostile talking up
A Distribution of Knowlege
--------------------------
If one were to make a distribution of the amount of knowledge different people have about, say, macroeconomics, I would suspect the distribution to be somewhat lognormal; they would have tails to both ends... | https://www.lesswrong.com/posts/m4a8jFjWjx7Q5k24r/overconfident-talking-down-humble-or-hostile-talking-up |
# Taking vitamin D3 with K2 in the morning
_Epistemological status: I studied bioinformatics but I'm no domain expert and_ layout _my current view on the issue based on the facts I found in the literature. This is not health advice._
**tl,dr:**
There’s a strong evolutionary selection for producing enough Vitamin... | https://www.lesswrong.com/posts/c5aycbSsSc38XWPEc/taking-vitamin-d3-with-k2-in-the-morning |
# Intuitions about goal-directed behavior
One broad argument for AI risk is the Misspecified Goal argument:
> **The Misspecified Goal Argument for AI Risk:** Very intelligent AI systems will be able to make long-term plans in order to achieve their goals, and if their goals are even slightly misspecified then the AI ... | https://www.lesswrong.com/posts/DfcywmqRSkBaCB6Ma/intuitions-about-goal-directed-behavior |
# Anyone use the "read time" on Post Items?
Someone recently mentioned that they found the "read time" on home page post-items to be more of annoying clutter than a helpful indicator of whether to read something.
It occurred to me that I don't personally use them, and in general it's important for each UI element to ... | https://www.lesswrong.com/posts/PHL8qvX4QpM6sXKfK/anyone-use-the-read-time-on-post-items |
# Benign model-free RL
In my [last post](https://medium.com/ai-control/directions-and-desiderata-for-ai-control-b60fca0da8f4), I described three research areas in AI control that I see as central: reward learning, robustness, and deliberation.
In this post I argue that these three pieces may be *sufficient* to get a ... | https://www.lesswrong.com/posts/PRaxzmDJdvie46ahL/benign-model-free-rl-1 |
# Coherence arguments do not entail goal-directed behavior
One of the most pleasing things about probability and expected utility theory is that there are many _coherence arguments_ that suggest that these are the “correct” ways to reason. If you deviate from what the theory prescribes, then you must be executing a _d... | https://www.lesswrong.com/posts/NxF5G6CJiof6cemTw/coherence-arguments-do-not-entail-goal-directed-behavior |
# Book Review - Other Minds: The Octopus, the Sea, and the Deep Origins of Consciousness
*In this post:*
* *A true-false test about octopuses*
* *What is it like to be an octopus?*
* *An exercise in updating on surprising facts*
* *Experiments related to animal suffering and consciousness*
* *The evolution of a... | https://www.lesswrong.com/posts/QffdgbTEGNRoXMews/book-review-other-minds-the-octopus-the-sea-and-the-deep |
# Peanut Butter
This is part 2 of my interview with [deluks917](https://deluks917.wordpress.com/). In [part 1 we discussed](https://putanumonit.com/2018/11/20/deluks917-on-online-communities/) online weirdos, free thinkers, and Uyghurs. Stay tuned for part 3 on archetypes, the will to knowledge and Moana.
We talk abo... | https://www.lesswrong.com/posts/gRcBten4TBmCNduMZ/peanut-butter |
# Alignment Newsletter #35
Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through this [spreadsheet](https://docs.google.com/spreadsheets/d/1PwWbWZ6FPqAgZWOoOcXM8N_tUCuxpEyMbN1NYYC02aM/edit?usp=sharing) of ... | https://www.lesswrong.com/posts/Pe3aqWXJWLHoB6vc4/alignment-newsletter-35 |
# Open and Welcome Thread December 2018
If it’s worth saying, but not worth its own post, then it goes here.
Also, if you are new to LessWrong and want to introduce yourself, this is the place to do it. Personal stories, anecdotes, or just general comments on how you found us and what you hope to get from the site an... | https://www.lesswrong.com/posts/DYCNuqMpCfQtPcPSo/open-and-welcome-thread-december-2018 |
# Playing Politics
_Epistemic Status: Guesses Based on Personal Experience_
Lately I’ve been going through a family of learning experiences in the world of _how to get things done cooperatively_. It’s hard for me. Even very basic things in this area have been stumping me, overwhelming me, leaving me way more tired ... | https://www.lesswrong.com/posts/wQofcJZzySGLqiw5N/playing-politics |
# Factored Cognition
_Note: This post (originally published [here](https://ought.org/presentations/factored-cognition-2018-05)) is the transcript of a presentation about a project worked on at the non-profit Ought. It is included in the sequence because it contains a very clear explanation of some of the key ideas beh... | https://www.lesswrong.com/posts/DFkGStzvj3jgXibFG/factored-cognition |
# Why we need a *theory* of human values
There have been multiple *practical* suggestions for methods about how we should extract the values of a given human. Here are four common classes of such methods:
* Methods that put high weight on human (bounded) quasi-rationality, or revealed preferences. For example, we can... | https://www.lesswrong.com/posts/zvrZi95EHqJPxdgps/why-we-need-a-theory-of-human-values |
# Transhumanism as Simplified Humanism
_This essay was originally posted in 2007._
* * *
[Frank Sulloway](http://www.robertboynton.com/?art_id=119) once said: “Ninety-nine per cent of what Darwinian theory says about human behavior is so obviously true that we don’t give Darwin credit for it. Ironically, psychoanaly... | https://www.lesswrong.com/posts/Aud7CL7uhz55KL8jG/transhumanism-as-simplified-humanism |
# On Rationalist Solstice and Epistemic Caution
Since 2011, some LessWrong folk have observed the winter solstice, as a holiday ritual celebrating human achievement in the face of a confusing, often terrifying world.
I've written in the past about the [potential value, and danger](https://www.lesswrong.com/posts/rijo... | https://www.lesswrong.com/posts/CGkZEQdeBZZXbBT7o/on-rationalist-solstice-and-epistemic-caution |
# Hell Must Be Destroyed
\[Originally published in [Qualia Computing](https://qualiacomputing.com/2018/12/03/hell-must-be-destroyed/)\]
_Singer called the movement that grew up around him “effective altruism”, and its rallying cry was that one ought to spend every ounce of one’s energy doing whatever most relieves hu... | https://www.lesswrong.com/posts/pPssXmTGEin9Ri3An/hell-must-be-destroyed |
# COEDT Equilibria in Games
From Jessica's [earlier post about conditional oracles](https://www.alignmentforum.org/posts/Rcwv6SPsmhkgzfkDw/edt-solves-5-and-10-with-conditional-oracles), the question was asked: what is the equilibrium concept for games with more than one COEDT agent?
This post will (partially) answer ... | https://www.lesswrong.com/posts/4MLpRxz7ZoX8YXSY3/coedt-equilibria-in-games |
# Trivial Inconvenience Day (December 9th at 12 Noon PST)
Hi everybody, it's been 9 months since [I wrote my report on Trivial Inconvenience Day](https://www.lesswrong.com/posts/9GX49j5DMtasxTbwd/trivial-inconvenience-day-retrospective). I wasn't quite sure when to run a sequel event. I wanted to give people an opport... | https://www.lesswrong.com/posts/9hYaPDPJFSqfBv67y/trivial-inconvenience-day-december-9th-at-12-noon-pst |
# Worth keeping
_(Epistemic status: quick speculation which matches my intuitions about how social things go, but which I hadn’t explicitly described before, and haven’t checked.)_
If your car gets damaged, should you invest more or less in it going forward? It could go either way. The car needs more investment to be... | https://www.lesswrong.com/posts/pDcQKJXBLJTHBDbhK/worth-keeping |
# Is cognitive load a factor in community decline?
This is speculation; I had the thought and then ran in to trouble disentangling the question I am trying to answer from other research on a different question, and also the sources I know about are not conveniently available to me. Ideally I can either get a swift neg... | https://www.lesswrong.com/posts/KMvgXcrntHoNaPj7i/is-cognitive-load-a-factor-in-community-decline |
# Transhumanists Don't Need Special Dispositions
_This essay was originally posted in 2007._
* * *
I have [claimed](https://www.lesswrong.com/posts/Aud7CL7uhz55KL8jG/transhumanism-as-simplified-humanism) that transhumanism arises strictly from love of life. A bioconservative humanist says that it is good to save so... | https://www.lesswrong.com/posts/cq4DsXzGRXJBmYuyB/transhumanists-don-t-need-special-dispositions |
# LW Update 2018-12-06 – Table of Contents and Q&A
After a couple months of work, we're ready for a significant update to LessWrong:
* Post Page Redesign
* Open Questions (aka Q&A)
* Table of Contents (aka ToC)
* Comment Guidelines
Post Page Redesign
==================
The first thing you probably noticed i... | https://www.lesswrong.com/posts/mrGeJ4Wt66PxN9RQh/lw-update-2018-12-06-table-of-contents-and-q-and-a |
# Is the human brain a valid choice for the Universal Turing Machine in Solomonoff Induction?
I've recently been thinking about Solomonoff induction, and in particular the [free choice of Universal Turing Machine](https://www.lesswrong.com/posts/fC248GwrWLT4Dkjf6/open-problems-related-to-solomonoff-induction).
One va... | https://www.lesswrong.com/posts/3xnkw6JkQdwc8Cfcf/is-the-human-brain-a-valid-choice-for-the-universal-turing |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.