text stringlengths 300 320k | source stringlengths 52 154 |
|---|---|
# What I talk about when I talk about AI x-risk: 3 core claims I want machine learning researchers to address.
Recently, as PCSOCMLx, I (co-)hosted a session with the goal of explaining, debating, and discussing what I view as "the case for AI x-risk". Specifically, my goal was/is to make the case for the "out-of-cont... | https://www.lesswrong.com/posts/bJdaB2Mz4mBvwFBeb/what-i-talk-about-when-i-talk-about-ai-x-risk-3-core-claims-1 |
# A list of good heuristics that the case for AI x-risk fails
I think one reason machine learning researchers don't think AI x-risk is a problem is because they haven't given it the time of day. **And on some level, they may be right in not doing so!**
We all need to do meta-level reasoning about what to spend our ti... | https://www.lesswrong.com/posts/bd2K3Jdz82csjCFob/a-list-of-good-heuristics-that-the-case-for-ai-x-risk-fails |
# Open & Welcome Thread - December 2019
* If it’s worth saying, but not worth its own post, here's a place to put it.
* You can also make a [shortform post](http://www.lesslong.com/).
* And, if you are new to LessWrong, here's the place to introduce yourself.
* Personal stories, anecdotes, or just gene... | https://www.lesswrong.com/posts/X4nYiTLGxAkR2KLAP/open-and-welcome-thread-december-2019 |
# MIRI’s 2019 Fundraiser
(Crossposted from the [MIRI blog](https://intelligence.org/2019/12/02/miris-2019-fundraiser/))
**MIRI’s 2019 fundraiser** is now live, December 2–31!
Over the past two years, huge donor support has helped us double the size of our AI alignment research team. Hitting our $1M fundraising goal ... | https://www.lesswrong.com/posts/XJiNtvxoiLCpBn6FH/miri-s-2019-fundraiser |
# Russian x-risks newsletter #2, fall 2019
Russia could be seen as an x-risks wonderland, with exploding nuclear facilities and bioweapons storage sites, doomsday weapons, frequent asteroid impacts, Siberian volcanic traps, Arctic methane bomb, military AI programs, and crazy scientists wanting to drill to the Earth’s... | https://www.lesswrong.com/posts/v8TH7Ax6vYBicfgCM/russian-x-risks-newsletter-2-fall-2019 |
# Long-lasting Effects of Suspensions?
I recently read "The School to Prison Pipeline: Long-Run Impacts of School Suspensions on Adult Crime" (Bacher-Hicks et. al. 2019, [pdf](https://www.jefftk.com/bacherhicks2019.pdf), [via Rob Wiblin](https://www.facebook.com/robert.wiblin/posts/872782182395)) which argues that a p... | https://www.lesswrong.com/posts/y6BvzBfQMN2XiMEwq/long-lasting-effects-of-suspensions |
# Symbiotic Conflicts
A meme is a self-replicating pattern of information. Some change. Some survive. Some die out.
The most virulent memes often reproduce themselves via their own inverses. Talking about how Flat Earthers are wrong increases awareness of the Flat Earther meme. Increasing awareness of the Flat Earthe... | https://www.lesswrong.com/posts/Myi7eyEGTPiPa3pbm/symbiotic-conflicts |
# A letter on optimism about human progress
*This open letter was originally posted on Letter.wiki and is part of a [longer conversation](https://letter.wiki/conversation/202) with Andrew Glover about sustainability and progress.*
Dear Andrew,
Thanks for a thoughtful reply. Reading over it, it seems the biggest diff... | https://www.lesswrong.com/posts/rDYgsng7rdBJnY2wL/a-letter-on-optimism-about-human-progress |
# In which ways have you self-improved that made you feel bad for not having done it earlier?
It can be a decision, a skill, a habit, etc.
Can be because the improvement was very valuable, obvious in insight, a moral imperative, or any other reason.
Note that I'm specifically looking for things *that made you feel b... | https://www.lesswrong.com/posts/KcDQXiAd7cCnsPA75/in-which-ways-have-you-self-improved-that-made-you-feel-bad |
# "Fully" acausal trade
[Acausal trade](https://wiki.lesswrong.com/wiki/Acausal_trade) happens when two agents manage to reach a deal with each other, despite not being able to interact causally (and, in some cases, not being sure the other one exists). Consider, for example, the prisoner's dilemma played against anot... | https://www.lesswrong.com/posts/MHHzLfAQBZzieGBjq/fully-acausal-trade |
# [AN #76]: How dataset size affects robustness, and benchmarking safe exploration by measuring constraint violations
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.g... | https://www.lesswrong.com/posts/SXoHj7DTAjAsfJrcs/an-76-how-dataset-size-affects-robustness-and-benchmarking |
# Paper-Reading for Gears
Lesswrong has a fair bit of advice on [how to evaluate](https://www.lesswrong.com/posts/gxbGKa2AnQsrn3Gni/how-do-you-assess-the-quality-reliability-of-a-scientific) the claims made in scientific papers. Most of this advice seems to focus on a single-shot use case - e.g. a paper claims that ta... | https://www.lesswrong.com/posts/TPjbTXntR54XSZ3F2/paper-reading-for-gears |
# Recent Progress in the Theory of Neural Networks
It's common wisdom that neural networks are basically ["matrix multiplications that nobody understands"](https://twitter.com/ESYudkowsky/status/1083434318142173184) , impenetrable to theoretical analysis, which have achieved great results largely through trial-and-err... | https://www.lesswrong.com/posts/KrQvZM8uFjSTJ7hq3/recent-progress-in-the-theory-of-neural-networks-1 |
# Karate Kid and Realistic Expectations for Disagreement Resolution
There’s an [essay](https://www.cracked.com/article_18544_how-the-karate-kid-ruined-modern-world.html) that periodically feels deeply relevant to a situation:
> Someday I want to write a self-help book titled “F*k The Karate Kid: Why Life is So Much H... | https://www.lesswrong.com/posts/W8vSrHAM9qoWdzFoP/karate-kid-and-realistic-expectations-for-disagreement |
# What are some non-purely-sampling ways to do deep RL?
Conventionally in machine learning, if you want to learn to minimize some loss or maximize some expected return, you do so by sampling a bunch of losses/rewards and training on those. Since the model only ever sees the loss or reward function through the lens of ... | https://www.lesswrong.com/posts/Ca3sCRGfWvXvYC5YC/what-are-some-non-purely-sampling-ways-to-do-deep-rl |
# Elementary Statistics
Our elementary school has a directory listing kids and parents, and since we live in the future it's a spreadsheet, which means I can count things. A typical family at this K-5 school has one child enrolled (76%). The child has two parents (96%) with different last names (59%), but they share a... | https://www.lesswrong.com/posts/6S7YnJuvN4TEXAk3T/elementary-statistics |
# Seeking Power is Often Convergently Instrumental in MDPs
**Edit, 5/16/23: I think this post is beautiful, correct in its narrow technical claims, and practically irrelevant to alignment. This post presents a cripplingly unrealistic picture of the role of reward functions in reinforcement learning. I expect this post... | https://www.lesswrong.com/posts/6DuJxY8X45Sco4bS2/seeking-power-is-often-convergently-instrumental-in-mdps |
#
Oracles: reject all deals - break superrationality, with superrationality
tl;dr Acausal trade between Oracles is a problem even if our Oracle is a causal decision theorist. To avoid this, we need an AFDT Oracle - an Oracle that reasons partially like a functional decision theorist, partially like a causal decision ... | https://www.lesswrong.com/posts/6XCTppoPAMdKCPFb4/oracles-reject-all-deals-break-superrationality-with-1 |
# Reading list: Starting links and books on studying ontology and causality
**_(This article will be revised, as I add more to it. This is a living document.)_**
I recently read gwern's excellent "[Why Correlation Usually ≠ Causation](https://www.gwern.net/Causality)" notes, and, like any good reading, felt a profoun... | https://www.lesswrong.com/posts/bHR5aqEQPPTG9JvEY/reading-list-starting-links-and-books-on-studying-ontology |
# LW For External Comments?
There are several blogs which (a) cross-post to LessWrong and (b) have a relatively quiet comment section on their "canonical" sites: Meteuphoric ( [canonical](https://meteuphoric.com/2019/11/10/pieces-of-time/), [LW](https://www.lesswrong.com/posts/BuqaN3T6gwcj7bhrb/pieces-of-time)), Don't... | https://www.lesswrong.com/posts/Y3apSc55ZE9wc8Df8/lw-for-external-comments |
# LW Team Updates - December 2019
This is the [once-monthly updates post](https://www.lesswrong.com/s/GTpv3mjKqFot9pxT3) for LessWrong team activities and announcements.
Summary
=======
In the past month we rolled out floating comment guidelines and launched the inaugural Lesswrong 2018 review. Work has continued o... | https://www.lesswrong.com/posts/rYLfaj2nSRJWEnQpQ/lw-team-updates-december-2019 |
# Values, Valence, and Alignment
I have previously advocated for finding a mathematically precise theory for [formally approaching AI alignment](https://www.lesswrong.com/posts/7dvDgqvqqziSKweRs/formally-stating-the-ai-alignment-problem). Most recently I couched this [in terms of predictive coding](https://www.lesswro... | https://www.lesswrong.com/posts/ALvnz3DrjHwmLG29F/values-valence-and-alignment |
# Tapping Out In Two
I'm one of those people who feels personally called out by xkcd's "[Duty Calls](https://xkcd.com/386/)" ("someone is wrong on the internet").
(Not as much as I used to. At some point I stopped reading most of the subreddits that I would argue on, partly for this reason, and Hacker News, for unrel... | https://www.lesswrong.com/posts/DctWqpGLZwTnrGdLT/tapping-out-in-two |
# Understanding “Deep Double Descent”
If you're not familiar with the double descent phenomenon, I think you should be. I consider double descent to be one of the most interesting and surprising recent results in analyzing and understanding modern machine learning. Today, Preetum et al. released a new paper, “[Deep Do... | https://www.lesswrong.com/posts/FRv7ryoqtvSuqBxuT/understanding-deep-double-descent |
# The Actionable Version of "Keep Your Identity Small"
(cross posted on my [roam](https://roamresearch.com/) [blog](https://roamresearch.com/#/app/hazard_blog/page/hYLCprCMM))
There's an old Paul Graham Essay, ["Keep Your Identity Small"](http://www.paulgraham.com/identity.html). It's short so it's worth it to read t... | https://www.lesswrong.com/posts/BXQsZmubkovJ76Ldo/the-actionable-version-of-keep-your-identity-small |
# Comment on Coherence arguments do not imply goal directed behavior
In [coherence arguments do not imply goal directed behavior](https://www.alignmentforum.org/s/4dHMdK5TLN6xcqtyc/p/NxF5G6CJiof6cemTw) Rohin Shah argues that a system's merely being at all model-able as an EU maximizer does not imply that it has "goal ... | https://www.lesswrong.com/posts/EnN7cm3KaRrEAuWfa/comment-on-coherence-arguments-do-not-imply-goal-directed |
# What is Abstraction?
Let's start with a few examples (borrowed from [here](https://www.lesswrong.com/posts/hLFD6qSN9MmQxKjG5/embedded-agency-via-abstraction)) to illustrate what we're talking about:
* We have a gas consisting of some huge number of particles. We throw away information about the particles themselv... | https://www.lesswrong.com/posts/wuJpYLcMEBz4kcgAn/what-is-abstraction-1 |
# The New Age of Social Engineering
Why have so many online social networks failed to form healthy communities, and instead gained notoriety as hostile spaces? I argue that the reason these platforms have failed is because they didn’t learn the lessons taught by the High Moderns when humans were first faced with the c... | https://www.lesswrong.com/posts/256rkLpg3CQ6KCbtr/the-new-age-of-social-engineering |
# Ungendered Spanish
Spanish has gramatical gender in a way English doesn't:
> una amiga ruidosa — *a loud (female) friend*
> un amigo ruidoso — *a loud (male) friend*
> unas amigas ruidosas — *some loud (female) friends*
> unos amigos ruidosos — *some loud (not-all-female) friends*
I remember when I was stu... | https://www.lesswrong.com/posts/rKXimnpLYfHg7Cskp/ungendered-spanish |
# Confabulation
When someone asks me why I did or said something I usually lie because the truthful answer is "I don't know". I literally don't know why I make >99% of my decisions. I think through none of these decisions rationally. It's usually some mixture of gut instinct, intuition, cultural norms, common sense an... | https://www.lesswrong.com/posts/ZvLPnJrgHxhkEyXKu/confabulation |
# Progress and preservation in IDA
# Overview
This post arose out of my attempts to understand IDA and ways it could fail. It might help you do the same and could provide useful vocabulary for discussing desiderata for IDA.
We want IDA to satisfy progress---decomposition should make answering questions easier---and ... | https://www.lesswrong.com/posts/di8H7rEAnzXC97Dvu/progress-and-preservation-in-ida |
# What determines the balance between intelligence signaling and virtue signaling?
Lately I've come to think of human civilization as largely built on the backs of intelligence and virtue signaling. In other words, civilization depends very much on the positive side effects of ([not necessarily conscious](https://www.... | https://www.lesswrong.com/posts/vA2Gd2PQjNk68ngFu/what-determines-the-balance-between-intelligence-signaling |
# Books on the zeitgeist of science during Lord Kelvin's time.
> There is nothing new to be discovered in physics now. All that remains is more and more precise measurement.
_(edit: looks like this question was based on a false impression of history, so nvm)_
That Lord Kelvin [apocryphal](https://www.lesswrong.com/p... | https://www.lesswrong.com/posts/wczp6eqwYRWX5mkxw/books-on-the-zeitgeist-of-science-during-lord-kelvin-s-time |
# The Review Phase
_LessWrong is currently doing a [major review of 2018](https://www.lesswrong.com/posts/qXwmMkEBLL59NkvYR/the-lesswrong-2018-review) — looking back at old posts and considering which of them have stood the test of time. Info about what features we added to the site for writing reviews is [in December... | https://www.lesswrong.com/posts/5DcJw3LHNimzezrdC/the-review-phase |
# What are the best arguments and/or plans for doing work in "AI policy"?
I'm looking to get oriented in the space of "AI policy": interventions that involve world governments (particularly the US government) and existential risk from strong AI.
When I hear people talk about "AI policy", my initial reaction is skepti... | https://www.lesswrong.com/posts/sH2pNHmaqqHQA2zzi/what-are-the-best-arguments-and-or-plans-for-doing-work-in |
# Long Bets by Confidence Level
If you want to make a [long-term bet](https://www.jefftk.com/p/long-term-bets) one of your options is to register your bet with the Long Now Foundation as a Long Bet. They have some rules, which are [roughly](http://longbets.org/rules/):
* Both parties put up the same amount, at leas... | https://www.lesswrong.com/posts/TTd2EJ2cSJ2DiJerr/long-bets-by-confidence-level |
# Is Rationalist Self-Improvement Real?
[Cross-posted from Putanumonit](https://putanumonit.com/2019/12/08/rationalist-self-improvement/) where the images show up way bigger. I don't know how to make them bigger on LW.
* * *
Basketballism
=============
Imagine that tomorrow everyone on the planet forgets the concep... | https://www.lesswrong.com/posts/8xLtE3BwgegJ7WBbf/is-rationalist-self-improvement-real |
# Bayesian examination
A few months ago, Olivier Bailleux, a Professor of computer science and reader of my book on Bayesianism, sent me an email. He suggested to apply some of the ideas of the book to examine students. He proposed **Bayesian examination**_._
I believe it to be a brilliant idea, which could have an i... | https://www.lesswrong.com/posts/8wzKawHmh4d3h2otw/bayesian-examination |
# Were vaccines relevant to 20th century US mortality improvements?
I heard a vaccine skeptic claim that “90% of the decline in infectious disease mortality in the 20th century in the US was due to factors other than vaccines.” I wondered, is that right?
My guess is yes—but at the same time, I think this is very misl... | https://www.lesswrong.com/posts/o6o2JDHpZwgdREE8f/were-vaccines-relevant-to-20th-century-us-mortality |
# [Review] On the Chatham House Rule (Ben Pace, Dec 2019)
*This is a brief review of* [*On the Chatham House Rule*](https://www.lesswrong.com/posts/sWof2zGexwwJ8Q4ND/on-the-chatham-house-rule) *by Scott Garrabrant.*
I tend to be very open about my thoughts and beliefs. However, I naturally am still discrete about a l... | https://www.lesswrong.com/posts/aE5q2Mb8zQNo4eKxy/review-on-the-chatham-house-rule-ben-pace-dec-2019 |
# [Review] Meta-Honesty (Ben Pace, Dec 2019)
*\[Edit: I'm going to re-write this at some point, I don't think I managed to say much of anything very clearly. I've removed the first section.\]*
The essay takes the same starting assumption as I do. Honesty is deadly important and something you work hard at. There are s... | https://www.lesswrong.com/posts/y4NxDJHLbdjicDYTF/review-meta-honesty-ben-pace-dec-2019 |
# Predictive coding = RL + SL + Bayes + MPC
*(Update much later (2021-06): I put in some minor updates and retractions below. Also, if I were writing this today I would probably have framed it as "here's how I disagree with predictive coding" rather than "here's a version of predictive coding that I like". Also, where... | https://www.lesswrong.com/posts/cfvBm2kBtFTgxBB7s/predictive-coding-rl-sl-bayes-mpc |
# Applications of Economic Models to Physiology?
Applying economic models to physiology seems really obvious. For instance:
* Surely the body uses price signals to match production to consumption of various metabolites. [Insulin as a price signal for glucose](https://www.lesswrong.com/posts/bNXdnRTpSXk9p4zmi/book-r... | https://www.lesswrong.com/posts/W6g9a7sFaRycm5FcJ/applications-of-economic-models-to-physiology |
# ToL: Foundations
_(These are the touched up notes from a class I took with CMU's [Kevin Kelly](http://www.andrew.cmu.edu/user/kk3n/homepage/kelly.html) this past semester on the Topology of Learning. Only partially optimized for legibility)_
Possible World Semantics
========================
Let W be the set of al... | https://www.lesswrong.com/posts/bFXkz9PsSgWaWe3BT/tol-foundations |
# ToL: Introduction
_(These are the touched up notes from a class I took with CMU's [Kevin Kelly](http://www.andrew.cmu.edu/user/kk3n/homepage/kelly.html) this past semester on the Topology of Learning. Only partially optimized for legibility)_
Feasibility Contextualism
=========================
A large amount of ph... | https://www.lesswrong.com/posts/s4RGYpADPhtbqaXPH/tol-introduction |
# ToL: The Topological Connection
_(These are the touched up notes from a class I took with CMU's [Kevin Kelly](http://www.andrew.cmu.edu/user/kk3n/homepage/kelly.html) this past semester on the Topology of Learning. Only partially optimized for legibility)_
Now that we've got the basic formalism, it's time to go fro... | https://www.lesswrong.com/posts/MbPrpAQ6smruK7jKJ/tol-the-topological-connection |
# ToL: This ONE WEIRD Trick to make you a GENIUS at Topology!
_(These are the touched up notes from a class I took with CMU's [Kevin Kelly](http://www.andrew.cmu.edu/user/kk3n/homepage/kelly.html) this past semester on the Topology of Learning. Only partially optimized for legibility)_
Time to introduce some new Topo... | https://www.lesswrong.com/posts/LBYthCbh4xefz4MjD/tol-this-one-weird-trick-to-make-you-a-genius-at-topology |
# Generalizing Experimental Results by Leveraging Knowledge of Mechanisms
In a [recent post (and papers)](https://www.lesswrong.com/posts/wwbrvumMWhDfeo652), Anders Huitfeldt and co-authors have discussed ways of achieving external validity in the presence of “effect heterogeneity.” These results are not immediately i... | https://www.lesswrong.com/posts/pbuDzeqHSXkkAdWY7/generalizing-experimental-results-by-leveraging-knowledge-of |
# Causal Abstraction Toy Model: Medical Sensor
*Author's Note: This post is a bunch of mathy research stuff with very little explanation of context. Other posts in this sequence will provide more context, but you might want to skip this one unless you're looking for mathy details.*
Suppose we have a medical sensor me... | https://www.lesswrong.com/posts/S8WZ2rav9BqFAZoRM/causal-abstraction-toy-model-medical-sensor |
# The Cybersecurity Dilemma in a Nutshell
_The Cybersecurity Dilemma: Hacking, Trust, and Fear Between Nations_ organizes its arguments in a fairly no-nonsense, premise –> premise –> conclusion manner that I thought would be good to summarize with a blog post. You can buy the book [here](https://www.amazon.com/Cyberse... | https://www.lesswrong.com/posts/oA8anbAiopGEzz8aj/the-cybersecurity-dilemma-in-a-nutshell |
# Examples of Causal Abstraction
I’m working on a [theory of abstraction](https://www.lesswrong.com/posts/wuJpYLcMEBz4kcgAn/what-is-abstraction-1) suitable as a [foundation for embedded agency](https://www.lesswrong.com/posts/hLFD6qSN9MmQxKjG5/embedded-agency-via-abstraction) and specifically multi-level world models.... | https://www.lesswrong.com/posts/Expvyb6nndbjqigRL/examples-of-causal-abstraction |
# Fractional Sweets
We have a rule with the kids that they can have two sweets a day: one before dinner and one after. They don't roll over, so you can't skip your morning sweet to have two evening ones. You can eat just part of a sweet, though, and Lily (5y) has gotten really into fractional sweets.
It started with ... | https://www.lesswrong.com/posts/iaBh8TKaMhqoRfpj2/fractional-sweets |
# A Re-Interpretation of the Normative Foundations of Majority Rule
This is a working paper on group rationality by Mahendra Prasad, who has previously published, among other things, "Social Choice and the Value Alignment Problem" in _Artificial Intelligence Safety and Security_, edited by Roman V. Yampolskiy. He's pr... | https://www.lesswrong.com/posts/zWpRpS2AekmmiSYtq/a-re-interpretation-of-the-normative-foundations-of-majority |
# ialdabaoth is banned
[ialdabaoth](https://www.lesswrong.com/users/ialdabaoth) is banned from LessWrong, because I think he is manipulative in ways that will predictably make the epistemic environment worse. This ban is unusual in several respects: it relies somewhat heavily on evidence from in-person interactions an... | https://www.lesswrong.com/posts/uwBKaeQzsvkcErmBm/ialdabaoth-is-banned |
# CHAI Internship applications are due by 12/15
Hi everyone,
I'm the Assistant Director at [CHAI](https://humancompatible.ai/) and as some of you may know, CHAI is currently accepting applications for our 2020 internship program.
The application closes at 11:59pm PST on 12/15 (this upcoming Sunday). If you are an un... | https://www.lesswrong.com/posts/W4fCJqBvKD6vrMAet/chai-internship-applications-are-due-by-12-15 |
# Under what circumstances is "don't look at existing research" good advice?
In [How I do research](https://www.lesswrong.com/posts/e3Db4w52hz3NSyYqt/how-i-do-research), TurnTrout writes:
> [I] Stare at the problem on my own, ignoring any existing thinking as much as possible. Just think about what the problem is, w... | https://www.lesswrong.com/posts/oJfyksr2YAqezHpav/under-what-circumstances-is-don-t-look-at-existing-research |
# Preface to CLR's Research Agenda on Cooperation, Conflict, and TAI
The [Center on Long-Term Risk (CLR)](https://longtermrisk.org/) is focused on [reducing risks of astronomical suffering](https://foundational-research.org/cause-prioritization-downside-focused-value-systems/), or _s-risks_, from transformative artif... | https://www.lesswrong.com/posts/DbuCdEbkh4wL5cjJ5/preface-to-clr-s-research-agenda-on-cooperation-conflict-and |
# [Link] "Doing being rational: polymerase chain reaction" by David Chapman
This is a link post for:
- [Doing being rational: polymerase chain reaction | Meaningness](https://meaningness.com/metablog/rational-pcr)
I'm not going to quote the content at the link itself. [Should I?]
David Chapman – the author of the ... | https://www.lesswrong.com/posts/Cs6xXWjN8W4xzQe9w/link-doing-being-rational-polymerase-chain-reaction-by-david |
# Book Recommendations for social skill development?
I have come to a realisation a bit later than I should have. Although I am still quite young and definitely have time to act on this realisation now, I wish I had started sooner.
I am studying to become a teacher, and I hope to go into education policy later, with ... | https://www.lesswrong.com/posts/G4d6qh6imcpwSqsJY/book-recommendations-for-social-skill-development |
# [Personal Experiment] One Year without Junk Media
I wrote in a [previous post](https://www.lesswrong.com/posts/koa48Qio5LP48xMM2/mediums-overpower-messages) how my life is better when I avoid certain kinds of media. It increases my happiness, lowers my stress and makes me smarter.
I should follow my own advice. I ... | https://www.lesswrong.com/posts/wQnJ4ZBEbwE9BwCa3/personal-experiment-one-year-without-junk-media |
# Is the term mesa optimizer too narrow?
In the [post introducing mesa optimization](https://www.alignmentforum.org/s/r9tYkB2a8Fp4DN8yB/p/FkgsxrGf3QxhfLWHG), the authors defined an _optimizer_ as
> a system \[that is\] internally searching through a search space (consisting of possible outputs, policies, plans, strat... | https://www.lesswrong.com/posts/nFDXq7HTv9Xugcqaw/is-the-term-mesa-optimizer-too-narrow |
# When would an agent do something different as a result of believing the many worlds theory?
One of the things impeding the many worlds vs wavefunction-collapse dialogue is that nobody seems to be able to point to a situation in which the difference clearly matters, where we would make a different decision depending ... | https://www.lesswrong.com/posts/crhgwfgYKN6CcA737/when-would-an-agent-do-something-different-as-a-result-of |
# Approval Extraction Advertised as Production
Paul Graham has a new essay out, [The Lesson to Unlearn](http://paulgraham.com/lesson.html), on the desire to pass tests. It covers the basic points made in Hotel Concierge's [The Stanford Marshmallow Prison Experiment](https://hotelconcierge.tumblr.com/post/113360634364/... | https://www.lesswrong.com/posts/ypXHnS3vWeDRmDsK9/approval-extraction-advertised-as-production |
# Boston Solstice 2019 Retrospective
This was the second time we hosted the Boston Secular Solstice at our house ( [2018 retrospective](https://www.jefftk.com/p/boston-solstice-2018-retrospective)), and my first time running it. I'd been in a "music director" sort of role for Boston Solstices going back to 2013, but w... | https://www.lesswrong.com/posts/wAF5zaGsKKTerwFmT/boston-solstice-2019-retrospective |
# Taxi Industry Regulation, Deregulation, and Reregulation: The Paradox of Market Failure
([from the March 2018 Gwern.net newsletter](https://www.gwern.net/newsletter/2018/03))
Uber/Lyft are disrupting the near-total regulatory capture of the taxi industry. But why does that regulatory mess exist in the first place? ... | https://www.lesswrong.com/posts/M8HZrxW5zpMkKnGzh/taxi-industry-regulation-deregulation-and-reregulation-the |
# A parable in the style of Invisible Cities
Cities & Belief: Goreme
=======================
The city of Goreme hides a terrible secret. At first, innocent visitor, you will admire the old bridge over the river, the colorful graffiti on alley walls, and the flags waving prettily in the public square. You will be impr... | https://www.lesswrong.com/posts/Aut78T9pv4pPhdcKe/a-parable-in-the-style-of-invisible-cities |
# How time tracking can help you prioritize
_This piece is cross-posted on my blog [here](https://effectivealtruismcoaching.com/blog/2019/12/15/how-time-tracking-can-help-you-prioritize)._
Ann told her supervisor that she would get a draft of her research proposal to them by Friday. However, it’s Sunday afternoon and... | https://www.lesswrong.com/posts/4vtZMJWDCa4gLSWSA/how-time-tracking-can-help-you-prioritize |
# How’s that Epistemic Spot Check Project Coming?
Quick context: [Epistemic spot checks](https://acesounderglass.com/tag/epistemicspotcheck/) started as a process in which I did quick investigations a few of a book’s early claims to see if it was trustworthy before continuing to read it, in order to avoid wasting time... | https://www.lesswrong.com/posts/qs3PcCMBnxwv994jW/how-s-that-epistemic-spot-check-project-coming |
# T-Shaped Organizations
Back in 2012, the Valve [New Employee Handbook](https://steamcdn-a.akamaihd.net/apps/valve/Valve_NewEmployeeHandbook.pdf) made the rounds on the internet, and claimed that Valve was looking for T-shaped individuals, who were broad generalists (the horizontal top of the T) with a depth of exper... | https://www.lesswrong.com/posts/bBJzkbzmFRGpx9QbM/t-shaped-organizations |
# Voluntourism
Someone recently asked me what I thought of the idea of using extra vacation days for volunteering, from an effective altruism perspective. Here's what I wrote to them:
In the effective altruism community people tend to be relatively down on this approach. It's relatively common for, say, a church grou... | https://www.lesswrong.com/posts/mnXRFBxvQWYsdLTDu/voluntourism |
# Counterfactual Induction
One of the first attempts at defining counterfactuals was the conjecture that ϕ is a counterfactual consequence of ψ when there is a short proof of ψ→ϕ.
This post will present a variant of logical induction that implements these proof-length-based counterfactuals, so there are well-defined... | https://www.lesswrong.com/posts/EAqHkKtbefvyRs4nw/counterfactual-induction |
# Sections 1 & 2: Introduction, Strategy and Governance
_This post is part of the sequence version of the Effective Altruism Foundation's [research agenda on Cooperation, Conflict, and Transformative Artificial Intelligence](https://www.lesswrong.com/s/p947tK8CoBbdpPtyK)._
### 1 Introduction
Transformative artifici... | https://www.lesswrong.com/posts/KMocAf9jnAKc2jXri/sections-1-and-2-introduction-strategy-and-governance |
# Sections 3 & 4: Credibility, Peaceful Bargaining Mechanisms
_This post is part of the sequence version of the Effective Altruism Foundation's [research agenda on Cooperation, Conflict, and Transformative Artificial Intelligence](https://www.lesswrong.com/s/p947tK8CoBbdpPtyK)._
### 3 Credibility
*Credibility* is ... | https://www.lesswrong.com/posts/8xKhCbNrdP4gaA8c3/sections-3-and-4-credibility-peaceful-bargaining-mechanisms |
# A dilemma for prosaic AI alignment
_Epistemic status: I predict that people who focus on [prosaic AI alignment](https://ai-alignment.com/prosaic-ai-control-b959644d79c2) have thought of this before, in some way at least. But I don’t know what they would say in response, so I’m writing this up so I can find out! I’m ... | https://www.lesswrong.com/posts/jYdAxH8BarPT4fqnb/a-dilemma-for-prosaic-ai-alignment |
# Counterfactual Mugging: Why should you pay?
The LessWrong Wiki defines Counterfactual Mugging as follows:
> [Omega](https://wiki.lesswrong.com/wiki/Omega) appears and says that it has just tossed a fair coin, and given that the coin came up tails, it decided to ask you to give it $100. Whatever you do in this situa... | https://www.lesswrong.com/posts/h9qQQA3g8dwq6RRTo/counterfactual-mugging-why-should-you-pay |
# Is Causality in the Map or the Territory?
steve2152 [brought up a great example](https://www.lesswrong.com/posts/mZy6AMgCw9CPjNCoK/computational-model-causal-diagrams-with-symmetry?commentId=JF9jmPNkNkH8yJX9g):
> Consider a 1kΩ resistor, in two circuits. The first circuit is the resistor attached to a 1V \[voltage\... | https://www.lesswrong.com/posts/ZBYE2F5DBiZtj6m95/is-causality-in-the-map-or-the-territory |
# On the role of abstraction in mathematics and the natural sciences
I'm a mathematics undergraduate in the UK, in my final year, and I have been thinking a lot about Platonism. I largely find myself completely unconvinced by attempts to naturalise the metaphysics behind the ontology of mathematical objects, though I ... | https://www.lesswrong.com/posts/NFMPftEhagCcHcmQJ/on-the-role-of-abstraction-in-mathematics-and-the-natural |
# Big Community Solstice
In 2011, Ray [started](https://www.lesswrong.com/posts/jES7mcPvKpfmzMTgC/ritual-report-nyc-less-wrong-solstice-celebration) a tradition in the broader rationalist community of having a gathering around the winter solstice. As the person who started the thing, if he says we should do it differe... | https://www.lesswrong.com/posts/cHujateswTyJfMkPT/big-community-solstice |
# [AN #77]: Double descent: a unification of statistical theory and modern ML practice
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/1PwWbW... | https://www.lesswrong.com/posts/LYdvzXF6E4iXM2ZSD/an-77-double-descent-a-unification-of-statistical-theory-and |
# Inductive biases stick around
_This post is a follow-up to [Understanding “Deep Double Descent”](https://www.alignmentforum.org/posts/FRv7ryoqtvSuqBxuT/understanding-deep-double-descent)._
I was talking to Rohin at NeurIPS about my [post on double descent](https://www.alignmentforum.org/posts/FRv7ryoqtvSuqBxuT/unde... | https://www.lesswrong.com/posts/nGqzNC6uNueum2w8T/inductive-biases-stick-around |
# Against Premature Abstraction of Political Issues
A few days ago romeostevensit [wrote](https://www.lesswrong.com/posts/WoaKd6bZsXN5LCnB4/steelmanning-social-justice?commentId=QRkvxRLXjq3dyGQMp) in response to me asking about downvotes on a post:
>I didn’t downvote, but I do think that conversations like this... | https://www.lesswrong.com/posts/bFv8soRx6HB94p5Pg/against-premature-abstraction-of-political-issues |
# Abstraction, Causality, and Embedded Maps: Here Be Monsters
In a lot of examples of abstract causality, the abstract causal structure generally mirrors the ground-level causal structure. Think about fluid flow, for instance. At the (classical) ground level, we have a bunch of particles whose interactions are local i... | https://www.lesswrong.com/posts/ipCAL4tx7jcJsFasY/abstraction-causality-and-embedded-maps-here-be-monsters |
# "You can't possibly succeed without [My Pet Issue]"
There's a particular conversational move that I've noticed people making over the past couple years. I've also noticed *myself* making it. The move goes:
> "You can't *possibly* succeed without X", where X is whatever principle the person is arguing for.
(Where ... | https://www.lesswrong.com/posts/gxbJfvb5eejrGkA9w/you-can-t-possibly-succeed-without-my-pet-issue |
# One Million Dollars
On Giving Tuesday, Julia and I [donated](https://www.jefftk.com/donations) our millionth dollar to [GiveWell's top charities](https://www.jefftk.com/p/my-effective-altruism-timeline) and [growing](https://www.jefftk.com/p/more-meta-funding-thoughts) the Effective Altruism movement. It was more dr... | https://www.lesswrong.com/posts/w4wQzyYSvSGBv8oWS/one-million-dollars |
# 2019 AI Alignment Literature Review and Charity Comparison
C_ross-posted to the EA forum [here](https://forum.effectivealtruism.org/posts/dpBB24QsnsRnkq5JT/2019-ai-alignment-literature-review-and-charity-comparison)._
Introduction
============
As in [2016](https://forum.effectivealtruism.org/posts/nSot23sAjoZRgaEw... | https://www.lesswrong.com/posts/SmDziGM9hBjW9DKmf/2019-ai-alignment-literature-review-and-charity-comparison |
# Propagating Facts into Aesthetics
*Epistemic status: Tentative. I’ve been practicing this on-and-off for a year and it’s seemed valuable, but it’s the sort of thing I might look back on and say “hmm, that wasn’t really the right frame to approach it from.”*
In doublecrux, the focus is on “what observations would ch... | https://www.lesswrong.com/posts/YN6daWakNnkXEeznB/propagating-facts-into-aesthetics |
# CFAR: Progress Report & Future Plans
_Context: This is the first in a series of [year-end updates](https://rationality.org/fundraiser) from CFAR. This post mainly describes what CFAR did over the past two years, and what it plans to do next year; other posts in the series will describe more about its mission, [finan... | https://www.lesswrong.com/posts/vj6CYLuDPw3ieCB4A/cfar-progress-report-and-future-plans-1 |
# Neural networks as non-leaky mathematical abstraction
I find it rather weird that mathematics is usually thought to people on a historical basis. It's likely possible for a 4th century Alexandrian mathematician to teach it in a modern primary school and get decent results. An 18th century Venetian tutor could carry ... | https://www.lesswrong.com/posts/qRtbjHsJiwYggYhP4/neural-networks-as-non-leaky-mathematical-abstraction |
# When Goodharting is optimal: linear vs diminishing returns, unlikely vs likely, and other factors
In this post, I'll argue that some of the behaviours that seem to be clear examples of the [Goodhart problem](https://www.lesswrong.com/posts/EbFABnst8LsidYs5Y/goodhart-taxonomy#Extremal_Goodhart), are in fact not. And ... | https://www.lesswrong.com/posts/megKzKKsoecdYqwb7/when-goodharting-is-optimal-linear-vs-diminishing-returns |
# [Part 2] Amplifying generalist research via forecasting – results from a preliminary exploration
_This post covers the set-up and results from our exploration in amplifying generalist research using predictions, in detail. It is accompanied by [a second post](https://www.lesswrong.com/posts/cLtdcxu9E4noRSons) with a... | https://www.lesswrong.com/posts/FeE9nR7RPZrLtsYzD/part-2-amplifying-generalist-research-via-forecasting |
# [Part 1] Amplifying generalist research via forecasting – Models of impact and challenges
_This post covers our models of impact and challenges with our exploration in amplifying generalist research using forecasting. It is accompanied by [a second post](https://www.lesswrong.com/posts/FeE9nR7RPZrLtsYzD) with a high... | https://www.lesswrong.com/posts/cLtdcxu9E4noRSons/part-1-amplifying-generalist-research-via-forecasting-models |
# We run the Center for Applied Rationality, AMA
CFAR recently launched its 2019 [fundraiser](https://rationality.org/fundraiser), and to coincide with that, we wanted to give folks a chance to ask us about our mission, plans, and strategy. Ask any questions you like; we’ll respond to as many as we can from 10am PST o... | https://www.lesswrong.com/posts/96N8BT9tJvybLbn5z/we-run-the-center-for-applied-rationality-ama |
# Causal Abstraction Intro
I haven't been terribly satisfied by the first few posts in this sequence; they don't do a very good job introducing things. I think part of the problem is the format, so I [decided to invest in a high-end studio](https://www.lesswrong.com/posts/KzAG4yWQJosmEjHe2/blatant-lies-are-the-best-ki... | https://www.lesswrong.com/posts/yD9GLtQgp8vAfndL8/causal-abstraction-intro |
# Should We Still Fly?
I've seen a lot of discussion about plane travel from a climate perspective lately, with people arguing that we should try to restructure our lives to fly much less. Avoid business travel, vacation closer to home, visit relatives less, etc. After looking at the numbers, though, I think this most... | https://www.lesswrong.com/posts/mFeGmtdLLaeHqyxuk/should-we-still-fly |
# Sections 5 & 6: Contemporary Architectures, Humans in the Loop
_This post is part of the sequence version of the Effective Altruism Foundation's [research agenda on Cooperation, Conflict, and Transformative Artificial Intelligence](https://www.lesswrong.com/s/p947tK8CoBbdpPtyK)._
### 5 Contemporary AI architectures... | https://www.lesswrong.com/posts/4GuKi9wKYnthr8QP9/sections-5-and-6-contemporary-architectures-humans-in-the |
# Understanding Machine Learning (I)
[Understanding Machine Learning](https://www.amazon.com/Understanding-Machine-Learning-Theory-Algorithms/dp/1107057132) is another book on [Miri's research guide](https://intelligence.org/research-guide/). I've previously written about [Topology](https://www.lesswrong.com/posts/rYu... | https://www.lesswrong.com/posts/Z29etNPef6yvrPteK/understanding-machine-learning-i |
# Polio and the controversy over randomized clinical trials
I’m currently reading _[Polio: An American Story](https://rootsofprogress.org/books/polio-an-american-story)_, by David Oshinsky, and I came across a fascinating story:
In 1954, when it was time for large-scale human trials of the first polio vaccine, some r... | https://www.lesswrong.com/posts/N2fsGN8FLE766gQZB/polio-and-the-controversy-over-randomized-clinical-trials |
# Clarifying Power-Seeking and Instrumental Convergence
Previously: _[Seeking Power is Provably Instrumentally Convergent in MDPs](https://www.lesswrong.com/posts/6DuJxY8X45Sco4bS2/seeking-power-is-instrumentally-convergent-in-mdps)_
Rohin Shah and Vanessa Kosoy pointed out a subtle problem with my interpretation of ... | https://www.lesswrong.com/posts/cwpKagyTvqSyAJB7q/clarifying-power-seeking-and-instrumental-convergence |
# Quadratic voting for the 2018 Review
LessWrong is currently [reviewing the posts from 2018](https://www.lesswrong.com/posts/qXwmMkEBLL59NkvYR/the-lesswrong-2018-review), and I'm trying to figure out how voting should happen. The new hotness that all your friends are talking about is [quadratic voting](https://vitali... | https://www.lesswrong.com/posts/qQ7oJwnH9kkmKm2dC/quadratic-voting-for-the-2018-review |
# Free Speech and Triskaidekaphobic Calculators: A Reply to Hubinger on the Relevance of Public Online Discussion to Existential Risk
In response to [Wei Dai's claim that](https://www.lesswrong.com/posts/bFv8soRx6HB94p5Pg/against-premature-abstraction-of-political-issues#mp4uYsP9cfukFhMSd) a multi-post 2009 _Less Wron... | https://www.lesswrong.com/posts/yaCwW8nPQeJknbCgf/free-speech-and-triskaidekaphobic-calculators-a-reply-to |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.