id stringlengths 36 36 | source stringclasses 15
values | formatted_source stringclasses 13
values | text stringlengths 2 7.55M |
|---|---|---|---|
70375b15-9dd5-4fce-9d9f-0bfc905a654e | trentmkelly/LessWrong-43k | LessWrong | Meetup : LW Munich Meetup in August
Discussion article for the meetup : LW Munich Meetup in August
WHEN: 10 August 2013 03:00:00PM (+0200)
WHERE: Hirschgarten, 80639 Munich, Germany
The next Munich meetup will take place on August 10th. You are highly welcome to come and say hi, no matter how long you’ve been readi... |
d7108f54-a754-45af-b216-b9f9b857253b | trentmkelly/LessWrong-43k | LessWrong | Exploring Notions of "Utility Approximation" and testing quantilizers
|
976e1af8-34a8-451f-b485-dfad943ac600 | trentmkelly/LessWrong-43k | LessWrong | Neuroscience basics for LessWrongians
The origins of this article are in my partial transcript of the live June 2011 debate between Robin Hanson and Eliezer Yudkowsky. While I still feel like I don't entirely understand his arguments, a few of his comments about neuroscience made me strongly go, "no, that's not right.... |
5c831d1e-3a5a-44c0-80e9-b55a19274190 | trentmkelly/LessWrong-43k | LessWrong | Notes From an Apocalypse
(This is a loose adaptation of a talk I sometimes give on the Cambrian Explosion, smoothed a bit for popular consumption. That talk, in turn, draws heavily from a 2006 paper by Charles Marshall, titled “Explaining the Cambrian ‘Explosion’ of Animals”. It can be read in full here. I’m mostly ju... |
737daa10-d1b9-4f90-8d63-bf4b47aa5b0c | trentmkelly/LessWrong-43k | LessWrong | Would a tax on AI products be useful?
On one hand, it would capture part of the negative externality in the market pricing. And the money could be used to finance AI safety research.
On the other hand, maybe AI companies would think their responsability are limited to paying that tax and that's it.
It might also be ... |
473c6c42-6418-40ed-8036-1f7fee4f97c4 | trentmkelly/LessWrong-43k | LessWrong | Meetup : Washington D.C. - Mini Talks
Discussion article for the meetup : Washington D.C. - Mini Talks
WHEN: 19 October 2014 03:00:00PM (-0400)
WHERE: National Portrait Gallery
We will be meeting in the Kogod Courtyard of the National Portrait Gallery (8th and F Sts or 8th and G Sts NW, go straight past the informa... |
bac56a17-3152-4324-95a3-bb7335029e0c | trentmkelly/LessWrong-43k | LessWrong | Rational Feed
=== Updates:
I have been a little more selective about which articles make it onto the feed. I have not been overly selective and all of the obviously general interest rationalsit articles still make it.
Unless people object I am going to try a "weekly feed". The bi-weekly feed is pretty long. I curren... |
cfd6cc36-7677-4873-ac18-2357ee4af593 | trentmkelly/LessWrong-43k | LessWrong | Karma votes: blind to or accounting for score?
Normally I try to use karma votes in a practical way to try to help promote content that I want to see more of on LW, or to downrank content that I want to see less of. (As opposed to using them as a kind of social signal, which is a separate issue.)
One thing I have nev... |
0feeb954-c851-4d6a-8fc0-fc6401fd1771 | trentmkelly/LessWrong-43k | LessWrong | Weekly LW Meetups: Pittsburgh, Melbourne
There are upcoming irregularly scheduled Less Wrong meetups in:
* Pittsburgh Holiday Meetup: 16 December 2011 08:00PM
* Indianapolis, Potential: 08 January 2012 02:00PM
* Salt Lake City, Late January 2012
* First Brussels meetup: 18 February 2012 11:00AM
The following mee... |
65975444-eaa1-401c-9d98-1ea1f8d7453b | trentmkelly/LessWrong-43k | LessWrong | Covid 12/1/22: China Protests
This week China experienced their largest protests since 1989, and according to many sources the most dangerous as well. China has painted itself into a corner. It cannot abandon Zero Covid, yet the costs of Zero Covid rise with each passing day and the policy is also slowly failing. What... |
285be01b-388b-4f23-af2a-a14bdc2a6c67 | trentmkelly/LessWrong-43k | LessWrong | Only say 'rational' when you can't eliminate the word
Almost all instances of the word "true" can be eliminated from the sentences in which they appear by applying Tarski's formula. For example, if you say, "I believe the sky is blue, and that's true!" then this can be rephrased as the statement, "I believe the sky i... |
244cde54-bcfb-4c68-95f7-131df58d4275 | trentmkelly/LessWrong-43k | LessWrong | Noticing the Value of Noticing Confusion
Crossposted from spacelutt.com
> Your strength as a rationalist is your ability to be more confused by fiction than by reality. Either your model is wrong or this story is false.
~ Your Strength as a Rationalist, Eliezer Yudkowsky
Your model of the world is how you understan... |
286d61d2-92b9-48b8-a34d-63d5eddac022 | trentmkelly/LessWrong-43k | LessWrong | When there are too few people to donate money to
Just a comment on how some people decide the money/time allocation problem when helping their charity of choice. I almost posted it to the open thread, but it is longish, so I posted it here.
1. The ordinary hour.
When I read about the lawyer working at the soup ki... |
c400ae35-58a8-4912-a3a6-9cfc74e0d99a | trentmkelly/LessWrong-43k | LessWrong | Monthly Roundup #17: April 2024
As always, a lot to get to. This is everything that wasn’t in any of the other categories.
BAD NEWS
You might have to find a way to actually enjoy the work.
> Greg Brockman (President of OpenAI): Sustained great work often demands enjoying the process for its own sake rather than onl... |
842c5dee-28a4-483d-98be-6f9f29813b9f | trentmkelly/LessWrong-43k | LessWrong | Meetup : Washington DC Megametup debriefing/planning meetup
Discussion article for the meetup : Washington DC Megametup debriefing/planning meetup
WHEN: 21 July 2013 03:00:00PM (-0400)
WHERE: National Portrait Gallery, Washington, DC 20001, USA (courtyard)
We'll be meeting to talk about the megameetup last week, an... |
7ffcadbd-5bb2-4ea4-88ee-87c1c5aef941 | trentmkelly/LessWrong-43k | LessWrong | New paper: Corrigibility with Utility Preservation
I am pleased to announce the availability of a long-format paper with new results on AGI safety: Corrigibility with Utility Preservation.
You can get the paper at https://arxiv.org/abs/1908.01695 , and in the related software repository at https://github.com/kholtman... |
ba1440f5-647f-4bc6-a6b6-d3304dd42b14 | trentmkelly/LessWrong-43k | LessWrong | [Crosspost] Unveiling the American Public Opinion on AI Moratorium and Government Intervention: The Impact of Media Exposure
This is a summary of a follow-up study conducted by the Existential Risk Observatory, which delves into a greater number media items. To access our previous study, please follow this link. This ... |
2428c100-1c4d-4ac8-abfe-604b8f4c547a | trentmkelly/LessWrong-43k | LessWrong | Theories of Modularity in the Biological Literature
Introduction
This post is part of a sequence describing our team’s research on selection theorems for modularity, as part of this year's AI Safety Camp, under the mentorship of John Wentworth. Here, we provide some background reading for the discussion of modularity ... |
ca9fa412-8078-4ccf-94a3-bf51c364cae2 | trentmkelly/LessWrong-43k | LessWrong | Experience with dual n-back?
I've run across a couple of questions about the value of dual n-back, and this is the best place I know of to find people who've worked with it.
If you've tried it, has it noticeably improved your working memory? Was it enough to be worth the time? |
f5098495-ddb2-4a8d-bb70-3b723582863a | StampyAI/alignment-research-dataset/blogs | Blogs | Benchmarking the next generation of never-ending learners
#### Learning how to build upon knowledge by tapping 30 years of computer vision research
In just a few years, large-scale deep learning (DL) models have achieved unprecedented success in a variety of domains, from predicting protein structures to natural lang... |
adfb3d29-2c37-4fc4-8d2a-6c63a2343cd9 | trentmkelly/LessWrong-43k | LessWrong | Making every researcher seek grants is a broken model
When Galileo wanted to study the heavens through his telescope, he got money from those legendary patrons of the Renaissance, the Medici. To win their favor, when he discovered the moons of Jupiter, he named them the Medicean Stars. Other scientists and inventors o... |
7eb09e76-e1b2-4a23-a81b-6831391d02bf | trentmkelly/LessWrong-43k | LessWrong | Optimization, speculations on the X and only X problem.
Epistemic status: Speculative. Rambling stream of ideas.
Recomended semiprerequisite https://www.lesswrong.com/posts/DkcdXsP56g9kXyBdq/coherence-arguments-imply-a-force-for-goal-directed-behavior
How do we tell if an arbitrary system is an optimizer or not?
W... |
545c3cc2-234e-4743-a577-b2bcee3c6282 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Assessing Policy, Loss and Planning Combinations in Reinforcement Learning using a New Modular Architecture
I Introduction
---------------
In 2016, a program called AlphaGo [AlphaGo] beat a Go world champion for the first time. Go was seen as the new milestone of artificial intelligence since the former world chess... |
44c4eb67-6b9d-44cd-b2c4-e1db8c8a1b63 | trentmkelly/LessWrong-43k | LessWrong | You don't need to be a genius to be in AI safety research
The aim of this article is to share my experiences within the AI safety community to those in research and operations in the field, with the goal of creating a more inclusive, supportive, and positive environment. The intention of this article is to allow those... |
5e5c7035-7198-40a9-a243-87cd76b2a753 | StampyAI/alignment-research-dataset/arbital | Arbital | Shutdown utility function
A special case of [low impact](https://arbital.com/p/2pf) which probably seems deceptively trivial - how would you create a utility function such that an agent with this utility function would harmlessly shut down? Without, for example, creating an environmental subagent that assimilated all... |
2bc36ce4-4ec5-4a02-aeaf-14d543e8a1ef | trentmkelly/LessWrong-43k | LessWrong | New Year Review Resources
I’ve gathered together a handful of resources here on yearly reviews and setting new years goals. …Just in time for everyone to have wrapped up their new years reflections. Oh well, better late than never, right?
These aren’t particularly cohesive or polished. I recommend you skim through an... |
d09dbace-5173-4767-8727-8814d2d1ba8b | trentmkelly/LessWrong-43k | LessWrong | It's all in your head-land
From David Foster Wallace's Infinite Jest:
> He could do the dextral pain the same way: Abiding. No one single instant of it was unendurable. Here was a second right here: he endured it. What was undealable-with was the thought of all the instants all lined up and stretching ahead, glitteri... |
d5edc4a1-4c30-4b11-abe9-fc06be1c9f4d | trentmkelly/LessWrong-43k | LessWrong | Long Term Future Fund application is closing this Friday (October 11th)
Just a reminder that the Long Term Future Fund application is closing in two days. In past rounds we received 80% of our applications the day of the deadline, but I am not sure whether that is contingent on us reminding everyone that the applicati... |
e04c1076-686a-47ca-a6e7-033ab79471f8 | StampyAI/alignment-research-dataset/special_docs | Other | Complex Value Systems in Friendly AI
J. Schmidhuber, K.R. Thórisson, and M. Looks (Eds.): AGI 2011, LNAI 6830, pp. 388–393, 2011.
© Springer-Verlag Berlin Heidelberg 2011 Complex Value Systems in Friendly AI\*
Eliezer Yudkowsky
Singularity Institute for Artificial Intelligence, San Francisco, CA
yudkowsky@singinst... |
285d30b6-099a-4c9d-aaee-417fd1186324 | trentmkelly/LessWrong-43k | LessWrong | Savanna Poets
> "Poets say science takes away from the beauty of the stars—mere globs of gas atoms. Nothing is "mere". I too can see the stars on a desert night, and feel them. But do I see less or more?
> "The vastness of the heavens stretches my imagination—stuck on this carousel my little eye can catch o... |
f9265fb9-e084-4133-af06-da5f3545679d | trentmkelly/LessWrong-43k | LessWrong | Price dikes
I like this comment, from Scott Alexander:
> I think there’s a general principle that once you pass dumb regulations, it’s going to make bad things happen, and then if you try to solve those bad things by passing further regulations, you’re just going to get caught in an endless trap.
>
> So first they r... |
67f34f15-cb59-4692-aa4f-2b105df83dbc | trentmkelly/LessWrong-43k | LessWrong | A Dragon Confronts the Terasem Movement
I wrote a previous blog takedown of Malatora, which was somewhat shooting fish in a barrel. In the spirit of moving on to harder targets, I now turn my meager attentions and talents to the subject of the Terasem Movement. This is a marginally more difficult movement to vivisect,... |
caef58c6-d15b-44b2-a8f7-21be8f4a07c4 | trentmkelly/LessWrong-43k | LessWrong | My submission to the ALTER Prize
This is a linkpost for my submission to the ALTER AI Alignment Prize. As such, it's 8 and a half pages of pretty unapologetically dense math and/or philosophy the whole way through. It represents my own work, and I stand by the results and thinking in it. As such, I welcome actionable ... |
1e2794a8-ee31-4eb9-8a95-f1bf7fb23c29 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Cryptocurrency Exploits Show the Importance of Proactive Policies for AI X-Risk
When technology takes a leap in innovation, such as with cryptocurrency, existing institutions are forced to play catch-up with education and regulations.
Cryptocurrency and Web3 enable improved privacy, censorship resistance, fewer inte... |
f06674dd-f81b-4b73-a59a-fbadd5fd7a17 | StampyAI/alignment-research-dataset/special_docs | Other | Artificial General Intelligence: Coordination and Great Powers
Artificial General
Intelligence:
Coordination
& Great Powers
A white paper based on the
2018 Foresight Institute Strategy Meeting on AGI
Allison Duettmann , Foresight Institute
Olga Afanasjeva , GoodAI
Stuart Armstrong , Future of Humanity Institute
Ry... |
68c7aced-c992-4d56-a6c0-af84a8e5fa21 | trentmkelly/LessWrong-43k | LessWrong | [Help]: Social cost of cryonics?
Over the past few months I've been doing a lot of reading about cryonics, and though I agree with the arguments of Eliezer Yudkowsky and Robin Hanson on the issue, I still feel uncomfortable about actually signing up. Upon reflection, my true rejection is my fear of the social cost of... |
6fdecbfd-2fea-470a-b976-109d4a387477 | trentmkelly/LessWrong-43k | LessWrong | Quantifying Human Suffering and "Everyday Suffering"
In the case of humans, it seems self-evident that suffering is a consciously experienced, mental or psychological phenomenon. This makes it difficult to quantify, given our lack of access to other beings’ qualia. However, the science of neuropsychology seeks to corr... |
717235fa-cd90-4b14-8401-0f3d8330051a | trentmkelly/LessWrong-43k | LessWrong | Naive Decision Theory
I am posting this is because I'm interested in self-modifying agent decision theory but I'm too lazy to read up on existing posts. I want to see a concise justification as to why a sophisticated decision theory would be needed for the implementation of an AGI. So I'll present a 'naive' decision... |
6e6761fd-9b04-45b9-980f-f4e82d7a592a | trentmkelly/LessWrong-43k | LessWrong | Power Outage Chances
On yesterday's post about generators, Chris asked whether power outages are worth planning for. That's a good question: very roughly, you can think of whether it is worth spending $X to handle an emergency with probability P as whether you'd be willing to spend $X/P if the emergency were certain. ... |
904dd8b9-4059-46f5-a34b-82f01d9be842 | StampyAI/alignment-research-dataset/arxiv | Arxiv | A multi-component framework for the analysis and design of explainable artificial intelligence
1 Introduction
---------------
Fueled by a growing need for trust and ethical artificial intelligence (AI) by design, the wake of the last decade of machine learning is crowded with a broad spectrum of research on explain... |
646883eb-819a-4272-b0dd-ef64b6220947 | trentmkelly/LessWrong-43k | LessWrong | Snake Eyes Paradox
Disclaimer: this is a personal blog post that I'm posting for my benefit and do not expect to be valuable to LessWrong readers in general. If you're reading this, consider not.
This post analyzes the "Snake Eyes Paradox" as specified by Daniel Reeves (2013). This is a variation on the "Shooting Roo... |
83f0b34f-2bc2-447b-9353-62dbbb28d4f7 | trentmkelly/LessWrong-43k | LessWrong | Abstracted Idealized Dynamics
Followup to: Morality as Fixed Computation
I keep trying to describe morality as a "computation", but people don't stand up and say "Aha!"
Pondering the surprising inferential distances that seem to be at work here, it occurs to me that when I say "computation", some of my listeners ma... |
2081ad35-7834-459d-a32b-c4a10186b5a5 | trentmkelly/LessWrong-43k | LessWrong | Email tone and status: !s, friendliness, 'please', etc.
Do the following things in email tone lower status?:
* Exclamation marks
* Friendliness
* Saying 'please', or 'it'd be great if'
* Saying 'you don't have to do this, but...'
* Saying 'sorry' (e.g. 'sorry to bother you')
* Signing of with 'Thanks!'
My im... |
2991c6eb-c0cc-4def-b714-6a8d8595dd59 | trentmkelly/LessWrong-43k | LessWrong | What’s the future of AI hardware?
Since the AI mania erupted, I haven’t made up my mind about the century's hottest topic. I’ve been experimenting with building “AI-inside” products for a year now, but I’ve cautiously tried not to form a strong opinion about it.
As a software native, the recent AI hardware discourse ... |
b7405a40-b3a1-465d-b2d5-9e253c470ba5 | trentmkelly/LessWrong-43k | LessWrong | Paper reading as a Cargo Cult
I have come across various people (including my past self) who meet up regularly to study, e.g., alignment forum posts and discuss them. This helps people bond over their common believes, fears, and interests, which I think is good, but in no way is this ever going to lead anyone to find ... |
28ae0f57-6744-4cc3-82d2-92dc4acab71c | trentmkelly/LessWrong-43k | LessWrong | Prediction markets are consistently underconfident. Why?
look at this calibration s-curve from calibration.city
markets are consistently closer to 50% than should be. why?
I have my own guesses but curious to hear yours. |
fd5faff5-7b35-4827-b7e3-79b07341cab7 | awestover/filtering-for-misalignment | Redwood Research: Alek's Filtering Results | id: post513
This is a link post for the Anthropic Alignment Science team's first "Alignment Note" blog post. We expect to use this format to showcase early-stage research and work-in-progress updates more in the future. Twitter thread here. Top-level summary: In this post we present "defection probes": linear classifi... |
59f9dc40-47ee-4b12-b2f9-96eaac32e015 | awestover/filtering-for-misalignment | Redwood Research: Alek's Filtering Results | id: post924
YouTube link Is there some way we can detect bad behaviour in our AI system without having to know exactly what it looks like? In this episode, I speak with Mark Xu about mechanistic anomaly detection: a research direction based on the idea of detecting strange things happening in neural networks, in the h... |
f65fdd60-5b75-4842-b8e7-2e131c5903ce | trentmkelly/LessWrong-43k | LessWrong | Meetup : Skepticon IV meetup: Saturday night
Discussion article for the meetup : Skepticon IV meetup: Saturday night
WHEN: 19 November 2011 09:30:00PM (-0600)
WHERE: University Plaza Hotel, 333 S John Q. Hammons Parkway, Springfield, Missouri 65806
Following the last talk on Saturday, we'll meet up in the lobby of ... |
5da2839a-6389-47a6-8834-65c1d8463980 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Preventing Language Models from hiding their reasoning
When working with powerful AI models, one concern is that the AIs [steganographically encode secret messages](https://www.alignmentforum.org/posts/yDcMDJeSck7SuBs24/steganography-in-chain-of-thought-reasoning) when communicating with each other or in long Chains-o... |
0f6032f8-e300-46a2-8a4f-5a8dba756f65 | trentmkelly/LessWrong-43k | LessWrong | Ben Goertzel interviews Michael Anissimov regarding existential risk [link]
http://hplusmagazine.com/2011/04/20/mitigating-the-risks-of-artificial-superintelligence/ |
0fadc868-5d32-47a8-9c05-242e27fef9b6 | trentmkelly/LessWrong-43k | LessWrong | Meetup : Washington, D.C.: Fun & Games Meetup
Discussion article for the meetup : Washington, D.C.: Fun & Games Meetup
WHEN: 17 August 2014 03:00:00PM (-0400)
WHERE: National Portrait Gallery
We will be meeting in the Kogod Courtyard of the National Portrait Gallery (8th and F Sts or 8th and G Sts NW, go straight p... |
7f850f8b-cf8b-4050-be81-0438120cd511 | trentmkelly/LessWrong-43k | LessWrong | Factorio, Accelerando, Empathizing with Empires and Moderate Takeoffs
I started planning this post before Cousin It's post on a similar subject. This is a bit of a more poetic take on moderate, peaceful AI takeoffs.
Spoilers for the game Factorio, and the book Accelerando. They are both quite good. But, if you're not... |
944f9b23-2560-40e1-aa52-24869975e5b2 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Tradeoffs in Constructing and Evaluating Temporal Influence Diagrams
40 Provan
TRADEOFFS IN CONSTRUCTING AND EVALUA TING TEMPORAL
INFLUENCE DIAGRAMS
Gregory M. Provan*
Computer and Information Science Department
University of Pennsylvania
301C 3401 Walnut St. Philadelphia PA 19104-6228
Abstract
This paper addr... |
56cb62aa-d56d-4efc-927a-8b8e75678082 | trentmkelly/LessWrong-43k | LessWrong | Come for the productivity, stay for the philosophy
There's a large class of things that eventually make you more effective when you've studied them for a while, but which are challenging and seem like they produce no gains while being learned.
Formal logic, for instance, seems that way for a lot of people.
Many of t... |
ef8e8c27-ecb5-49c3-92a4-5becf90c9039 | trentmkelly/LessWrong-43k | LessWrong | On the OpenAI Economic Blueprint
TABLE OF CONTENTS
1. Man With a Plan.
2. Oh the Pain.
3. Actual Proposals.
4. For AI Builders.
5. Think of the Children.
6. Content Identification.
7. Infrastructure Week.
8. Paying Attention.
MAN WITH A PLAN
The primary Man With a Plan this week for government-guided AI pro... |
dcd15c2e-4489-4a33-8079-a3c41399a969 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | A measure-theoretic generalization of logical induction
.mjx-chtml {display: inline-block; line-height: 0; text-indent: 0; text-align: left; text-transform: none; font-style: normal; font-weight: normal; font-size: 100%; font-size-adjust: none; letter-spacing: normal; word-wrap: normal; word-spacing: normal; white-spa... |
938bc87d-1f25-4900-a961-77692becf075 | trentmkelly/LessWrong-43k | LessWrong | How can I make money?
Hey everyone,
I'd like too see if I can utilize the rational powers of Less Wrong
Currently I work full time. Am well paid for my age. Have approximately $20,000 AU in savings.
My spouse is in uni. Will finish this year and begin working next year. But currently I am supporting our household.
... |
a3e72f15-0498-43f7-ab03-3a95db6f8cf3 | trentmkelly/LessWrong-43k | LessWrong | SSC Meetups Everywhere: Community Map on frontpage this month
To celebrate and help people find meetups for this year's SSC meetups everywhere, we are pinning the big community map to the frontpage map again for the month of September (as we did last year).
If you want to get rid of the map, you can do so in your acc... |
2843f57c-cbae-4eae-99cd-11241fd633ed | StampyAI/alignment-research-dataset/arbital | Arbital | Image of the identity under a group homomorphism is the identity
For any [https://arbital.com/p/-47t](https://arbital.com/p/-47t) $f: G \to H$, we have $f(e_G) = e_H$ where $e_G$ is the identity of $G$ and $e_H$ the identity of $H$.
Indeed, $f(e_G) f(e_G) = f(e_G e_G) = f(e_G)$, so premultiplying by $f(e_G)^{-1}$ we ... |
35e9deec-2210-4de3-8ded-e533d1663caf | StampyAI/alignment-research-dataset/special_docs | Other | Can the Singularity Be Patented? (And Other IP Conundrums for Converging Technologies)
Chapter 10
Can the Singularity Be Patented?
(And Other IP Conundrumsfor Converging Technologies)
David Koepsell
10.1 Introduction
Many of the societal controls and policy considerations discussed in relation to the
development of “T... |
837e3208-0667-49a2-9dc9-883170b9859b | trentmkelly/LessWrong-43k | LessWrong | What are some exercises for building/generating intuitions about key disagreements in AI alignment?
I am interested in having my own opinion about more of the key disagreements within the AI alignment field, such as whether there is a basin of attraction for corrigibility, whether there is a theory of rationality that... |
7ef76f1e-2e80-44af-bc37-3f70afbcfa3a | trentmkelly/LessWrong-43k | LessWrong | Asymmetric Justice
Related and required reading in life (ANOIEAEIB): The Copenhagen Interpretation of Ethics
Epistemic Status: Trying to be minimally judgmental
Spoiler Alert: Contains minor mostly harmless spoiler for The Good Place, which is the best show currently on television.
The Copenhagen Interpretation of ... |
47f0343a-4d46-492f-b9c9-6b4614f57394 | trentmkelly/LessWrong-43k | LessWrong | Retrospective: Practical Social Networking
Practical Social Networking was the third Guild of the ROSE course. Taught by Councilor David Youssef, the course was an introduction to the art of growing and nurturing a healthy social web. While some people got a lot out of it, we also struggled with engagement and there w... |
59333b1b-7c87-4928-9c64-65ea0f7bf2ae | trentmkelly/LessWrong-43k | LessWrong | [SEQ RERUN] Terminal Values and Instrumental Values
Today's post, Terminal Values and Instrumental Values was originally published on 15 November 2007. A summary (taken from the LW wiki):
> Discusses a formalism for a discussion of the relationship between terminal and instrumental values. Terminal values are worl... |
440ff720-2df1-41ac-b7be-017e1ab379e4 | trentmkelly/LessWrong-43k | LessWrong | Learning to build conviction
People sometimes tell me that they want to join a startup, so that they can learn how it works, and eventually start one themselves. I usually end up suggesting that they skip straight to step 2 and start one themselves.
Why is that? Isn’t it better to learn from someone else’s mistakes t... |
24f74173-954c-4725-bfe4-d0e84db51636 | trentmkelly/LessWrong-43k | LessWrong | Bye Bye Benton: November Less Wrong Meetup
As some of you may know, SIAI is in the process of moving our Visiting Fellows Program to a larger and more permanent location in Berkeley. Nothing is final yet but, however things turn out, November will be the last month we spend at 3755 Benton street. In honor of the hou... |
c4d201ce-352d-4aa4-bbe5-d55a878f98a0 | trentmkelly/LessWrong-43k | LessWrong | GWWC Should Require Public Charity Evaluations
One of the roles of Giving What We Can (GWWC) is to help its members and other interested people figure out where to give. If you go to their site and click " start giving" they list charitable funds, including GiveWell's All Grants Fund and the EA Infra Fund, both of whi... |
c02d7e48-2138-4bd4-9497-9f113195ceaa | trentmkelly/LessWrong-43k | LessWrong | Luna Lovegood and the Chamber of Secrets - Part 2
None |
6b321974-ee77-462c-ae62-38cbdc15fc30 | trentmkelly/LessWrong-43k | LessWrong | "Is Science Broken?" is underspecified
http://fivethirtyeight.com/features/science-isnt-broken/
This is an interesting article-- it's got an overview of what's currently seen as the problems with replicability and fraud, and some material I haven't seen before about handing the same question to a bunch of scientists,... |
dd324a97-f82c-4a9c-bd7b-93cbf7707f3c | trentmkelly/LessWrong-43k | LessWrong | Meetup : Urbana-Champaign: Fun and Games
Discussion article for the meetup : Urbana-Champaign: Fun and Games
WHEN: 15 December 2013 02:00:00PM (-0600)
WHERE: 300 S Goodwin Ave Apt 102, Urbana, IL.
There will also probably be HPMOR discussion.
Discussion article for the meetup : Urbana-Champaign: Fun and Games |
36c891d9-1624-479b-952e-79ebccb4510d | trentmkelly/LessWrong-43k | LessWrong | Experts vs. parents
I'm reviewing the literature on the link between food dyes and hyperactivity. Studies evaluate hyperactivity largely by observations made by teachers, psychologists, and/or parents. Observations by trained professionals using defined scales are sometimes considered superior to observations by par... |
5313cc9c-1334-4578-9d9e-989a1509d9a0 | StampyAI/alignment-research-dataset/youtube | Youtube Transcripts | Predicting AI - Shanghai
thanks a lot for inviting me to the
conference and thanks a lot especially
for that introduction at the future of
humanity Institute as you said we try
and look at the big pictures the big
impacts in the world and hope to
convince you by the end of this talk
that one of the very biggest is in ... |
574bcebe-af2f-4493-8116-58d8e2d1d2d4 | trentmkelly/LessWrong-43k | LessWrong | Freakonomics Study Investigates Decision-Making and Estimated Prior Probabilities
The Freakonomics web site is currently conducting online research that appears, to this properly hypothesis-blinded participant, to be investigating decision-making and estimated prior probability of success. You can participate yourself... |
79d0b1d6-e56a-4f18-b366-3c3a9ed15660 | trentmkelly/LessWrong-43k | LessWrong | Why is o1 so deceptive?
The o1 system card reports:
> 0.8% of o1-preview’s responses got flagged as being ‘deceptive’ [...] roughly two thirds of which appear to be intentional (0.38%), meaning that there was some evidence in the chain of thought that o1-preview was aware that the answer was incorrect [...] Intention... |
87c01d7b-6694-469c-8166-91c00b837f3b | trentmkelly/LessWrong-43k | LessWrong | Thoughts on Iterated Distillation and Amplification
Alignment researcher Paul F. Christiano has written several posts on what he refers to as Iterated Distillation and Amplification (IDA). In this post, I will argue that IDA is a general method of adaptation and that it can be found in various different guises in a wi... |
90f8a0d9-3421-4b81-8ceb-f82e03991655 | trentmkelly/LessWrong-43k | LessWrong | What's the minimal additive constant for Kolmogorov Complexity that a programming language can achieve?
This is a question that I got from these posts, which essentially talked about how the constant issue impacts discussions of simplicity, and thus I wanted to ask a question about Kolmogorov Complexity:
https://www.... |
af4d40dc-80e3-40eb-99de-166b5598c4f9 | trentmkelly/LessWrong-43k | LessWrong | What money-pumps exist, if any, for deontologists?
Suppose I'm a classical utilitarian except that I have some deontological constraint I always obey, e.g. I never kill anyone with my bare hands. Is there a way to money-pump me?
(This question came out of a conversation with So8res) |
ebd06596-d895-4040-b3e3-c0f526009c80 | trentmkelly/LessWrong-43k | LessWrong | Logical Inductors contain Logical Inductors over other complexity classes
In the Logical Induction paper, we give a definition of logical inductors over polynomial time traders. It is clear from our definition that our use of polynomial time is rather arbitrary, and we could define e.g. an exponential time logical ind... |
80be3a4f-2bb5-4ba6-b116-33f50ed54c51 | trentmkelly/LessWrong-43k | LessWrong | Deconfusing Human Values Research Agenda v1
On Friday I attended the 2020 Foresight AGI Strategy Meeting. Eventually a report will come out summarizing some of what was talked about, but for now I want to focus on what I talked about in my session on deconfusing human values. For that session I wrote up some notes sum... |
05f644b3-8d77-4e69-957f-22ee1a41c124 | trentmkelly/LessWrong-43k | LessWrong | Bayesian Buddhism: a path to optimal enlightenment
I see epistemic rationality as a connection between Bayesian inference and Buddhism. Bayesian inference focuses on the mathematics for updating belief, Bayes' theorem. Buddhism focuses on the human factors of belief, how belief can lead to dukkha (suffering, disconten... |
a155a62d-413e-496e-9ba1-8418242307f3 | StampyAI/alignment-research-dataset/arbital | Arbital | Behaviorist genie
A behaviorist [genie](https://arbital.com/p/6w) is an AI that has been [averted from modeling](https://arbital.com/p/1g4) minds in more detail than some whitelisted class of models.
This is *possibly* a good idea because many [possible difficulties](https://arbital.com/p/6r) seem to be associated wi... |
fd07a768-81ce-4c07-b1c4-35f33ecd4555 | trentmkelly/LessWrong-43k | LessWrong | A Cruciverbalist’s Introduction to Bayesian reasoning
Status: Hopefully a nice introduction to some of the basics of Bayesian reasoning for newcomers.
> Clue: Mathematical methods inspired by an eighteenth century minister (8)
“Bayesian” is a word that has gained a lot of attention recently, though my experience te... |
615dd1d3-3055-4ebb-bdb6-b41bdaf77dea | trentmkelly/LessWrong-43k | LessWrong | The Spousal Chain of Succession
I told a bunch of people I'd post this idea here when I was writing it, then procrastinated on that. So here it is in all its glory. (Epistemic Status: Silly. Read at your own risk of wasting time). |
6c99f957-eaf2-4871-a6c8-4214995438f3 | trentmkelly/LessWrong-43k | LessWrong | Emotional regulation Part II: research summary
Abstract: Emotional regulation is a topic currently being studied in the field of psychology. Five different types of emotional regulation strategies have been identified, distinguished by the stage of the emotion-response process in which they occur. To drastically simpl... |
f86554b6-4f53-45b5-89e7-57919689cfac | trentmkelly/LessWrong-43k | LessWrong | 1. Premise one: Values are malleable
In this post, I will defend the first premise of the Value Change Problem: that realistic models of human values take them to be malleable, rather than fixed. I call this the ’value malleability’ claim (VM).
I start by clarifying what I mean by value malleability before making a... |
cd0a6335-56fa-49f1-b6f1-6841078a7350 | trentmkelly/LessWrong-43k | LessWrong | 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, are in fact not. And they are, in some sense, perfectly optimal.
And, somewhat conversely, that the reason we f... |
205584fe-a5b1-41cc-aaa0-ce53c77d6ada | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Reflexive decision theory is an unsolved problem
By "reflexive decision theory", hereafter RDT, I mean any decision theory that can incorporate information about one's own future decisions into the process of making those decisions. RDT is not itself a decision theory, but a class of decision theories, or a property o... |
2009515b-79a3-417a-a324-374b9075c423 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Microsoft Research Paper Claims Sparks of Artificial Intelligence in GPT-4
Microsoft Research (conflict of interest? what’s *that?)* has issued a 154-page report [entitled Sparks of Artificial Intelligence: Early Experiments With GPT-4](https://arxiv.org/pdf/2303.12712.pdf), essentially saying that GPT-4 could reasona... |
9a79de70-2102-4a8e-9197-e06c96a97640 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | [Link] Introducing OpenAI
From their site:
[OpenAI](https://openai.com/) is a non-profit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return.
The money quote is a... |
cfca9beb-905c-4f93-8a76-eb5b555f6506 | trentmkelly/LessWrong-43k | LessWrong | You are Dissociating (probably)
You are probably dissociating all the time and don't realize it. Let's look at just what I mean by this bold claim.
Disclaimer: I'm not a medical professional, and although I'm talking about "dissociation" here I definitely don't mean to give medical explanation or advice about clinica... |
c3db9ade-abde-40fc-86ed-66a11c1d5dff | trentmkelly/LessWrong-43k | LessWrong | [Linkpost] TrojanNet: Embedding Hidden Trojan Horse Models in Neural Networks
TrojanNet: Embedding Hidden Trojan Horse Models in Neural Networks
TL;DR:
> TrojanNet embeds a neural network within another neural network, and it's NP-hard to detect - which limits transparency.
Abstract:
> The complexity of large-sca... |
fc2c8743-f0e2-49b6-b8cf-a7f14af67388 | trentmkelly/LessWrong-43k | LessWrong | Felicifia: a Utilitarianism Forum
Utilitarianism seems to be a common theme on this site. I suggest checking out felicifia.org, a Utilitarianism forum. That is all. |
203d8123-3292-47ba-bab8-91c7d1ee69fa | trentmkelly/LessWrong-43k | LessWrong | Debate: Is it ethical to work at AI capabilities companies?
Epistemic status: Soldier mindset. These are not (necessarily) our actual positions, these are positions we were randomly assigned, and for which we searched for the strongest arguments we could find, over the course of ~1 hr 45 mins.
Sides: Ben was assigned... |
d3145c3d-cc1c-4fbe-b060-40f30bfc6015 | trentmkelly/LessWrong-43k | LessWrong | How to solve the argument about what the algorithm should do
Should YouTube's algorithm recommend conspiracy theory videos? Should twitter recommend tweets that disagree with the WHO? Should Facebook recommend things that'll make you angry?
There's countless such questions about recommender algorithms, usually with g... |
102d317d-cf4e-4a18-b4c3-900fb9441741 | trentmkelly/LessWrong-43k | LessWrong | [SEQ RERUN] Decoherence is Falsifiable and Testable
Today's post, Decoherence is Falsifiable and Testable was originally published on 07 May 2008. A summary (taken from the LW wiki):
> (Note: Designed to be standalone readable.) An epistle to the physicists. To probability theorists, words like "simple", "falsifia... |
6c294cd7-c9c6-47b7-9b75-a6dc10920f68 | trentmkelly/LessWrong-43k | LessWrong | Two explanations for variation in human abilities
In My Childhood Role Model, Eliezer Yudkowsky argues that people often think about intelligence like this, with village idiot and chimps on the left, and Einstein on the right.
However, he says, this view is too narrow. All humans have nearly identical hardware. There... |
48c9d366-1abd-4dd9-a495-21075edbb20d | trentmkelly/LessWrong-43k | LessWrong | Where is the Meaning?
> "You're being mean."
> "No, I'm not."
> "This album is amazing!"
> "What are you talking about? It's clearly trash."
> "Tomatoes are my favorite vegetable!"
> "Dude, they're fruits."
People often argue about what things are, and they often do it with words. Sometimes this can create another... |
c09b55b7-528e-4799-beb8-426247561d59 | trentmkelly/LessWrong-43k | LessWrong | A poem co-written by ChatGPT
Let's write a poem together!
That sounds wonderful! What should our poem be about?
I'll give you some lines and you continue, okay?
Sure, that sounds like a great way to collaborate! Please go ahead and give me your opening lines.
The ChatGPT Poem
How many books have you read in your ... |
2cee4249-965b-4474-be6c-a6ef6c34e62b | trentmkelly/LessWrong-43k | LessWrong | AI, Consciousness, and the problem of Moral Considerability
The problem I intend to work through in this short (hopefully) post is that of moral considerability in regard to artificial intelligence systems. This is not a problem that we should wait for advanced AI to consider and it doesn’t even really hinge on “advan... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.