id stringlengths 36 36 | source stringclasses 15
values | formatted_source stringclasses 13
values | text stringlengths 2 7.55M |
|---|---|---|---|
52fa4962-175a-4885-9bec-cd88f23aac25 | trentmkelly/LessWrong-43k | LessWrong | We Have No Plan for Preventing Loss of Control in Open Models
Note: This post is intended to be the first in a broader series of posts about the difficult tradeoffs inherent in public access to powerful open source models. While this post highlights some dangers of open models and discusses the possibility of global r... |
d4be2ecb-1803-476c-84cf-a6727b8f38b4 | StampyAI/alignment-research-dataset/youtube | Youtube Transcripts | AI Safety Landscape - Panel 1: The Challenge of Achieving Consensus
perspectives
and and perhaps come closer together in
their conceptions of habitation tree
structure but because of different
different different perspectives
so for instance your framework that the
core of it was that we need to
essentially constantly... |
60def4e1-f761-4c7b-bf4f-922f9a09f3d2 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Reward is Enough for Convex MDPs.
1 Introduction
---------------
In Reinforcement Learning (RL), an agent learns how to map situations to actions so
as to maximize a cumulative numerical reward signal. The learner is not told which actions to take, but instead must discover which actions lead to the most cumulative... |
809f7c88-918c-471a-adba-804a8e539198 | trentmkelly/LessWrong-43k | LessWrong | Covid-19 4/30: Stuck in Limbo
Epistemic Status: As with all my Covid-19 posts I’m not any kind of expert and am doubtless making a lot of mistakes. Yet it seems worthwhile to persist.
In previous posts I’ve been focused my direct modeling on New York. That was for various reasons. It’s the epicenter, it’s where I liv... |
f5bb4807-e95a-40d6-9974-94db9740257b | trentmkelly/LessWrong-43k | LessWrong | Are there good ways to find expert reviews of popular science books?
As the title says. Non-expert reviews only evaluate how good the writing was, or how impressed the reviewer was by the ideas; but I typically want something that evaluates the actual scientific claims.
My one trick has been to look at Google Scholar... |
2bd3a0b1-9398-46e1-8a79-66db120b4a61 | trentmkelly/LessWrong-43k | LessWrong | A Selection of Randomly Selected SAE Features
Epistemic status - self-evident.
In this post, we interpret a small sample of Sparse Autoencoder features which reveal meaningful computational structure in the model that is clearly highly researcher-independent and of significant relevance to AI alignment.
-----------... |
c81da72f-fdf8-4732-8789-db8b96a515bb | StampyAI/alignment-research-dataset/special_docs | Other | CHAI Newsletter #2 2021
5/4/22, 3:16 PM Center for Human-Compatible AI (CHAI) Newsletter
https://mailchi.mp/humancompatible.ai/aqhgtk12mu-9352805 1/8CHAI Newsletter
This newsletter includes CHAI activities from June 2021 to August 2021.
Research Highlight
On 7-8 June, CHAI held its fifth annual workshop. Over 150 atten... |
d9c9bfbb-34f3-4eb6-ac7c-ccaef6788d8d | trentmkelly/LessWrong-43k | LessWrong | Curing the World of Men
This is a post about the American Psychological Association's recent guidelines to treating men and boys. This post is not the bloodless academic assessment of those guidelines I would write on LessWrong, which is why it's a link post. This is a condemnation of those guidelines as an ideologica... |
0283cf14-576d-40cb-b8fa-7ce3b996a0b1 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Restoration of marker occluded hematoxylin and eosin stained whole slide histology images using generative adversarial networks
1 Introduction
---------------
Fig. 1:
The pipeline employed for training the model
It is a common practice for pathologists to an... |
7c32a7ba-c3af-4eb7-814d-b327c1b25fa4 | trentmkelly/LessWrong-43k | LessWrong | Vingean Agency
I've been involved with several discussions about different notions of agency (and their importance/relationships) lately, especially with the PIBBSS group including myself, Daniel, Josiah, and Ramana; see here.
There's one notion of agency (not necessarily "The" notion of agency, but a coherent and si... |
1214c880-84e4-4a13-9afc-bf8222f4c93e | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Apply to the ML for Alignment Bootcamp (MLAB) in Berkeley [Jan 3 - Jan 22]
We ([Redwood Research](https://www.redwoodresearch.org/) and [Lightcone Infrastructure](https://www.lightconeinfrastructure.com/)) are organizing a bootcamp to bring people interested in AI Alignment up-to-speed with the state of modern ML engi... |
9cdb1367-7df9-47f4-89ed-30f724573cf9 | trentmkelly/LessWrong-43k | LessWrong | Sometimes, talking the issue through *works*
Michael Nielsen's new book Reinventing Discovery is invigorating. Here's one passage on how a small group talked an issue through and had a large impact on scientific progress:
> Why is it that biologists share genetic data in GenBank in the first place? When you think abo... |
a13d33ff-fb51-471a-9bed-51a5573897bb | trentmkelly/LessWrong-43k | LessWrong | What spiritual experiences have you had?
The recent adversarial collaboration on spiritual experiences on Slate Star Codex includes this paragraph:
> It was also discovered that people in the United States, Australia, the United Kingdom, and Scandinavia do not tend to share their spiritual experiences with others. Ho... |
10ed512a-4730-4a95-b15d-ae1e062cacee | trentmkelly/LessWrong-43k | LessWrong | Internalizing Internal Double Crux
> In sciences such as psychology and sociology, internalization involves the integration of attitudes, values, standards and the opinions of others into one's own identity or sense of self.
Internal Double Crux is one of the most important skills I've ever learned. In the last two w... |
8680322c-b655-44f4-b0e9-6bd2b75ea3fc | StampyAI/alignment-research-dataset/arxiv | Arxiv | Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
|
8d6df549-ae36-4789-af02-ee8ede26a441 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Behavioral Sufficient Statistics for Goal-Directedness
*Note: this is a new version -- with a new title -- of my recent post "A Behavioral Definition of Goal-Directedness". Most of the formulas are the same, except for the triviality one that deals better with what I wanted; the point of this rewrite is to present the... |
3225e152-8e0c-4c5c-b84e-7c55203f8917 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | My thoughts on the social response to AI risk
A common theme implicit in many AI risk stories has been that broader society will either fail to anticipate the risks of AI until it is too late, or do little to address those risks in a serious manner. In my opinion, there are now clear signs that this assumption is fals... |
f91e65ad-1146-416c-91e0-91425d6cbb64 | trentmkelly/LessWrong-43k | LessWrong | OpenAI #12: Battle of the Board Redux
Back when the OpenAI board attempted and failed to fire Sam Altman, we faced a highly hostile information environment. The battle was fought largely through control of the public narrative, and the above was my attempt to put together what happened.
My conclusion, which I still b... |
ddcfc4cf-05bd-492b-8482-6095ef8100f0 | trentmkelly/LessWrong-43k | LessWrong | Genetic fitness is a measure of selection strength, not the selection target
Alternative title: "Evolution suggests robust rather than fragile generalization of alignment properties."
A frequently repeated argument goes something like this:
1. Evolution has optimized humans for inclusive genetic fitness (IGF)
2. ... |
9b6cb9c2-b86d-4c04-ac25-2c71a74d5a20 | trentmkelly/LessWrong-43k | LessWrong | Rashomon - A newsbetting site
Hello Rationalists!
A while ago I created a concept to evaluate political biases that I call "newsbetting" and made a demo site at rashomonnews.com. Essentially, one receives a news article stripped of identifying information (so stuff like title, author, news source, etc) so that they ... |
ef6ffcb4-8c29-46d9-917c-d74e240c8d30 | trentmkelly/LessWrong-43k | LessWrong | The time you have
Original post: http://bearlamp.com.au/the-time-you-have/
----------------------------------------
Part 1: Exploration-Exploitation
Part 2: Bargaining Trade-offs to your brain.
Part 2a: Empirical time management
Part 3: The time that you have
----------------------------------------
There is a ... |
74f1f044-36bb-428f-8719-2da577e6d7fe | trentmkelly/LessWrong-43k | LessWrong | March 16th: Daily Coronavirus Links
As part of the LessWrong Coronavirus Link Database, Ben, Elizabeth and I are publishing daily update posts with all the new links we are adding each day that we ranked a 3 or above in our importance rankings. Here are all the top links that we added yesterday (March 16th), by topic.... |
428e5520-950b-4e9b-ae51-542ec3be91e9 | trentmkelly/LessWrong-43k | LessWrong | Meetup : Baltimore / UMBC Meetup - usefulness and meaning of "truth"
Discussion article for the meetup : Baltimore / UMBC Meetup - usefulness and meaning of "truth"
WHEN: 12 February 2017 11:00:00AM (-0500)
WHERE: Performing Arts and Humanities Bldg 4th floor, 1000 Hilltop Cir, Baltimore, MD 21250
This week we'll b... |
ef148c82-decf-4d0d-9613-9f32e7f45049 | trentmkelly/LessWrong-43k | LessWrong | Childhood Roundup #2
Recently: The Kids are Not Okay, Childhood Roundup #1, Fertility Roundup #1
It seems like a good time to go over the interesting relevant stuff I’ve found since the last time I did one of these.
Executive Summary and Table of Contents:
1. Why Are the Kids are Not Okay. Additional fleshing out ... |
92248bde-6149-45c8-a6b9-2f8b0d66c5e2 | trentmkelly/LessWrong-43k | LessWrong | Prisoner's Dilemma (with visible source code) Tournament
After the iterated prisoner's dilemma tournament organized by prase two years ago, there was discussion of running tournaments for several variants, including one in which two players submit programs, each of which are given the source code of the other player's... |
b589029d-7172-4a86-a790-eaf58c2d765f | trentmkelly/LessWrong-43k | LessWrong | Entering At the 11th Hour (Babble & Anaylsis)
Background
It isn’t obvious what contributing to The Good Future from a non-standard background looks like.
Especially on a short timeline.
I am a first time poster with a background in the arts. I became aware of AI safety as a cause area relatively recently, and despit... |
bedb9e06-6c2f-4393-bd9f-d033eaeb4d97 | trentmkelly/LessWrong-43k | LessWrong | Some simple existential risk math
What's the best way to allocate resources towards decreasing existential risk?
Let's say we have n different risks. Define the functions r1(s1), r2(s2), ... , rn(sn), one for each risk. Each takes USD spent mitigating that particular risk as its input and outputs the probability that... |
a512ec85-3a17-4ed5-807a-590590161a1c | trentmkelly/LessWrong-43k | LessWrong | September 2010 Southern California Meetup
The second LessWrong meetup for Southern California will happen on Saturday September 25th, 2010! The meetup will start at 1PM and probably run for 4 or 5 hours in the Platt Campus Center of Harvey Mudd College. Thanks are due to the Harvey Mudd Future Tech Club for the loca... |
7b71da16-acb7-40f3-886d-3b2c1bb801ae | trentmkelly/LessWrong-43k | LessWrong | Abstractions as Redundant Information
What is it that makes the concept of “pencil” a good abstraction?
One way to frame it: there are lots of “pencils” in the world - lots of little chunks of the world which I could look at which all contain information about pencils-in-general. Many different places from which I co... |
f986e245-e4cf-4781-983d-dad7da770d71 | trentmkelly/LessWrong-43k | LessWrong | Starting Over: What to tell Sarah, at the edge of professional oblivion.
I came across a sad post recently from a woman on LinkedIn— let’s call her Sarah—who spent over a decade, in her words, as a top 2% recruiter, and was now finding it impossible to land even the most basic of roles. It was one of those earnest, h... |
6f39c1ca-9b2f-4db3-a3e9-616b1b66a6d9 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Understanding how hard alignment is may be the most important research direction right now
[Epistemic status: some fairly rough thoughts on alignment strategy. I am moderately confident (60%) that these claims are largely correct.]
---
As of today, most alignment research is directed at solving the problem of alig... |
9ed15167-ea91-41bc-a2ca-54b6900f8b9d | StampyAI/alignment-research-dataset/lesswrong | LessWrong | The Urgent Meta-Ethics of Friendly Artificial Intelligence
Barring a major collapse of human civilization (due to nuclear war, asteroid impact, etc.), many experts expect the [intelligence explosion Singularity](http://intelligence.org/media/thehumanimportanceoftheintelligenceexplosion) to occur within 50-200 years.
... |
bd47bc5a-a22b-4501-9cbf-2dd57a979b10 | trentmkelly/LessWrong-43k | LessWrong | How do you know it's time for a break?
My original question was something like "how much productive work can people do in a day?", but even if I got a signed piece of paper from God telling me the exact average, that wouldn't be very useful to individuals except as cover for stopping once they'd reached the average. S... |
53436add-a116-4d02-889c-6e2b74735e52 | trentmkelly/LessWrong-43k | LessWrong | A Rocket–Interpretability Analogy
1.
4.4% of the US federal budget went into the space race at its peak.
This was surprising to me, until a friend pointed out that landing rockets on specific parts of the moon requires very similar technology to landing rockets in soviet cities.[1]
I wonder how much more enthusias... |
a3182a98-c543-4f4d-9ec2-3fdb5a19bfde | trentmkelly/LessWrong-43k | LessWrong | Do Not Mess With Scarlett Johansson
I repeat. Do not mess with Scarlett Johansson.
You would think her movies, and her suit against Disney, would make this obvious.
Apparently not so.
> Andrej Karpathy (co-founder OpenAI, departed earlier), May 14: The killer app of LLMs is Scarlett Johansson. You all thought it wa... |
3b0d0f6c-99b2-44cc-8f97-9c33b76c58f8 | trentmkelly/LessWrong-43k | LessWrong | (Report) Evaluating Taiwan's Tactics to Safeguard its Semiconductor Assets Against a Chinese Invasion
This was written by Gaurav Yadav and Robert Reason for the Bristol AI Safety Centre.
Thank you to David Staley, Gabriel Kulp, Lucy Farnik, Oliver Guest, and Oliver Marks for their feedback and help on this report. Th... |
eba17d60-cf89-4ca1-a17b-5b3495c2327d | StampyAI/alignment-research-dataset/special_docs | Other | Artificial intelligence in a crisis needs ethics with urgency
[Download PDF](/articles/s42256-020-0195-0.pdf)
### Subjects
\* [Ethics](/subjects/ethics)
\* [Infectious diseases](/subjects/infectious-diseases)
\* [SARS-CoV-2](/subjects/sars-cov-2)
\* [Science, technology and society](/subjects/science-technolog... |
120a184e-ef6f-4035-9ae9-21e67204f929 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Delegative Reinforcement Learning with a Merely Sane Advisor
.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; whit... |
59eaa8cc-9908-4fff-8176-500a3b3b767f | trentmkelly/LessWrong-43k | LessWrong | Substitute goods for leisure are abundant
Meta: This is intended to be opinionated. I think opinionated things are often useful, they just need to be read with the right perspective. The arguments aren't air-tight, but to a certain audience the ideas might "click". Mr. Money Mustache comes to mind as an example of thi... |
cef605ac-dd03-4835-9b84-13cd28401e9a | trentmkelly/LessWrong-43k | LessWrong | Childhood and Education #11: The Art of Learning
In honor of the latest (always deeply, deeply unpopular) attempts to destroy tracking and gifted and talented programs, and other attempts to get children to actually learn things, I thought it a good time to compile a number of related items.
TABLE OF CONTENTS
1. L... |
9dfd7e1b-e549-4ab8-8454-96264fc7700a | trentmkelly/LessWrong-43k | LessWrong | “Artificial General Intelligence”: an extremely brief FAQ
(Crossposted from twitter for easier linking.) (Intended for a broad audience—experts already know all this.)
When I talk about future “Artificial General Intelligence” (AGI), what am I talking about? Here’s a handy diagram and FAQ:
“Are you saying that ChatG... |
5587bc0b-bf41-4385-abea-78ff091d00c4 | trentmkelly/LessWrong-43k | LessWrong | Housing and Transit Roundup #3
Previously: #1, #2
In honor of much of northern California coming face to face with the Builder’s Remedy, I’m doing this first, ahead of the monthly general roundup. Here we go.
The Federal Option
Biden announces he will hold federal housing funding hostage to the filing of an ‘equity... |
af06b178-0de2-4850-b2dc-8a3f7272e4f5 | trentmkelly/LessWrong-43k | LessWrong | The Field of AI Alignment: A Postmortem, and What To Do About It
> A policeman sees a drunk man searching for something under a streetlight and asks what the drunk has lost. He says he lost his keys and they both look under the streetlight together. After a few minutes the policeman asks if he is sure he lost them her... |
36422b30-81dc-42c8-8e44-15dad69c802f | trentmkelly/LessWrong-43k | LessWrong | Intent of Experimenters; Halting Procedures; Frequentists vs. Bayesians
|
44ca591c-56f4-4fa4-b60d-9bd660772ee9 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | A simple way of exploiting AI's coming economic impact may be highly-impactful
(This is an attempt at crossposting from LessWrong. Not sure if I'm doing this right)
Scott Alexander ended [A Lesswrong Crypto Autopsy](https://www.lesswrong.com/posts/MajyZJrsf8fAywWgY/a-lesswrong-crypto-autopsy) with these words
... |
2ffdf5e8-f53c-46b0-9b2e-9fc2eddaadf2 | trentmkelly/LessWrong-43k | LessWrong | [SEQ RERUN] Artificial Mysterious Intelligence
Today's post, Artificial Mysterious Intelligence was originally published on 07 December 2008. A summary (taken from the LW wiki):
> Attempting to create an intelligence without actually understanding what intelligence is, is a common failure mode. If you want to make... |
ad7aece2-0ad7-4041-b8f6-3f861c80c517 | trentmkelly/LessWrong-43k | LessWrong | Selective, Corrective, Structural: Three Ways of Making Social Systems Work
(A reference post for a concept that comes up often enough to warrant such a thing.)
----------------------------------------
> If men were angels, no government would be necessary. If angels were to govern men, neither external nor internal... |
265b678e-27f3-403a-8fe7-9156e846af21 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Transcript: Yudkowsky on Bankless follow-up Q&A
**Head over to Rob Bensinger's** [**improved**](https://www.lesswrong.com/posts/e4pYaNt89mottpkWZ/yudkowsky-on-agi-risk-on-the-bankless-podcast) **transcript, combined with the original podcast.** ***(This one has been updated correspondingly)*****.**
**---**
*This fol... |
98dedb37-a991-4f27-813f-273907500614 | StampyAI/alignment-research-dataset/youtube | Youtube Transcripts | AiTech Agora: Emma van Zoelen - Human AI Co-Learning Mutual Understanding and Smooth Collaboration
but before i started my phd because i
was looking for nice places that worked
on like human centered ai
uh type of stuff
um
so yeah very nice um so today i'll talk
about my phd work um and also just some
my general thou... |
05a20621-3a69-4206-ba93-d50ef06d432d | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Preserving and continuing alignment research through a severe global catastrophe
*[Epistemic status: Shallow dive into research questions, backed by some years of on-and-off thinking about this kind of plan.]*
Introduction
------------
There is some chance that civilization will cease to function before we hit an in... |
25248821-4bbd-4c08-b6a2-630cc2546c2a | trentmkelly/LessWrong-43k | LessWrong | Concrete vs Contextual values
The concept of recursive self-improvement is not an accepted idea outside of the futurist community. It just does not seem right in some fashion to some people. I am one of those people, so I'm going to try and explain the kind of instinctive skepticism I have towards it. It hinges on th... |
74e7d613-6099-4816-b581-c9907892558d | trentmkelly/LessWrong-43k | LessWrong | Meetup : Oslo Meetup at Bitraf.
Discussion article for the meetup : Oslo Meetup at Bitraf.
WHEN: 29 May 2015 05:30:00PM (+0200)
WHERE: Youngstorget 6, Oslo
Welcome to the second Lesswrong meetup, this time at a hackerspace.
The entrance is a bit quirky: a big green double door With "NILZ" in yellow writing over it... |
9fee0472-3f7d-4d81-9eb3-b8b6bde8fec8 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Decision Theories, Part 3.75: Hang On, I Think This Works After All
**Followup to:** [Decision Theories, Part 3.5: Halt, Melt and Catch Fire](/lw/e94/decision_theories_part_35_halt_melt_and_catch_fire/), [Decision Theories: A Semi-Formal Analysis, Part III](/lw/b7w/decision_theories_a_semiformal_analysis_part_iii/)
... |
a0908cf8-7150-466a-b22b-095a3ac94aab | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Practical anthropics summary
This is a simple summary of how to "do anthropics":
* If there are no issues of exact copies, or advanced decision theory, and the questions you're asking aren't weird, then use SIA. And by "use SIA", I mean "ignore the [definition of SIA](https://www.lesswrong.com/tag/self-indication-as... |
6c7537bd-de70-435c-b9c6-ef4e1a45bb02 | trentmkelly/LessWrong-43k | LessWrong | [link, 2019] AI paradigm: interactive learning from unlabeled instructions
Sent by a friend, I thought this was an interesting reference folks on lesswrong should know about. I have not attempted to grok this paper. If anyone thinks I should, please reply and let's schedule a time to read and discuss it in order to ex... |
4bbe368d-e277-4a56-8e45-a2b601eb54bd | trentmkelly/LessWrong-43k | LessWrong | Recommender Alignment for Lock-In Risk
Epistemic status: my own research and reasoning about lock-in risk threat models, and how recommender systems connect to the threat model outlined. I'm fairly confident in the claims about the contribution of recommender systems to filter bubbles, less so on extreme and persuasiv... |
21c95e6d-a9e7-4708-95b6-b5c972adc078 | trentmkelly/LessWrong-43k | LessWrong | "Drinking Alcohol May Significantly Enhance Problem Solving Skills"
saw this on reddit. Thought people here would be interested.
http://medicaldaily.com/news/20120411/9496/alcohol-solving-skills-analytical-thinking-creativity-study.htm
edit: from a comment in the reddit thread
http://www.reddit.com/r/scienc... |
fd391319-cc3c-4b6a-a686-7f68bb4c311d | trentmkelly/LessWrong-43k | LessWrong | The Generalized Anti-Zombie Principle
> "Each problem that I solved became a rule which served afterwards to solve other problems."
> —Rene Descartes, Discours de la Methode
"Zombies" are putatively beings that are atom-by-atom identical to us, governed by all the same third-party-visible physical laws, excep... |
ce451010-4927-4266-93b4-15126b84b469 | trentmkelly/LessWrong-43k | LessWrong | General-Purpose Questions Thread
Similar to the Crazy Ideas Thread and Diaspora Roundup Thread, I thought I'd try making a General-Purpose Questions Thread.
The purpose is to provide a forum for asking questions to the community (appealing to the wisdom of this particular crowd) in things that don't really merit t... |
1ac669dd-d451-4544-abcd-d062ed0b663f | trentmkelly/LessWrong-43k | LessWrong | Exercise in dissolving
A fun little exercise in dissolving a problem. Relatively quick, but it can wake you up on a slow day.
http://www.memedroid.com/share-meme/337837/8000
|
fc2c551b-69f5-4fd5-9f1a-06b4a0a7179e | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Finite Factored Sets: Inferring Time
The [fundamental theorem](https://www.lesswrong.com/posts/jr5kyRhNriCX2Ayyg/finite-factored-sets-polynomials-and-probability#5_5__The_Fundamental_Theorem_of_Finite_Factored_Sets) of [finite factored sets](https://www.lesswrong.com/s/kxs3eeEti9ouwWFzr) tells us that ([conditional](h... |
2ddef33a-1e20-478d-baba-6e16557dc79b | trentmkelly/LessWrong-43k | LessWrong | Tiling Agents for Self-Modifying AI (OPFAI #2)
An early draft of publication #2 in the Open Problems in Friendly AI series is now available: Tiling Agents for Self-Modifying AI, and the Lobian Obstacle. ~20,000 words, aimed at mathematicians or the highly mathematically literate. The research reported on was conduc... |
3f9b36fb-538c-4525-86b9-b2acad826f99 | trentmkelly/LessWrong-43k | LessWrong | What problem would you like to see Reinforcement Learning applied to?
We often talk about the dangers and challenges of AI and self-improving agents, but I'm curious what you view as potential beneficial applications of AI - if any! As a ML researcher I encounter a lot of positivity and hype in the field, so the very ... |
7503400e-0e26-4874-ae21-ab56a0abca5d | trentmkelly/LessWrong-43k | LessWrong | Infohazards, hacking, and Bricking—how to formalize these concepts?
This is an attempt to formulate a thought that’s been rolling around in my brain for a few years now. I want to formalize a hypothesis, but I’m not sure how to convey what the hypothesis even is. Hopefully by writing this out I’ll help make that happe... |
8329b86e-0aa2-41af-baa9-ad804939d382 | trentmkelly/LessWrong-43k | LessWrong | Live Sound: Big-O Improvements
In cost-conscious live sound scenes (ex: contra dance) for decades one monitor mix was standard and two was bonus. Then two was standard. Now it's common for even relatively low end systems to be able to give you more mixes than you need, and in some scenes a separate stereo mix for each... |
e1b71f29-ddb3-4a61-8ce2-05fd6dcd7971 | trentmkelly/LessWrong-43k | LessWrong | List of potential cognitive enhancement methods
Participants in the Singularity Summit 2011 workshops held on October 17-18 brainstormed a list of cognitive enhancement methods they would like to see tested — some of them for the first time, many of them more thoroughly than has been done so far. Here is that list:
... |
8b08bf87-11c6-4854-94ef-e74622c39c05 | trentmkelly/LessWrong-43k | LessWrong | Meetup : Pittsburgh: Harry Potter and the Methods of Rationality
Discussion article for the meetup : Pittsburgh: Harry Potter and the Methods of Rationality
WHEN: 18 May 2012 06:00:00PM (-0400)
WHERE: Cafe Phipps, 1 Schenley Park, Pittsburgh, PA
In time for happy hour and/or dinner time at Cafe Phipps. Phone 412-30... |
2c20dd15-8ce1-41dc-9b78-c5f0d5ef2cd0 | trentmkelly/LessWrong-43k | LessWrong | Rational insanity
My theory on why North Korea has stepped up its provocation of South Korea since their nuclear missle tests is that they see this as a tug-of-war.
Suppose that North Korea wants to keep its nuclear weapons program. If they hadn't sunk a ship and bombed a city, world leaders would currently be press... |
902f8946-461d-4aff-b66a-f9d9c353437d | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Rethink Priorities’ 2022 Impact, 2023 Strategy, and Funding Gaps
Key Takeaways and Introduction
==============================
Rethink Priorities’ mission is:
> *We address global priorities by researching solutions and strategies, mobilizing resources, and empowering our team and others.*
>
>
Our vision is:
... |
3c16e4e9-21d5-4a8d-971a-a3ea0cc0c5ae | trentmkelly/LessWrong-43k | LessWrong | Fighting Akrasia: Incentivising Action
Related To: Incremental Progress and the Valley, Silver Chairs, Paternalism, and Akrasia, How a pathological procrastinator can lose weight
Akrasia can strike anywhere, but one place it doesn't seem to strike too often or too severely, assuming you are employed, is in the work ... |
2ab27440-2c45-45e1-8cfa-193f31c93eda | StampyAI/alignment-research-dataset/lesswrong | LessWrong | ChatGPT's "fuzzy alignment" isn't evidence of AGI alignment: the banana test
*TL;DR; ChatGPT achieves seeming alignment by entirely sidestepping the hard questions of alignment and instead refusing to engage with prompts that even border on unaligned territory, instead repeating warm-and-fuzzy platitudes about helpful... |
a1833add-b864-4246-8d7b-7db43d7e6cd6 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Three pillars for avoiding AGI catastrophe: Technical alignment, deployment decisions, and coordination
Generated by Midjourney*Epistemic status: This model is loosely inspired by a conversation with Nate... |
3c3c506a-238f-4e1d-a77c-9aba65269da1 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | EleutherAI's GPT-NeoX-20B release
> GPT-NeoX-20B is a 20 billion parameter autoregressive language model whose weights will be made freely and openly available to the public through a permissive license. It is, to the best of our knowledge, the largest dense autoregressive model that has publicly available weights. In... |
0f73cec4-5233-496f-b66f-bb7f233c8863 | trentmkelly/LessWrong-43k | LessWrong | Speculations on the Future of Fiction Writing
If we look at major movies, writing is generally a relatively small chunk of the overall budget. A handful of examples from Wikipedia:
* Unbreakable: story rights + screenplay ~8.1% of budget
* Tomb Raider - Cradle of Life: story rights + screenplay ~3.4%
* Terminator ... |
ab1b15b7-4e07-4149-b877-30196519365f | trentmkelly/LessWrong-43k | LessWrong | Direct effects matter!
Note: this is from my personal blog here.
A strange phenomenon plagues public discourse. Subtle and largely detached from the culture war, it often manages to evade detection. Can you spot it in each of the following arguments or discussion points?
1. Drugs like cocaine and heroin are bad. Th... |
59d82438-024a-49a8-b4f4-ceaec43ad5e8 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | [Intro to brain-like-AGI safety] 12. Two paths forward: “Controlled AGI” and “Social-instinct AGI”
*Part of the* [*“Intro to brain-like-AGI safety” post series*](https://www.alignmentforum.org/s/HzcM2dkCq7fwXBej8)*.*
12.1 Post summary / Table of contents
=====================================
Thus far in the series, ... |
01af328c-e3c7-4438-8d6f-e93d4169d66c | trentmkelly/LessWrong-43k | LessWrong | How would you take math notes to make the most of them?
I've been heavily influenced by reading about the zettelkasten method (and its siblings); it seems to, theoretically, solve many problems that I (a bioinformatics student) encounter on a daily basis.
However, I'm struggling with implementing said method in pract... |
903ae191-ddbf-4855-b386-e29ba42c203b | StampyAI/alignment-research-dataset/arxiv | Arxiv | TanksWorld: A Multi-Agent Environment for AI Safety Research
1 Introduction
---------------
Emerging paradigms in machine learning (e.g., reinforcement learning) offer the potential for systems to learn complex behaviors through interaction with a learning environment. These advancements could help overcome the lim... |
8ceafbfb-53aa-4c79-ab93-9d16d7134f30 | trentmkelly/LessWrong-43k | LessWrong | Sneaking Suspicion
Over time I've learned to pay more attention to a sensation I think of as "Sneaking Suspicion."
Sneaking Suspicion is a "flavor" that some of my thoughts come with: it's a quiet little (metaphorical) voice that whispers "wait, but, are you sure?"
Lots of other voices in my head are much louder tha... |
3fe1db6c-63bc-4041-9250-153bd55b423d | trentmkelly/LessWrong-43k | LessWrong | Book Review: Reinforcement Learning by Sutton and Barto
Note: I originally submitted this for the Slate Star Codex (RIP) book review contest, but given the current drama it probably won't happen for a while (if at all), so I decided to share this review with the LW community in the meantime.
Cross-posted on my blog: ... |
9868ac03-f5b3-442d-b292-b14ddb06a726 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | [Cross-post] Change my mind: we should define and measure the effectiveness of advanced AI
Epistemic status: trying to bait you into responding ([original](https://clarifyingconsequences.substack.com/p/half-baked-ideas-defining-and-measuring?s=w)).
Artificial Intelligence safety researchers are concerned with the beh... |
2f19827c-d679-4d30-ac94-ec24d5f06a38 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Half-baked ideas thread (EA / AI Safety)
*[Note that I have also* [*cross-posted a version of this post*](https://www.lesswrong.com/posts/5WPwNvfYnh78QSxCA/half-baked-ai-safety-ideas-thread) *to LessWrong. This version of the post is for both half-baked EA ideas and half-baked AI Safety ideas, whereas the LessWrong ve... |
82f78254-2b6a-459e-a48e-be69c6b24894 | trentmkelly/LessWrong-43k | LessWrong | Fractals and time management
As you might know, fractal structures appear in a variety of natural situations and have found many technical applications (see Wikipedia for more information and examples). In this short article I want to ask the question, whether it makes sense to structure various activities according t... |
f225d314-a38c-4e64-a7c4-1f102808d440 | trentmkelly/LessWrong-43k | LessWrong | Monthly Roundup #2
It’s that time again. This is everything that doesn’t fall into a category and isn’t worthy of its own post.
Bad News
Scam sophistication, especially against the most vulnerable like the mentally impaired, continues to ramp up as one would expect given current technology, and AI is only going to m... |
320edc13-a22b-48a8-8741-7bab9ceb274c | trentmkelly/LessWrong-43k | LessWrong | Spaced Repetition Systems for Intuitions?
I often see discussion here of Anki or other Spaced Repetition Systems for learning. To me, these are great for "higher levels" of knowledge -- things more closely associated with System 2 than System 1.
Are there similar advantages to be had (and systems that exist for) "lo... |
0e787030-f92e-4a7a-ba25-3aca552ad1ca | trentmkelly/LessWrong-43k | LessWrong | Graceful Degradation
There’s a concept I think about when teaching, which I call Graceful Degradation. The basic idea is, how well does this lesson work if someone doesn’t remember it very well or if I teach it badly?
I picked up Graceful Degradation as an engineer, and you might be familiar with it from that milieu.... |
75bc3e9b-bdaa-4986-9cdf-61e1e1b1164a | trentmkelly/LessWrong-43k | LessWrong | How hard would it be to attack coronavirus with CRISPR?
Crisper can snip genomes in particular spots, we probably have the genome of corona virus. Viruses probably don't work if their genome is chopped up. What would happen if you loaded random chunks of the coronavirus genome into crispr, and injected it into someone... |
ca72a192-81f4-4243-bd4d-c0f3a98c637d | trentmkelly/LessWrong-43k | LessWrong | Being a donor for Fecal Microbiota Transplants (FMT): Do good & earn easy money (up to 180k/y)
TLDR
Help to find super donors for Fecal Microbiota Transplants (FMT)! By donating stool for FMT, exceptionally healthy people can earn $180,000/yr with little effort. Simultaneously they can help chronically ill people ... |
b9374ff7-2864-46f0-9819-ef7595531f24 | trentmkelly/LessWrong-43k | LessWrong | Bragging Thread, June 2014
Your job, should you choose to accept it, is to comment on this thread explaining the most awesome thing you've done this month. You may be as blatantly proud of yourself as you feel. You may unabashedly consider yourself the coolest freaking person ever because of that awesome thing you're ... |
131fc136-ff86-43fa-949b-da42e284e41c | trentmkelly/LessWrong-43k | LessWrong | Weekly LW Meetups
This meetup summary was posted to LW main on April 11th. The following week's summary is here.
New meetups (or meetups with a hiatus of more than a year) are happening in:
* Las Vegas/Henderson, NV: 01 May 2014 06:00PM
Irregularly scheduled Less Wrong meetups are taking place in:
* Atlanta: Pro... |
905d59db-5419-43a0-8185-8092ea809545 | trentmkelly/LessWrong-43k | LessWrong | How LDT helps reduce the AI arms race
(Epistemic status: I think this is right?)
Alice is the CEO of ArmaggAI, and Bob is the CEO of BigModelsAI, two major AI capabilities organizations. They're racing to be the first to build a superintelligence aligned to their respective CEV which would take over the universe and ... |
9463cd39-8b4d-47c8-9fa9-58ff943a82b3 | trentmkelly/LessWrong-43k | LessWrong | Use and misuse of models: case study
Some time ago, I discovered a post comparing basic income and basic job ideas. This sought to analyse the costs of paying everyone a guaranteed income versus providing them with a basic job with that income. The author spelt out his assumptions and put together a two models with a ... |
d7d4f5e5-504f-46ac-8674-494840d513ac | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Unsafe AI as Dynamical Systems
*[Thanks to Valerie Morris for help editing this post.]*
Overview
--------
Large Language Models (LLMs) and their safety properties are often studied from the perspective of [a single pass](https://transformer-circuits.pub/): what is the single next token the LLM will produce? But this... |
2cbbc7f4-9373-46f9-a8a5-2e175415a11a | trentmkelly/LessWrong-43k | LessWrong | The Danger of Stories
Tyler Cowen argues in a TED talk (~15 min) that stories pervade our mental lives. He thinks they are a major source of cognitive biases and, on the margin, we should be more suspicious of them - especially simple stories. Here's an interesting quote about the meta-level:
> What story do you ta... |
52317f16-c108-4e1e-91eb-427c0a7ee8e7 | StampyAI/alignment-research-dataset/blogs | Blogs | October 2017 Newsletter
“So far as I can presently estimate, now that we’ve had AlphaGo and a couple of other maybe/maybe-not shots across the bow, and seen a huge explosion of effort invested into machine learning and an enormous flood of papers, we are probably going to occupy our present epistemic state until very ... |
8394205b-e349-41f1-8b53-a80523f14517 | trentmkelly/LessWrong-43k | LessWrong | My scorched-earth policy on New Year’s resolutions
(forgive me if these concepts are too basic; this was originally written for normies)
Welp, at this time of year everyone either writes a “How to set and keep your New Year’s resolution” article, or a “Why you shouldn’t set New Year’s resolutions” article.
This is t... |
69485f99-bafc-42ce-944b-47caed4eb9df | StampyAI/alignment-research-dataset/blogs | Blogs | systems and diversity
systems and diversity
---------------------
as i've said in [a previous post](lets-not-generalize-politics.html): i really like culture; and, to that end, i like diversity (by which i mean people being more weird and different from one another).
there are many systems that exist today that af... |
453e4a14-7a85-41d2-a5bf-21894fefd523 | trentmkelly/LessWrong-43k | LessWrong | Are the bad epistemic conditions global?
You've set this in the context of USA politics and society. Clearly it is not just limited to the USA. I seem to recall similar types of things occurring in Europe over the past 20 - 30 years.
I also suspect one might find this type of problem more prevalent within the environ... |
44e3ef54-26de-4512-8946-35be387ad77f | trentmkelly/LessWrong-43k | LessWrong | Does anyone have an anki deck on ageing / rejuvenation biotechnology that they could share?
Thanks in advance! |
ec88eff1-5d9d-4da7-9987-931ba4a394cb | trentmkelly/LessWrong-43k | LessWrong | [Letter] Advice for High School #2
> This is amazing. Thank you, so, so much. I'll start coding in Hy with Vim. I'll start doing weekly blogs once my tennis season ends (three weeks from now), and I likely wouldn't have the guts to start if it wasn't for your advice. On the general high school advice; it's invaluable.... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.