id stringlengths 36 36 | source stringclasses 15
values | formatted_source stringclasses 13
values | text stringlengths 2 7.55M |
|---|---|---|---|
4681652e-3986-47cb-b96b-685f84345dc3 | trentmkelly/LessWrong-43k | LessWrong | Insights from Euclid's 'Elements'
Presumably, I was taught geometry as a child. I do not remember.
Recently, I'd made my way halfway through a complex analysis textbook, only to find another which seemed more suitable and engaging. Unfortunately, the exposition was geometric. I knew something was wrong – I knew somet... |
998d664c-18ea-410d-acc9-9e8142d03457 | trentmkelly/LessWrong-43k | LessWrong | I Think Eliezer Should Go on Glenn Beck
Glenn Beck is the only popular mainstream news host who takes AI safety seriously. I am being entirely serious. For those of you who don't know, Glenn Beck is one of the most trusted and well-known news sources by American conservatives.
Over the past month, he has produced tw... |
5ee14d63-32c7-45ee-b5d9-1254d26d2bca | trentmkelly/LessWrong-43k | LessWrong | Any interest in a one-time meetup in Charlotte, NC this Tuesday?
I am in Charlotte, NC for the week due to a family emergency and wondered if anyone from Less Wrong wanted to get together for dinner this Tuesday night? There is a small chance that my emergency will come to a head on Tuesday and I won't be able to atte... |
4f7fc039-b122-49ad-9432-b7119311ccd9 | trentmkelly/LessWrong-43k | LessWrong | Pascal’s Muggle Pays
Reply To (Eliezer Yudkowsky): Pascal’s Muggle Infinitesimal Priors and Strong Evidence
Inspired to Finally Write This By (Lesser Wrong): Against the Linear Utility Hypothesis and the Leverage Penalty.
The problem of Pascal’s Muggle begins:
> Suppose a poorly-dressed street person asks you for f... |
a6ea58e7-5091-4c3f-85f6-78c3849ac532 | trentmkelly/LessWrong-43k | LessWrong | Why self-improvement?
I spend a lot of time on self-improvement related things. For example, I listen to a lot of podcasts like Lex Friedman, Peter Attia and Andrew Huberman. In fact, I just finished listening to Lex's episode with expert negotiator Chris Voss.
Why do I do this? Well, it feels good. Why does it feel ... |
4063b6f7-f3a6-465e-84ad-e6135fba4fb2 | StampyAI/alignment-research-dataset/youtube | Youtube Transcripts | DeepMind x UCL | Deep Learning Lectures | 7/12 | Deep Learning for Natural Language Processing
hello and welcome to the UCL and
deepmind lecture series
my name's Felix Hill and I'm going to be
talking to you about deep learning and
language understanding so here's an
overview of the structure of today's
talk it's goi... |
f7e52160-6aaa-480e-9ceb-bc88d2d6d6ab | trentmkelly/LessWrong-43k | LessWrong | Request for feedback: paper on fine-tuning and the multiverse hypothesis
A while back, I posted in the "What are you working on?" thread about a paper I was working on. A few people wanted to see it once I have a complete draft, and I'm of course independently interested in obtaining feedback before I move on with it.... |
2149e313-f34d-4d20-bfe7-a86c4bb2a6a4 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Immobile AI makes a move: anti-wireheading, ontology change, and model splintering
Research projects
-----------------
I'm planning to start two research projects on [model splintering/reward generalisation](https://www.lesswrong.com/posts/k54rgSg7GcjtXnMHX/model-splintering-moving-from-one-imperfect-model-to-anothe... |
293dc99d-a9d6-4ff0-a959-c85ff5331c62 | trentmkelly/LessWrong-43k | LessWrong | How rapidly are GPUs improving in price performance?
None |
2d1e24e8-331e-48b1-8a1c-82b2440b3e82 | trentmkelly/LessWrong-43k | LessWrong | Maybe Antivirals aren’t a Useful Priority for Pandemics?
PLEASE KEEP COMMENTS GENERALLY ON THE TOPIC OF ANTIVIRALS.
Epistemic status: Building a better gears-level understanding of why antivirals don’t work very well, explaining why portfolio construction for technology isn’t the same as investing in markets, then sp... |
3d86e585-c018-4da5-a575-5a486dda457c | trentmkelly/LessWrong-43k | LessWrong | Inner and outer alignment decompose one hard problem into two extremely hard problems
TL;DR: One alignment strategy is to 1) capture “what we want” in a loss function to a very high degree (“robust grading”), 2) use that loss function to train the AI, and 3) get the AI to exclusively care about optimizing that objecti... |
ce6af68c-117c-4f6a-8352-e1899d246175 | StampyAI/alignment-research-dataset/arxiv | Arxiv | From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following
1 Introduction
---------------
0: Tl;dr.
* A problem with the obvious-seeming "wizard's code of honesty" aka "never say things that are false" is that it draws on high verbal intelligence and unusually permissive social embeddings. I.e., you can't always say "Fi... |
a0596e24-cb6d-4c69-b588-cb56e3474afd | trentmkelly/LessWrong-43k | LessWrong | How should a rational agent construct their utility function when faced with existence?
Intuitively there are some utility functions that should be discarded based on principle of indifference. For example, my utility function shouldn't change based on my angular position with respect to the sun (given everything else... |
34320acb-debf-4cd3-ba75-b47c14449016 | trentmkelly/LessWrong-43k | LessWrong | On What Selves Are - CEV sequence
The CEV Sequence Summary: The CEV sequence consists of three posts tackling important aspects of Coherent Extrapolated Volition (CEV). It covers conceptual, practical and computational problems of CEV's current form. On What Selves Are draws on analytic philosophy methods in order to ... |
9b56c623-a871-4efb-844d-9e236a88d9c9 | trentmkelly/LessWrong-43k | LessWrong | How I started believing religion might actually matter for rationality and moral philosophy
After the release of Ben Pace's extended interview with me about my views on religion, I felt inspired to publish more of my thinking about religion in a format that's more detailed, compact, and organized. This post is the fir... |
f86b689e-8eb2-416e-8820-9bdae2067d7a | trentmkelly/LessWrong-43k | LessWrong | Analysing Adversarial Attacks with Linear Probing
This work was produced as part of the Apart Fellowship. @Yoann Poupart and @Imene Kerboua led the project; @Clement Neo and @Jason Hoelscher-Obermaier provided mentorship, feedback and project guidance.
Here, we present a qualitative analysis of our preliminary resul... |
58bfc493-e26d-4583-8cfa-c55b31c2bcb7 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Meta-execution
This post describes meta-execution, my current proposal for [capability amplification](https://medium.com/ai-control/policy-amplification-6a70cbee4f34) and [security amplification](https://medium.com/ai-control/security-amplification-f4931419f903#.xb6xc8jaa).
(*Meta-execution is* *[annotated functional... |
078e1d6a-55c8-49d5-91ae-d49afada64e9 | trentmkelly/LessWrong-43k | LessWrong | Debug Headers with GET
HTTP offers a convenient way to download only the headers: send a HEAD request:
$ telnet www.example.com 80
Trying 93.184.216.34...
Connected to www.example.com.
Escape character is '^]'.
HEAD / HTTP/1.1
Host: www.example.com
HTTP/1.1 200 OK
Content-Encoding: gzip
Accept-Ranges: bytes
Age: 32... |
80231c17-cf71-4c5b-a50d-8bcd3ee9db9c | trentmkelly/LessWrong-43k | LessWrong | Covid 6/23/22: Under Five Alive
America has now approved the Covid vaccine for children as young as six months old. Should you vaccinate your very young child?
Yes, I believe that you should. I don’t think you absolutely have to do that. Children that young are at minimal risk from Covid. However, vaccination decreas... |
b7b16b8d-7c11-41f0-92d4-bd5607eb719b | trentmkelly/LessWrong-43k | LessWrong | [Letter] Russians are Welcome in America
Edit [to clarify]: American immigration policy is unwelcoming to immigrants. Often, the only ways to enter the US legally are to marry someone in the US or to get the sponsorship of a powerful tech company. It takes several months even under the best of circumstances. I just wa... |
523a7721-9ddb-48eb-b626-77e465845290 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | The Blackmail Equation
*This is Eliezer's model of blackmail in decision theory at the recent workshop at SIAI, filtered through my own understanding. Eliezer help and advice were much appreciated; any errors here-in are my own.*
The mysterious stranger blackmailing the Countess of Rectitude over her extra-marital a... |
4461f098-aadd-4cd3-954e-7d6c9226b781 | trentmkelly/LessWrong-43k | LessWrong | Meetup : Houston Hackerspace Meetup
Discussion article for the meetup : Houston Hackerspace Meetup
WHEN: 07 August 2011 01:00:00PM (-0700)
WHERE: 2010 Commerce St, Houston, Tx. 77002
A social meetup at the Houston hackerspace. We'll probably be playing munchkin, Settlers of Cataan, or some other type of game until ... |
d64f31dc-7d8c-473d-95a2-293fd6e70d38 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | On dollars, utility, and crack cocaine
The lottery came up in a recent comment, with the claim that the expected return is negative - and the implicit conclusion that it's irrational to play the lottery. So I will explain why this is not the case.
It's convenient to reason using units of equivalent value. Dollars,... |
9d208a10-ed44-4a89-9abb-d160d57cf260 | trentmkelly/LessWrong-43k | LessWrong | The environment as infrastructure
A good metaphor for the ideal relationship between humanity and the environment is that the environment is like critical infrastructure.
Infrastructure is valuable, because it provides crucial services. You want to maintain it carefully, because it’s bad if it breaks down.
But infra... |
f657a314-affc-4668-aae9-21379015cfb1 | trentmkelly/LessWrong-43k | LessWrong | [SEQ RERUN] The Magnitude of His Own Folly
Today's post, The Magnitude of His Own Folly was originally published on 30 September 2008. A summary (taken from the LW wiki):
> Eliezer considers his training as a rationalist to have started the day he realized just how awfully he had screwed up.
Discuss the post her... |
61fa9a47-331b-4e28-bc57-7a39b2e04ca7 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Shh, don't tell the AI it's likely to be evil
If you haven't been online in the past week: [ChatGPT](https://openai.com/blog/chatgpt/) is OpenAI's new fine-tuned language model that is really much better at doing all sorts things that you'd expect. Here's it [debugging some code](https://twitter.com/amasad/status/1598... |
523a4b62-ed2a-4673-8856-d2200b53c02f | trentmkelly/LessWrong-43k | LessWrong | Is Morality a Valid Preference?
In general, the ethical theory that prevails here on Less Wrong is preference utilitarianism. The fundamental idea is that the correct moral action is the one that satisfies the strongest preferences of the most people. Preferences are discussed with units such as fun, pain, death, tort... |
1a8b6a6f-6e4d-4466-8de5-44f1ad9dd4b1 | trentmkelly/LessWrong-43k | LessWrong | New LW Meetup: Bay City, MI
This summary was posted to LW Main on June 24th. The following week's summary is here.
New meetups (or meetups with a hiatus of more than a year) are happening in:
* Bay City Meetup: 30 June 2016 03:23PM
Irregularly scheduled Less Wrong meetups are taking place in:
* Baltimore Weekly ... |
5fc4dc13-5c0e-41d6-8e46-c4129ef2dc64 | trentmkelly/LessWrong-43k | LessWrong | Git Re-Basin: Merging Models modulo Permutation Symmetries [Linkpost]
See Arxiv and Twitter. From the abstract:
> We argue that neural network loss landscapes contain (nearly) a single basin, after accounting for all possible permutation symmetries of hidden units. We introduce three algorithms to permute the units o... |
7ad6c532-1f9f-4cfe-a306-dd2a556ac618 | trentmkelly/LessWrong-43k | LessWrong | Michael Nielsen explains Judea Pearl's causality
Michael Nielsen has posted a long essay explaining his understanding of the Pearlean causal DAG model. I don't understand more than half, but that's much more than I got out of a few other papers. Strongly recommended for anyone interested in the topic. |
32869437-bc48-4b4f-8d6f-dfdaff50c395 | trentmkelly/LessWrong-43k | LessWrong | Dominant Assurance Contract Experiment #2: Berkeley House Dinners
As an afterthought at the end of my previous post on dominant assurance contracts, I decided on a whim as I was typing to append a small dominant assurance contract:
> In theory, writers could kickstart posts using dominant assurance contracts. An exam... |
4b84bd06-dbf6-419f-b937-dbefe2fe5f2e | trentmkelly/LessWrong-43k | LessWrong | A Platform for Falsifiable Conjectures and Public Refutation — Would This Be Useful?
What would online discourse look like if it enforced falsifiability as a first-class constraint?
I'm considering building a simple platform. You can:
• Post a conjecture — a falsifiable claim, written clearly.
• Include falsificati... |
34669504-4d7a-41fe-a3cf-65f234ed7f93 | trentmkelly/LessWrong-43k | LessWrong | Drinking to lower standards?
One way to be more satisfied with life is to lower your standards. People seem pretty hesitant to do this most of the time. And fair enough: who wants to be satisfied at the expense of everything else they care about? Happiness isn’t that great.
If only it were possible to feel like you h... |
b7f8f96b-25ea-4d95-8651-d1883dd7a7fe | trentmkelly/LessWrong-43k | LessWrong | Information vs Assurance
In contract law, there’s this thing called a “representation”. Example: as part of a contract to sell my house, I might “represent that” the house contains no asbestos. How is this different from me just, y’know, telling someone that the house contains no asbestos? Well, if it later turns out ... |
2fbc8e15-283b-47d5-8968-7773fce2095b | trentmkelly/LessWrong-43k | LessWrong | How I Learned That You Should Push Children Into Ponds
Crosspost of this, on my blog.
Effective altruism, the capitalist ultracapitalist movement in favor of capitalism and capitalism, is a white cishet settler colonialist movement. It talks a big game about doing good effectively but then some people in the moveme... |
edb97e75-b7cb-481b-af7c-884afad2c566 | trentmkelly/LessWrong-43k | LessWrong | [SEQ RERUN] Statistical Bias
Today's post, Statistical Bias was originally published on March 30, 2007. A summary (from the LW wiki):
> There are two types of error, systematic error, and random variance error; by repeating experiments you can average out and drive down the variance error.
This post is part of a se... |
a684258b-b61b-494a-81e6-5396aafe37bf | trentmkelly/LessWrong-43k | LessWrong | Concept extrapolation: key posts
Concept extrapolation is the skill of taking a concept, a feature, or a goal that is defined in a narrow training situation... and extrapolating it safely to a more general situation. This more general situation might be very extreme, and the original concept might not make much sense ... |
52883e4a-7ca7-481c-a69d-97f47a683e44 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Machine Learning Projects on IDA
### TLDR
We wrote a 20-page [document](http://owainevans.github.io/pdfs/evans_ida_projects.pdf) that explains IDA and outlines potential Machine Learning projects about IDA. This post gives an overview of the document.
### What is IDA?
Iterated Distillation and Amplification (IDA... |
68b870b9-9542-4265-ba12-8fda68db6086 | StampyAI/alignment-research-dataset/blogs | Blogs | The View from 30,000 Feet: Preface to the Second EleutherAI Retrospective
Over a year and a half have passed since EleutherAI's last retrospective, and a great deal of things have changed. [In the first year, what started off as a Discord server created by some TPU enthusiasts grew into a much larger and more vibrant ... |
f1d6b8d7-53b5-4893-ba3e-db70a6e87c6c | trentmkelly/LessWrong-43k | LessWrong | EDT with updating double counts
I recently got confused thinking about the following case:
Calculator bet: I am offered the opportunity to bet on a mathematical statement X to which I initially assign 50% probability (perhaps X = 139926 is a quadratic residue modulo 314159). I have access to a calculator that is 99% ... |
23a02491-fa2f-4697-9646-3d7a13ae1ac0 | trentmkelly/LessWrong-43k | LessWrong | Book Review: Ages Of Discord
I.
I recently reviewed Secular Cycles, which presents a demographic-structural theory of the growth and decline of pre-industrial civilizations. When land is plentiful, population grows and the economy prospers. When land reaches its carrying capacity and income declines to subsistence, t... |
79ed3099-f1b4-4af4-96b2-0f45ed6136e1 | trentmkelly/LessWrong-43k | LessWrong | Is there a website for tracking fads?
Also, clever ways to use google trends etc?
|
3f9059be-7148-4cec-9e5d-8d1751bf4206 | trentmkelly/LessWrong-43k | LessWrong | Compute Research Questions and Metrics - Transformative AI and Compute [4/4]
Cross-posted here on the EA Forum.
Transformative AI and Compute - A holistic approach - Part 4 out of 4
This is the Appendix (part four) of the series Transformative AI and Compute - A holistic approach. You can find the sequence here and ... |
57cbccc5-a1cc-4530-9aa7-eb711f64e3d2 | trentmkelly/LessWrong-43k | LessWrong | Live Kingfisher Album?
I'm pretty happy with how Kingfisher sounds at this point: with fifteen gigs since dances restarted and eight in the last six months I feel like our playing has gelled a lot. I've been recording our dances and listening back there's a lot of good stuff. Which has me wondering about making a live... |
e9629c74-2114-406c-81d3-56903735360b | trentmkelly/LessWrong-43k | LessWrong | The Three Little Piggies of Rationality
There are three related logical fallacies, which I call the Three Little Piggies of rationality. A strawman is when you argue against a simplified view that your opponent doesn’t have. A steelman is when you argue against a more sophisticated view than the one your opponent has.... |
a0aa7132-7068-4dd6-8a1b-00f3ecdb5176 | trentmkelly/LessWrong-43k | LessWrong | Machine Learning Model Sizes and the Parameter Gap [abridged]
Summary: The model size of notable Machine Learning systems has grown ten times faster than before since 2018. After 2020 growth has not been entirely continuous: there was a jump of one order of magnitude which persists until today. This is relevant for fo... |
079e4e25-c79c-49d7-900d-f3f1003b91d5 | trentmkelly/LessWrong-43k | LessWrong | Societal and technological progress as sewing an ever-growing, ever-changing, patchy, and polychrome quilt
Joel Z. Leibo [1], Alexander Sasha Vezhnevets [1], William A. Cunningham [1, 2], Sébastien Krier [1], Manfred Diaz [3], Simon Osindero [1]
[1] Google DeepMind, [2] University of Toronto, [3] Mila Québec AI Insti... |
9bffff5f-c97d-4dfb-b780-eb383e579202 | trentmkelly/LessWrong-43k | LessWrong | Getting 50% (SoTA) on ARC-AGI with GPT-4o
I recently got to 50%[1] accuracy on the public test set for ARC-AGI by having GPT-4o generate a huge number of Python implementations of the transformation rule (around 8,000 per problem) and then selecting among these implementations based on correctness of the Python progra... |
71bc275e-d47a-4134-8949-a25dbfb38a26 | trentmkelly/LessWrong-43k | LessWrong | What's the best way for me to improve my English pronounciation?
I'm a native German speaker and learned a good portion of my English not in school but by reading English content.
As part of a New Years event we did a Doom Circle and one essential feedback I got was that I have to improve my English accent.
When it ... |
890c024c-a4d7-4af1-bd61-0334626fee2f | StampyAI/alignment-research-dataset/arbital | Arbital | Decit
A decit, or "decimal digit" (also known as a "dit", "ban", or "hartley") is the amount of [information](https://arbital.com/p/3xd) required to single out one thing from a set of ten. That is, a decit is a measure of information equal to $\log_2(10)\approx 3.32$ [shannons](https://arbital.com/p/shannon). |
b8e7f017-d62d-47d5-832b-545578b04298 | trentmkelly/LessWrong-43k | LessWrong | Did the recent blackmail discussion change your beliefs?
Various rationalist blogs and Less Wrong have recently posted on and discussed blackmail, and specifically legality and acceptability of such. I found the discussion unsatisfying, and I'm trying to understand why that is, and whether I'm alone in that.
As it wa... |
74e17249-0549-4bc6-9855-e5e375513f19 | trentmkelly/LessWrong-43k | LessWrong | The Best Bits From Build, Baby, Build
Build, Baby, Build is George Mason Economist Bryan Caplan’s latest book and second graphic novel. This book makes the case that government regulation of housing is at the center of our most important economic and social problems. It’s a fun book and a quick read. If you’re already... |
c366dd45-fc49-404f-baad-60b653fbd92b | StampyAI/alignment-research-dataset/distill | Distill Scientific Journal | Visualizing Neural Networks with the Grand Tour
The Grand Tour is a classic visualization technique for high-dimensional point clouds that *projects* a high-dimensional dataset into two dimensions.
Over time, the Grand Tour smoothly animates its projection so that every possible view of the dataset is (eventually) p... |
917f421f-b59b-429e-969d-fcf07541baf7 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Update on Ought's experiments on factored evaluation of arguments
[Ought](https://ought.org/) has written a detailed update and analysis of recent experiments on factored cognition. These are experiments with human participants and don’t involve any machine learning. The goal is to learn about the viability of [IDA](h... |
9bb7eee3-7d3b-4b04-930f-e2bb5d534896 | trentmkelly/LessWrong-43k | LessWrong | Additional arguments for NIMBY
The following post was written in July of 2017 in response to a post by Julia Galef, but I decided not to post/finish it. I figured I’d piled on San Francisco enough. With the recent post on the subject by Scott Alexander, I figured I would dust it off, give it an editing pass to reflect... |
d71fcb1e-a4d3-48f7-b4e2-b991a82bd464 | trentmkelly/LessWrong-43k | LessWrong | Computational Morality (Part 1) - a Proposed Solution
Religions have had the Golden Rule for thousands of years, and while it's faulty (it gives you permission to do something to someone else that you like having done to you but they don't like having done to them), it works so well overall that it clearly must be ba... |
8c4f2f46-f849-4074-a7f0-8446a7736ec2 | trentmkelly/LessWrong-43k | LessWrong | Two Explorations
Much has been written about the fundamental opposition between explore and exploit, chaos and order, yin and yang. In this post I make two observations about the psychology of this opposition.
In the first part, I challenge the metaphor of the comfort zone: a slowly-changing region in activity-space ... |
8d6a446a-f318-47b2-96bc-23b73bb68eaf | StampyAI/alignment-research-dataset/blogs | Blogs | November 2019 Newsletter
I'm happy to announce that Nate Soares and Ben Levinstein's “[Cheating Death in Damascus](https://intelligence.org/2017/03/18/new-paper-cheating-death-in-damascus/)” has been accepted for publication in *The Journal of Philosophy* (previously voted the [second-highest-quality](https://leiterre... |
540bc932-7b22-4b1e-9d17-c3754277b0b2 | trentmkelly/LessWrong-43k | LessWrong | [Resolved] Is the SIA doomsday argument wrong?
[EDIT: I think the SIA doomsday argument works after all, and my objection to it was based on framing the problem in a misguided way. Feel free to ignore this post or skip to the resolution at the end.]
ORIGINAL POST:
Katja Grace has developed a kind of doomsday argumen... |
97cf85e5-f95b-4f4e-8342-4e6ada891df7 | trentmkelly/LessWrong-43k | LessWrong | Market-shaping approaches to accelerate COVID-19 response: a role for option-based guarantees?
This is a policy brief directed at decision makers in the UK government, with a view to accelerating production of tests, drugs and vaccines for COVID-19; but it could be adapted for a wide variety of countries, products and... |
f9fa2ca4-310d-49e8-b616-62630e4532a7 | trentmkelly/LessWrong-43k | LessWrong | Narrative Syncing
Partial attempt to make sense of: On Doing the Improbable
Epistemic status: On my inside view, this seems like a clear and useful distinction that could use a short, shared term. Hasn’t been checked much by others. Some edge cases I’m still confused about. Also I may have a suboptimal term; help ge... |
4cd589c9-6dd7-4d06-a1f7-f16eee05255d | StampyAI/alignment-research-dataset/lesswrong | LessWrong | [Preprint for commenting] Digital Immortality: Theory and Protocol for Indirect Mind Uploading
I would like to get useful input for the following text :
"Digital Immortality: Theory and Protocol for Indirect Mind Uploading"
**Abstract**. Future superintelligent AI will be able to reconstruct a model of the personali... |
0ac358fc-dfd4-48e0-bbc1-0551a83f6263 | trentmkelly/LessWrong-43k | LessWrong | Thoughts on the January CFAR workshop
So, the Center for Applied Rationality just ran another workshop, which Anna kindly invited me to. Below I've written down some thoughts on it, both to organize those thoughts and because it seems other LWers might want to read them. I'll also invite other participants to write do... |
a8e3a51c-a08b-4f14-821a-5afc06643d85 | trentmkelly/LessWrong-43k | LessWrong | Your Prioritization is Underspecified
If you were really convinced that the next task on your to do list was the very best thing to advance your goals you'd feel a lot more interested in it. Some part of you believes this to be the case and other parts very obviously don't. Thus, uncertainty about how to prioritize. M... |
9772e8fd-f136-4a9c-925e-8b465cc9b56e | trentmkelly/LessWrong-43k | LessWrong | House Party Dances
Julia and I are in CA for LessOnline this weekend, and we're staying with friends. It happened that they were hosting a party, and while this is not a group of dancing friends they asked if I'd be up for leading some dancing. One of my hosts let me borrow their violin (which I'll also have with me t... |
88de4e29-456b-4566-a622-624936322edb | trentmkelly/LessWrong-43k | LessWrong | Dissolving the zombie argument
Update: Upon reflection, I'm not entirely satisfied with this post. I think I definitely managed to identify some of the confusion around these kinds of discussion, but a smaller proportion than I would have liked.
The Zombie argument (David Chalmer's website, Stanford Encyclopedia of P... |
522f9b32-d5f8-48ab-8e5c-6504ec46fe23 | trentmkelly/LessWrong-43k | LessWrong | Simplifying Corrigibility – Subagent Corrigibility Is Not Anti-Natural
Max Harms recently published an interesting series of posts on corrigibility, which argue that corrigibility should be the sole objective we try to give to a potentially superintelligent AI. A large installment in the series is dedicated to catalog... |
e546b311-0a5a-4ad6-a391-b30876b7f66b | StampyAI/alignment-research-dataset/arxiv | Arxiv | Bayesian Optimization with Safety Constraints: Safe and Automatic Parameter Tuning in Robotics
I Introduction
---------------
Safety and the ability to operate within given constraints of the environment are critical pre-requisites for any algorithm that is applied on a real system. This holds true especially in ro... |
9b3c73f7-87f6-47a0-97b1-7cd8c87167e5 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | AI Alignment 2018-2019 Review
Rohin wrote this amazing summary of all the research happening in AI Alignment in 2018-2019. Here is the "short" version of the summary, though I do recommend reading the whole post, which I think conveys quite well what happened in the last two years in AI Alignment.
> While the fu... |
a437d573-898e-4f8b-bfcb-76b96dbcef62 | trentmkelly/LessWrong-43k | LessWrong | Value Notion - Questions to Ask
Continuing my previous post on the subject of value notion, in this post I will try to make another small step towards a big goal of value notion. Specifically, I attempt to explore what kind of major questions we can ask about the value of something. The list may not be comprehensive, ... |
87fab664-179a-4997-a509-0408b222b5ef | trentmkelly/LessWrong-43k | LessWrong | Double Crux
Author's note: There is a preexisting standalone essay on double crux (also written by Duncan Sabien) available here. The version in the handbook is similar, but has enough disoverlap that it seemed worth including it rather than merely adding the standalone post to the sequence.
Missing from this sequen... |
17f5cdc5-2e0b-477e-a0ff-99f41ca8d62c | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Meta Questions about Metaphilosophy
To quickly recap my main intellectual journey so far (omitting a lengthy [side trip](https://www.lesswrong.com/posts/m8FjhuELdg7iv6boW/work-on-security-instead-of-friendliness) into cryptography and Cypherpunk land), with the approximate age that I became interested in each topic in... |
052474bb-02b3-4e48-b610-cd532d7bbd72 | trentmkelly/LessWrong-43k | LessWrong | The Problem (TM) - Analyse a conversation
Originally published here: http://bearlamp.com.au/the-problem-tm-analyse-a-conversation/
Part 2: http://lesswrong.com/r/discussion/lw/nt8/the_problem_tm_part_2/
Part 2: http://bearlamp.com.au/the-problem-analyse-a-conversation-part-2/
I had a chat with a person who admitted t... |
033609bf-3345-495f-81cc-8368dedc8aa2 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Using Machine Learning to Guide Cognitive Modeling: A Case Study in Moral Reasoning.
1 Introduction
---------------
Explanatory and predictive power are hallmarks of any useful scientific theory. However, in practice, psychology tends to focus more on explanation (Yarkoni \BBA Westfall, [\APACyear2017](#bib.bib34))... |
ceabb8a4-1d2b-4f80-a30f-a58cd8d65518 | StampyAI/alignment-research-dataset/arbital | Arbital | Modalized modal sentence
A [modal sentence](https://arbital.com/p/) $A$ is said to be **modalized** in $p$ if every occurrence of $p$ happens within the scope of a $\square$.
As an example, $\square p \wedge q$ is modalized in $p$, but not in $q$.
If $A$ does not contain $p$, then it is trivially modalized in $p$.
... |
1290d66b-ec79-4165-9145-cb7a11da1b86 | awestover/filtering-for-misalignment | Redwood Research: Alek's Filtering Results | id: post912
Thanks to Buck Shlegeris, Ryan Greenblatt, Kshitij Sachan and Jenny Nitishinskaya for helpful discussions and feedback on drafts. This work was done at Redwood Research. The views expressed are my own and do not necessarily reflect the views of the organization. Password-locked models are trained to exhibi... |
1d0d1c40-eb9d-4954-a004-4c5a337fdb58 | trentmkelly/LessWrong-43k | LessWrong | What additional features would you like on LessWrong?
I'm not on the LessWrong team; I'm just curious, and might want to answer that question myself ^_^ |
13d71221-e72e-40f7-ae34-74dcdd8d5c6a | trentmkelly/LessWrong-43k | LessWrong | My Kind of Reflection
In "Where Recursive Justification Hits Bottom", I concluded that it's okay to use induction to reason about the probability that induction will work in the future, given that it's worked in the past; or to use Occam's Razor to conclude that the simplest explanation for why Occam's Razor works is ... |
583d6331-9f32-4952-9801-dcb578746bcd | trentmkelly/LessWrong-43k | LessWrong | Change my mind: Veganism entails trade-offs, and health is one of the axes
Introduction
To me, it is obvious that veganism introduces challenges to most people. Solving the challenges is possible for most but not all people, and often requires trade-offs that may or may not be worth it. I’ve seen effective altruist v... |
2835e83e-79ec-4357-85c0-8ce73d3adab0 | trentmkelly/LessWrong-43k | LessWrong | What's the easiest way to make a luminator?
My room is really dark. I just realized this after being outside in a car for something and then coming back to my room. (I usually don't go outside.)
I wouldn't have motivation to string a whole bunch of bulbs together, I just want something I can order and plug in. How ca... |
866352bf-f6c8-4dc9-9483-0cdeead16b3f | trentmkelly/LessWrong-43k | LessWrong | Dr. Jubjub predicts a crisis
Dr. Jubjub: Sir, I have been running some calculations and I’m worried about the way our slithy toves are heading.
Prof. Bandersnatch: Huh? Why? The toves seem fine to me. Just look at them, gyring and gimbling in the wabe over there.
Dr. Jubjub: Yes, but there is a distinct negative tre... |
30ed1b5f-6cdc-447e-a643-900d37be4800 | trentmkelly/LessWrong-43k | LessWrong | Any prior work on mutiagent dynamics for continuous distributions over agents?
Typical work on game theory, economics or multiagent negotiations assumes that there are a finite number of discrete agents. What if, instead, there's a continuous distribution over possible agents? Further, is there any work looking at the... |
ad67345f-d191-4109-a063-fba8d9a5b731 | trentmkelly/LessWrong-43k | LessWrong | Notes from Copenhagen Secular Solstice 2024
Last night, we held the Traditional Copenhagen Secular Solstice celebration. Here’s a brief overview of what went well, what could be improved and reflections on the songs.
What Went Well
Growth in Attendance
Attendance increased to 14 participants, which is exciting! Ri... |
838f8bc9-88dd-418d-871b-b90308164f3d | trentmkelly/LessWrong-43k | LessWrong | [Talk] Paul Christiano on his alignment taxonomy
Paul Christiano makes an amazing tree of subproblems in alignment. I saw this talk live and enjoyed it. |
0dde1381-7579-4019-852a-f8ab493b64b9 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Less activations can result in high corrigibility?
*This post is for people interested in model* [*interpretability*](https://www.lesswrong.com/tag/interpretability-ml-and-ai) *- one must know a fair bit on how LLMs work conceptually. Please consider everything as preliminary as this is part of an on-going project - t... |
ccf7e7b2-41be-40b5-8676-0f15cf30d294 | trentmkelly/LessWrong-43k | LessWrong | Who's afraid of impossible worlds?
In order to clarify the semantics of paraconsistent and relevance logics, we need to make a detour into impossible worlds - a fruitful detour opening up fun new vistas. Note that this is an intuitive introduction to the subject, and logic is probably the area of mathematics where it ... |
1c55911b-28c7-4273-901c-27bf53e1129a | trentmkelly/LessWrong-43k | LessWrong | Lead in Chocolate?
Consumer Reports recently tested a bunch of dark chocolate for lead and cadmium, and found that:
> For 23 of the bars, eating just an ounce a day would put an adult over a level that public health authorities and CR's experts say may be harmful for at least one of those heavy metals. Five of the ... |
d214c11d-4b69-4423-91c4-fd792c5ae57b | StampyAI/alignment-research-dataset/youtube | Youtube Transcripts | What is AI alignment?
what is a eye alignment a eye alignment
concerns the task of making sure that
artificially intelligent systems behave
in a way that is conducive to human
morality and human well-being the
problem of a eye alignment arises from
what is called the orthogonality thesis
first put forward by Oxford ph... |
0111d929-490f-4d1f-b3fe-930ffaf7a647 | trentmkelly/LessWrong-43k | LessWrong | Problem noticed in aspect of LW comunity bonding?
I have noticed that given how much I identify as a rationalist, how much I have in common with the community here, how important I consider it, etc. I have surprisingly little instant in group identification with community members compared to other online communitie... |
4787d6ca-b3ea-4406-9b38-8714070754ce | trentmkelly/LessWrong-43k | LessWrong | Link: WJS article that uses Steve Jobs' death to mock cryonics and the Singularity
And One Last Thing: Digital immortality is an app that probably wouldn't have interested Steve Jobs
Excerpt:
> Jobs made it clear that he did not welcome death, but also that life could be more interesting knowing that death would be ... |
43b2d706-956a-4324-9244-b3faa517f717 | trentmkelly/LessWrong-43k | LessWrong | Meetup Discussion
One of the most valuable services the Less Wrong community has to offer are the meetup groups. However, it strikes me that there isn't a lot of knowledge sharing between different meetup groups. Presumably there's a lot that the different groups could learn from each other -- things that can be done,... |
c266633b-6b58-4e15-9137-ef0ee88b1475 | trentmkelly/LessWrong-43k | LessWrong | O(“AGI Safety”)>O(“Stop Tyrants”)
~ AGI safety is at LEAST as hard as a protocol which prevents tyranny ~
When we want to keep ourselves safe from AGI, then one of the key criteria is “can we turn it off if it tries to go berserk?” That is the same requirement whenever we put some person in charge of an entire govern... |
aa81cf9e-f823-4483-bdf0-a135602417ee | trentmkelly/LessWrong-43k | LessWrong | Biomedical engineers analyze—and duplicate—the neural mechanism of learning in rats [link]
Restoring Memory, Repairing Damaged Brains (article @ PR Newswire)
> Using an electronic system that duplicates the neural signals associated with memory, they managed to replicate the brain function in rats associated with lon... |
58a15da4-83ba-4bd0-978a-482458ed31d5 | trentmkelly/LessWrong-43k | LessWrong | Capture today's market, capture tomorrow's game board
> Look at this lead pencil. There’s not a single person in the world who could make this pencil.
Goals for this post
* Bring Systems Theoretic Process Analysis (STPA) to the attention of the AI alignment community. This is amazing work being undergone for the pa... |
0a7d1296-20e7-4d24-bee6-8f5e7232a065 | trentmkelly/LessWrong-43k | LessWrong | Interactive exploration of LessWrong and other large collections of documents
TL;DR We made a tool that allows you to interactively explore LessWrong, EA Forum and - potentially - other large collections of documents. It seems like this part of the Information Retrieval field is highly underutilized in practice, so I ... |
bf01319e-48ba-4ecb-898f-145c471b2d10 | StampyAI/alignment-research-dataset/aisafety.info | AI Safety Info | Will AI be able to think faster than humans?
Whether current systems think faster than humans depends on the meaning of “think”. Computers already add and multiply numbers at over a billion times human speed. Problems like chess and Go, which once seemed to require human-like thought, have also become mere calculation... |
4310741f-bc76-49a3-8c30-38bd6c0d2e18 | trentmkelly/LessWrong-43k | LessWrong | Welcome to Cambridge/Boston Less Wrong
The Cambridge/Boston Less Wrong group meets on the first and third Sunday of each month, at 3:30pm, rotating between several houses. To get notifications of events, join either our Facebook group or our Meetup.com group. |
320db4ba-9ae3-4e1e-8848-6c924c5c03d7 | trentmkelly/LessWrong-43k | LessWrong | cost estimation for 2 grid energy storage systems
CO2 mitigation costs
With modern gas turbines, burning natural gas makes perhaps 0.4 kg CO2 per kWh. I generally use $75/ton as a baseline target for CO2 mitigation costs; that's around what you see from various reasonable approaches like biomass conversion. $75/ton * ... |
329e535f-23c7-4550-b5ef-831cda1bb0e5 | StampyAI/alignment-research-dataset/special_docs | Other | How social science research can inform AI governance _ Baobao Zhang _ EAGxVirtual 2020-by Centre for Effective Altruism-video_id eTkvtHymI9s-date 20200615
# Baobao Zhang How social science research can inform AI governance - EA Forum
\_Political scientist Baobao Zhang explains how social science research can help peo... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.