id
stringlengths
36
36
source
stringclasses
15 values
formatted_source
stringclasses
13 values
text
stringlengths
2
7.55M
81ba99b5-d53a-4353-a954-6fd434087e55
StampyAI/alignment-research-dataset/special_docs
Other
Artificial Intelligence: A Modern Approach Artificial Intelligence A Modern Approach Second Edition FORSYTH & PONCE GRAHAM JURAFSKY & MARTIN NEAPOLITAN RUSSELL & NORVIG Computer Vision: A Modern Approach ANSI Common Lisp Speech and Language Processing Learning Bayesian Networks Artificial Intelligence: A Modern ...
f388d820-b2e2-4606-8f60-272baf8cff06
trentmkelly/LessWrong-43k
LessWrong
Finally Entering Alignment Background Reading Death With Dignity combined with recent advancements shortening my timeline has finally made me understand on a gut level that nature is allowed to kill me. Because of this alignment has gone from "An interesting career path I might pursue after I finish studying" to "If I...
8fd8b7d2-13c6-4c55-a259-d42dcc9cc631
trentmkelly/LessWrong-43k
LessWrong
Trusting Yourself (Formally) Suppose I am a self-modifying AI reasoning about my own behavior (or about the behavior of another AI I am designing).  To a human, it seems like it is very important that we trust our own deductions. That is, humans seem to believe "things I believe are probably true." Thus a human would...
d108ca25-b13e-4a00-814d-ae6579b99e6d
trentmkelly/LessWrong-43k
LessWrong
Stupid Questions November 2015 This thread is for asking any questions that might seem obvious, tangential, silly or what-have-you. Don't be shy, everyone has holes in their knowledge, though the fewer and the smaller we can make them, the better. Please be respectful of other people's admitting ignorance and don't m...
3536fdca-75ca-4142-8c56-ea6ad7d0ba37
trentmkelly/LessWrong-43k
LessWrong
The Machine that Broke My Heart Obesity is one of the world's greatest health problems. It contributes to everything from diabetes to cancer. Obesity is hard to prevent because tracking calories requires effort. It's too easy to eat absent-mindedly if you're a middle-aged woman in charge of your household's food. Manu...
eb2baf9b-4aba-41f5-ac04-4b083391257e
StampyAI/alignment-research-dataset/lesswrong
LessWrong
The Utility of Human Atoms for the Paperclip Maximizer TL;DR: use of humans’ atoms will increase AI’s expected utility by 0.00000000000000002 of total U. The iconic example of the existential risk of superintelligence is the so-called paperclip maximizer, that is, a system which maximizes some random goal not aligne...
fcb37b4e-b493-4d6b-9e33-7f984663a5eb
trentmkelly/LessWrong-43k
LessWrong
OB Status Update Followup to:  Whither OB? Overcoming Bias currently plans to transition to a new format, including a new and more open sister site, tentatively entitled "Less Wrong".  The new site will be built out of Reddit's source code, but you won't be limited to posting links - the new site will include a WYSIW...
2c265054-e249-4afd-8968-98aeae72aedf
trentmkelly/LessWrong-43k
LessWrong
Decision-Making for Impact: A Guide The most important decision is how to decide. Photo by Ian Schneider on Unsplash For about a decade now, I’ve been geeking out hard on decision-making. To be fair, my fascination comes from an honest place: I’m a walking stereotype of a reluctant decision-maker. Back when I was a t...
ae11d53d-b3e0-4b53-9b77-4ec2a0af5c9a
trentmkelly/LessWrong-43k
LessWrong
Should we be concerned about eating too much soy? Parts of the internet say that, especially for men, eating too much soy is unhealthy while other parts of the internet advocate that soy is really great way for vegans to consume their protein. Has anyone made a deep dive into the evidence base and formed an opinion a...
59988418-bc7a-46a6-9a5d-c74ba20946a7
trentmkelly/LessWrong-43k
LessWrong
Are there any essays on what scares us? A study of fear, so-to-speak. None
45058a87-435a-4b0c-b45d-9a8658aa8a3e
trentmkelly/LessWrong-43k
LessWrong
AI security might be helpful for AI alignment I am a person new to AI alignment, and recently I am looking for a job at the AI governance field, but it seems like it is hard to find a long-term financially stable full-time job.  For me, it basically means that only a tiny amount of people consider AI alignment import...
8cb0233a-69ab-4537-94cf-621955f3665c
trentmkelly/LessWrong-43k
LessWrong
Founding a rationalist group at the University of Michigan The Vision As far as I know, there aren't any LW-style university rationalist groups. I'll be attending the University of Michigan (Ann Arbor) as a graduate student starting this fall, and plan to be there for 1.5-2 years. While I'm there, I'd like to start a ...
952c2305-7fc6-43fe-949d-a478db365864
trentmkelly/LessWrong-43k
LessWrong
A Generalization of ROC AUC for Binary Classifiers Suppose you have a binary classifier. It looks at things and tries to guess whether they’re Dogs or Not Dogs. More precisely, the classifier outputs a numeric score, which is higher for things it thinks are more likely to be Dogs. There are a bunch of ways to assess...
83a7dbe5-5235-4f62-b125-96232b88ee56
trentmkelly/LessWrong-43k
LessWrong
Meetup : Irvine Meetup Wednesday July 13 Discussion article for the meetup : Irvine Meetup Wednesday July 13 WHEN: 13 July 2011 06:00:00PM (-0700) WHERE: 4187 Campus Dr, University Center, Irvine, CA 92612 This continues the weekly meetups in Irvine. As always the meetup at the outdoor food court in the University ...
d01d7120-0c20-4736-8837-b1cb86f9d01b
trentmkelly/LessWrong-43k
LessWrong
Can you recognize a random generator? I can't seem to get my head around a simple issue of judging probability. Perhaps someone here can point to an obvious flaw in my thinking. Let's say we have a binary generator, a machine that outputs a required sequence of ones and zeros according to some internally encapsulated...
899c747a-7097-4610-97ba-dd7258b94270
trentmkelly/LessWrong-43k
LessWrong
Meetup : San Francisco Meetup: Fermi Estimates Discussion article for the meetup : San Francisco Meetup: Fermi Estimates WHEN: 12 October 2015 06:15:00PM (-0700) WHERE: 1597 Howard St. San Francisco, CA NOTE: THE LOCATION HAS CHANGED RECENTLY We'll be meeting to do some fermi estimates. You can call 301-458-0764 t...
d6145074-7e93-4bbb-aa67-8a6dcff5a470
StampyAI/alignment-research-dataset/distill
Distill Scientific Journal
Research Debt .mountain-illustration { margin-top: 20px; margin-bottom: 10px; width: 180%; margin-left: -40%; } @media (min-width: 768px) { .mountain-illustration { max-width: 1200px; margin-left: auto; width: 100%; } } dt-article > h1:first-of-type { margin-top: 0; } Achieving a research-level unders...
3af66838-835b-40c6-8326-659e7e827be3
StampyAI/alignment-research-dataset/blogs
Blogs
Computing hardware performance data collections This is a list of public datasets that we know of containing either measured or theoretical performance numbers for computer processors. List ---- 1. **[Top 500](https://www.top500.org/lists/2017/06/)** maintains a list of the top 500 supercomputers, updated every si...
53e3c93a-b6cc-47d5-b6d6-f919b56d433f
trentmkelly/LessWrong-43k
LessWrong
A Narrative History of Environmentalism's Partisanship This is the second in a sequence of four posts taken from my recent report: Why Did Environmentalism Become Partisan? Many of the specific claims made here are investigated in the full report. If you want to know more about how fossil fuel companies’ campaign con...
b9f040ee-6638-41f4-a829-40ce6949562c
trentmkelly/LessWrong-43k
LessWrong
Introducing the WeirdML Benchmark WeirdML website Related posts: How good are LLMs at doing ML on an unknown dataset? o1-preview is pretty good at doing ML on an unknown dataset Introduction How good are Large Language Models (LLMs) at doing machine learning on novel datasets? The WeirdML benchmark presents LLMs ...
fc85bca0-01ea-4f36-9c2d-4f7e2559279c
StampyAI/alignment-research-dataset/lesswrong
LessWrong
A decade of lurking, a month of posting This post is a look back on my first month or so as an active contributor on LessWrong, after lurking for over a decade. My experience so far has been overwhelmingly positive, and one purpose of this post is to encourage other lurkers to do the same. The reason I decided to sta...
db7ddac0-971d-4e8d-87c4-c267a6c3a45b
trentmkelly/LessWrong-43k
LessWrong
Chapter 86: Multiple Hypothesis Testing (International news headlines of April 7th, 1992:) Toronto Magical Tribune: ENTIRE BRITISH WIZENGAMOT REPORTS SEEING 'BOY-WHO-LIVED' FRIGHTEN A DEMENTOR EXPERT ON MAGICAL CREATURES: "NOW YOU'RE JUST LYING" FRANCE, GERMANY ACCUSE BRITAIN OF MAKING THE WHOLE THING UP New Zeal...
e3e9a183-43bc-4fb3-b02c-a83889a1deb7
trentmkelly/LessWrong-43k
LessWrong
Reconceptualizing the Nothingness and Existence Epistemic status: very unsure, just got a possibly interesting philosophical concept to share from my thoughts. Confidence: seems possible but hard to evaluate Might not be a very original thought but never heard of it. Traditional views see nothingness as the simplest ...
57a6b42b-c546-4985-97c9-aa43492973c5
trentmkelly/LessWrong-43k
LessWrong
Rationality Quotes April 2013 Another monthly installment of the rationality quotes thread. The usual rules apply: * Please post all quotes separately, so that they can be upvoted or downvoted separately. (If they are strongly related, reply to your own comments. If strongly ordered, then go ahead and post them toge...
d056b7df-a929-4d89-bd72-e5160a595350
trentmkelly/LessWrong-43k
LessWrong
The Kernel of Meaning in Property Rights A phenomenon I have often encountered when thinking about things is when everything seems to collapse to tautology. This is hard to define precisely, but I’ll give you some examples: * Bounded rationality: Bounded rationality can be thought of as “rationality conditional on s...
69284b49-727e-4081-a47c-031a578dd8d3
trentmkelly/LessWrong-43k
LessWrong
How I do research Someone asked me about this, so here are my quick thoughts. Although I've learned a lot of math over the last year and a half, it still isn't my comparative advantage. What I do instead is, Find a problem that seems plausibly important to AI safety (low impact), or a phenomenon that's secretly con...
56404f14-0567-4084-9612-41019169ca81
trentmkelly/LessWrong-43k
LessWrong
The LessWrong 2018 Book is Available for Pre-order For the first time, you can now buy the best new ideas on LessWrong in a physical book set, titled: > A Map that Reflects the Territory: Essays by the LessWrong Community It is available for pre-order here. The standard advice for creating things is "show, don't te...
e4923e8e-69b0-4d3f-8ded-8c9389186bee
trentmkelly/LessWrong-43k
LessWrong
Center for Applied Postrationality: An Update Previous post: https://www.lesswrong.com/posts/7X3BWzAt62yvvtyQX/announcing-the-center-for-applied-postrationality Content Notice: Mild to Moderate NSFW language, Defective Altruism, Infoblessings A lot of progress has been made here at CFAP in the past two years. After...
92fc8a50-40fc-4ecb-b700-d7e4f931af29
trentmkelly/LessWrong-43k
LessWrong
Why take notes: what I get from notetaking and my desiderata for notetaking systems
ce85fb5b-6315-4cab-af1a-f4212b4332fd
StampyAI/alignment-research-dataset/blogs
Blogs
uploading people for alignment purposes uploading people for alignment purposes --------------------------------------- as per [my utopian vision](%E2%88%80V.html), i've thought that an aligned AI would want to figure out how to upload us. but, thinking about it more, it could be the other way around: if we can up...
81d8f08b-62a9-49d4-98d9-8432c14cb84f
trentmkelly/LessWrong-43k
LessWrong
Implications for the likelihood of human extinction from the recent discovery of possible microbial life You might have seen the recent news that there are signatures that suggest the planet K2 18-b might be the home of microbial life. The Fermi Paradox asks: if space is so big, why haven't the aliens contacted us ye...
d76937eb-198a-4e9e-8465-50165dfbb53f
trentmkelly/LessWrong-43k
LessWrong
Toni Kurz and the Insanity of Climbing Mountains This post has been recorded as part of the LessWrong Curated Podcast, and can be listened to on Spotify, Apple Podcasts, and Libsyn. ---------------------------------------- Content warning: death I've been on a YouTube binge lately. My current favorite genre is disa...
1d2c45c3-da27-4e12-8115-b1e832624870
trentmkelly/LessWrong-43k
LessWrong
Digging the Bull's Horn Some time ago I learned of the metaphor of 'digging the bull's horn'. This might sound a little strange, since horns are mostly hollow, but imagine a bull's horn used to store black powder. In the beginning the work is easy and you can scoop out a lot powder with very little effort. As you dig ...
097d5121-e77a-49a4-995e-88c0efba9bad
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
An intervention to shape policy dialogue, communication, and AI research norms for AI safety A PDF version of this article can be found [here](https://drive.google.com/open?id=0B_diIvF4w7F_TG9DM3R1ckYxdDg).1 Abstract ======== Discourse on AI safety suffers from heated disagreement between those sceptical and those...
cd604203-1fb5-43dc-9473-a7bf4feb08ef
trentmkelly/LessWrong-43k
LessWrong
Comparing Alignment to other AGI interventions: Extensions and analysis In the last post I presented the basic, bare-bones model, used to assess the Expected Value of different interventions, and especially those related to Cooperative AI (as distinct from value Alignment). Here I briefly discuss important enhancement...
72d2a0fd-c940-44bb-9cad-4390346e1db6
trentmkelly/LessWrong-43k
LessWrong
Social Dynamics These are some summary notes on social dynamics. * conformity * try to fit in * pandering * pleasing people * avoiding conflict * do whatever the group thinks is high status * follow trends * you need to already have friends. people are impressed by people who other people alre...
aba84e61-3b8d-4603-97c6-b4f373d28784
trentmkelly/LessWrong-43k
LessWrong
Organ donation vs Cryonics (I wrote this as a comment, but it struck me as something that was potentially worth sharing with a wider audience. It seems overly specific for a main post, however :)) > 1) I think I can save more lives by being an organ donor [ed: rather than doing cryonics] There were about 6 thousand ...
a225ef05-d3dc-4018-8a10-d3e747350f09
trentmkelly/LessWrong-43k
LessWrong
A Parable Of Explainability > Ok, sure, machine learning is great for making predictions; But you can’t use it to replace scientific theory. Not only will it fail to reach generalizable conclusions, but the result is going to lack elegance and explainability. We won’t be able to understand it or build upon it! What m...
8f85f490-fa13-4319-bb0b-eff5f3c53648
trentmkelly/LessWrong-43k
LessWrong
Temporary Cognitive Hyperparameter Alteration Social anxiety is one hell of a thing. I used to struggle with it a lot — escaping pressure by fleeing to the toilet. I’ve reduced my levels of social anxiety by bashing it over the head with exposure therapy, repeatedly dealing with anxiety-provoking situations until the...
350a1e90-851f-4e62-936a-092f07791dfa
trentmkelly/LessWrong-43k
LessWrong
Fundamental Uncertainty: Prelude NB: The first draft of the book is now complete. This post is no longer very relevant, but I've left it up for posterity. You can find the most up-to-date info on the book's website. I have a plan to write a sequence that I can turn into a book about the problem of the criterion. I'm ...
8c3d7ca1-c4c7-4fc0-853b-1b9e599b398c
trentmkelly/LessWrong-43k
LessWrong
A Candidate Complexity Measure I'd like to introduce a mathematical definition that tries to capture some of our intuitive notion of 'complexity'. People use this word in lots of ways. What I mean by this word is the sort of 'complexity' that seems to have been increasing over the history of our planet for the past fe...
75275b32-e92d-43be-b653-6d07d2135da5
trentmkelly/LessWrong-43k
LessWrong
Weekly LW Meetups This summary was posted to LW Main on December 2nd. The following week's summary is here. The following meetups take place in cities with regular scheduling, but involve a change in time or location, special meeting content, or simply a helpful reminder about the meetup: * Bay Area Winter Solstice...
5b767d16-7eff-4b2a-8d45-766a7df3c47f
trentmkelly/LessWrong-43k
LessWrong
The Valley of Bad Theory An interesting experiment: researchers set up a wheel on a ramp with adjustable weights. Participants in the experiment then adjust the weights to try and make the wheel roll down the ramp as quickly as possible. The participants go one after the other, each with a limited number of attempts, ...
b99dab81-7986-4680-9fdb-15e82fb366eb
StampyAI/alignment-research-dataset/blogs
Blogs
Learning By Writing *Click lower right to download or find on Apple Podcasts, Spotify, Stitcher, etc.* I have very detailed opinions on lots of topics. I sometimes get asked how I do this, which might just be people making fun of me, but I choose to interpret it as a real question, and I’m going to sketch an answer...
1ed1f50f-643e-4bdd-bdee-8b3caaac2ec2
trentmkelly/LessWrong-43k
LessWrong
Enjoy ≠ want, but why should wants submit? There are things that we try to get. There are things that we enjoy when we have them. There’s overlap,  but they aren’t the same. When people notice they aren’t the same, they often try to change or override their wants – or more often others’ wants – in the direction of ref...
c11f8596-a4df-4ccf-9004-1216bc928992
StampyAI/alignment-research-dataset/special_docs
Other
The Liability Problem for Autonomous Artificial Agents The Liability Problem for Autonomous Artificial Agents Asaro, Peter M. School of Media Studies, The New School Center for Information Technology Policy, Princeton University Center for Internet and Society, Stanford Law School asaro@newschool.edu Abstract This pap...
9729ad87-f514-4f21-b555-7427d87f982e
trentmkelly/LessWrong-43k
LessWrong
How to measure (human) impact? I would like to more thoroughly investigate the impact I have on the world. This includes determining roughly how my interactions with other people affect their well-being; how much energy is required to sustain my habits (this would be the energy expenditure involved in the transportati...
cd46126f-2a25-446d-a84b-0dbe07d41d97
trentmkelly/LessWrong-43k
LessWrong
The underlying model of a morphism I've already talked about generalised models. The aim is not only to have a universal system for modelling any agent's mental model - universality is pretty easy to get - but a system where it's easy to recreate these mental models. And then analyse the transition between models. Th...
495a068a-9561-4447-92b5-ebc9de20e348
trentmkelly/LessWrong-43k
LessWrong
Productivity thoughts from Matt Fallshaw At the 2014 Effective Altruism Summit in Berkeley a few weeks ago, I had the pleasure of talking to Matt Fallshaw about the things he does to be more effective.  Matt is a founder of Trike Apps (the consultancy that built Less Wrong), a founder of Bellroy, and a polyphasic slee...
4e1ddb70-6cf5-4751-ba3d-c223268ee623
trentmkelly/LessWrong-43k
LessWrong
Rational philosophies Hello all, I'm working on a top-level post about how Stoicism is an instrumentally useful philosophy to adopt, and figured I should give other philosophies a fair shake as well. Does anyone know of any other philosophies out there that seem to be practically useful or otherwise provide strategie...
45b9489d-d631-445c-8ca8-f33d97eba671
trentmkelly/LessWrong-43k
LessWrong
What Ketamine Therapy Is Like Sniff-click. Sniff-click. One shot in each nostril, and here…we…go. A while ago, I started Ketamine therapy for depression. I didn’t finish - I changed jobs, which switched insurances, which messed everything up - but I was doing it for two and half weeks (five doses) before then, and h...
39fbbeaf-f4c3-4c45-ba87-b713279491b6
trentmkelly/LessWrong-43k
LessWrong
Finding a quote: "proof by contradiction is the closest math comes to irony" I've got a paraphrased quote floating around in my mind, and I'm trying to track down the source. I think it was an article online but I have no idea where. There was a sentence like "proof by contradiction is the closest math comes to iron...
2ff9a56c-177b-487f-990d-f2fbf75ce9a8
trentmkelly/LessWrong-43k
LessWrong
Study Hacks on Convenience as Anti-Productivity Distraction is a Symptom of a Deeper Problem: The Convenience Principle and the Destruction of American Productivity is a good article on distractions versus getting things done. With extra emphasis on how many of our distractions are the result of a desire for convenien...
3f332b9b-c4be-4712-8d1b-2536ac4b805c
trentmkelly/LessWrong-43k
LessWrong
Betrand Russell's Ten Commandments Betrand Russell's Ten Commandments for teachers. > 1. Do not feel absolutely certain of anything. > 2. Do not think it worth while to proceed by concealing evidence, for the evidence is sure to come to light. > 3. Never try to discourage thinking for you are sure to succeed. >...
8d338b7f-9aba-4bca-b754-ae58201faf9c
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Reducing sycophancy and improving honesty via activation steering *Produced as part of the* [*SERI ML Alignment Theory Scholars Program*](https://serimats.org/) *- Summer 2023 Cohort, under the mentorship of Evan Hubinger.* I generate an activation steering vector using Anthropic's [sycophancy dataset](https://huggin...
c060fc70-55f5-4dd6-a57b-cb3379171e0a
StampyAI/alignment-research-dataset/special_docs
Other
Ensuring safety and consistency in the age of machine learning _ Chongli Qin _ EAGxVirtual 2020-by Centre for Effective Altruism-video_id SS9DMr4VkbY-date 20200615 # Chongli Qin Ensuring safety and consistency in the age of machine learning - EA Forum \_Machine learning algorithms have become an essential part of tec...
086e4f9b-4e2d-4655-b7d0-05ee6b029786
trentmkelly/LessWrong-43k
LessWrong
Youtube channel devoted to the art of rationality For quite some time I've considered creating a Youtube channel devoted to rationality. I live in Italy, and the political situation here bears some similarities with the US': we've seen the rise of populist figures accompanied by the debate around fake news and post-t...
560205ea-65f8-4401-80b8-407ffcf38b89
trentmkelly/LessWrong-43k
LessWrong
Trustworthy answers via indifference A putative new idea for AI control; index here. It would be useful to have a mechanism for getting an AI to provide an honest answer to a specific question. This post presents such a mechanism. The question is: what is the impact of X/¬X on the expected utility of a utility funct...
f05a858e-d5e1-45dd-9949-de01602cdaba
StampyAI/alignment-research-dataset/special_docs
Other
Racing to the precipice: a model of artificial intelligence development  FHI TECHNICAL REPORT  Racing to the Precipice : a Model of Artificial Intelligence Development Stuart Armstrong Nick Bostrom Carl Shulman Technical Report #20 13-1 Cite as : Armstrong, S. & Bostrom, N. & Shulman, C. (201...
8d62e102-bfdc-448f-9b23-8c222135a4cb
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Futarchy Fix .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-space: nowrap; float: none; direction: ltr; ma...
c86a044a-cf24-4754-b4f0-937c6c5049a8
trentmkelly/LessWrong-43k
LessWrong
Meetup : Conflicted Brains & Competing Goals: The Non-Unitary Self Discussion article for the meetup : Conflicted Brains & Competing Goals: The Non-Unitary Self WHEN: 05 November 2016 03:30:00PM (+1100) WHERE: Ross House, 247 Flinders Lane, Melbourne Our next meetup for Less Wrong Melbourne will be on Saturday 5th ...
8693ba45-c0b7-4063-8330-5d4048e76634
StampyAI/alignment-research-dataset/lesswrong
LessWrong
[Hebbian Natural Abstractions] Introduction With this sequence, we (Sam + Jan) want to provide a principled derivation of the [natural abstractions hypothesis](https://www.alignmentforum.org/posts/Nwgdq6kHke5LY692J/alignment-by-default#Unsupervised__Natural_Abstractions) (which we will introduce in-depth in later post...
54191969-4578-4dd2-902c-61fecc6e846f
trentmkelly/LessWrong-43k
LessWrong
UFO Betting: Put Up or Shut Up Are you confident in your current ontology? Are you convinced that ultimately all ufos are prosaic in nature? If so, do you want some immediate free money? I suspect that LW's are overconfident in their views on ufos/uap. As such, I'm willing to offer what I think many will find to be ...
c0fa6fbd-026b-4464-b719-5e77eb0a7ddd
trentmkelly/LessWrong-43k
LessWrong
March 25: Daily Coronavirus Updates Aggregators Clusters C19 literature for easier filtering and digestion Given a target article, identifies similar articles Economics Balancing the cost in lives of C19 and shutdown Plea to consider both the cost in lives of doing nothing and of gutting the economy, and spend mor...
43b81ee2-80e5-449f-839b-81c96bea1290
trentmkelly/LessWrong-43k
LessWrong
Spoiler-Free Review: The Stanley Parable At Ben Pace’s recommendation, I recently played The Stanley Parable. Ben considers the game Tier 1.5. On reflection I consider this Tier 2. If you can spare the $15 for a short game, it’s Worth It to play. If there’s one game I’ve played that needs to be played blind, this is ...
7114a6ef-fba6-49f5-bd82-58c170ffdf7f
trentmkelly/LessWrong-43k
LessWrong
Meetup : Meetup : London - Paranoid Debating 2nd Feb, plus social 9th Feb Discussion article for the meetup : Meetup : London - Paranoid Debating 2nd Feb, plus social 9th Feb WHEN: 02 February 2014 02:00:00PM (+0000) WHERE: Shakespeare's Head, Holborn, WC2B 6BG The next London meetup will be a session of paranoid d...
3b29d192-3644-4f57-a40b-0709f6133f2c
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Will Machines Ever Rule the World? MLAISU W50 Watch this week's episode on [YouTube](https://youtu.be/saiqSZkDBSk) or listen to the audio version [here](https://share.transistor.fm/s/44069318). Hopes and fears of the current AI safety paradigm, GPU performance predictions and popular literature on why machines will n...
c2588f9d-3ae7-4ce2-bc25-15fa3b35fb7b
trentmkelly/LessWrong-43k
LessWrong
Are "non-computable functions" always hard to solve in practice? I am trying to get a better understanding of the philosophical implications of non-computable functions, and this is some fix idea I had to which I don't know the answer: For a Problem Π: Π undecidable => Π infeasible to calculate even when making the do...
c19630b1-ad4e-4d06-b2da-5482a1e5087b
trentmkelly/LessWrong-43k
LessWrong
Rationality Quotes January 2014 Rules: * Please post all quotes separately, so that they can be upvoted or downvoted separately. (If they are strongly related, reply to your own comments. If strongly ordered, then go ahead and post them together.) * Do not quote yourself. * Do not quote from Less Wrong itself, HPM...
36826b7c-99ef-4f11-beae-54ee955c769d
StampyAI/alignment-research-dataset/blogs
Blogs
July 2022 Newsletter MIRI has put out three major new posts: [**AGI Ruin: A List of Lethalities.**](https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities) Eliezer Yudkowsky lists reasons AGI appears likely to cause an existential catastrophe, and reasons why he thinks the current research co...
0a0d7922-e788-4d62-9853-528140cba005
StampyAI/alignment-research-dataset/lesswrong
LessWrong
What is OpenAI's plan for making AI Safer? Some claim that large AI companies like OpenAI aren’t doing enough to make AI safer, or are causing harm by speeding up AI progress. But what *are* OpenAI doing to make AI safer? In general, OpenAI is very optimistic about using AI to help with alignment research. Their bas...
c629678f-4e6f-4a79-bb5b-b1ce186d6b03
trentmkelly/LessWrong-43k
LessWrong
Virtue Ethics for Consequentialists Meta: Influenced by a cool blog post by Kaj, which was influenced by a cool Michael Vassar (like pretty much everything else; the man sure has a lot of ideas). The name of this post is intended to be taken slightly more literally than the similarly titled Deontology for Consequentia...
51cd37ff-5d70-455e-9617-1f69e7419a86
trentmkelly/LessWrong-43k
LessWrong
200 COP in MI: Techniques, Tooling and Automation This is the seventh post in a sequence called 200 Concrete Open Problems in Mechanistic Interpretability. Start here, then read in any order. If you want to learn the basics before you think about open problems, check out my post on getting started. Look up jargon in m...
b926fe0d-b8be-48b3-a6f4-dd9ab6bde099
trentmkelly/LessWrong-43k
LessWrong
Where do your eyes go? [Shoutout to the LW team for very helpful (and free!) feedback on this post.] I. Prelude When my wife first started playing the wonderful action roguelike Hades, she got stuck in Asphodel. Most Hades levels involve dodging arrows, lobbed bombs, melee monsters, and spike traps all whilst hackin...
ae7dd9d2-c568-4821-a9a7-51e202d0983f
trentmkelly/LessWrong-43k
LessWrong
Meetup: Bay Area: Jan 15th, 7pm Overcoming Bias / Less Wrong meetup in the San Francisco Bay Area at SIAI House on January 15th, 2010, starting at 7PM. Robin Hanson and possibly Michael Vassar will be present.
3a52b01d-35c0-42d6-a813-1a2f888a7e1a
trentmkelly/LessWrong-43k
LessWrong
Problems with p-values I'm taking a really cool course right now on statistics education, which covers: * replication * p-values being bad * bayesian statistics * meta-analyses A lot of our readings are really fun, and I'll likely write more posts based on them, maybe into a "why social science studies are ba...
c78862dd-e2d9-4d02-b4aa-c268f9f1a841
trentmkelly/LessWrong-43k
LessWrong
Meetup : Fort Collins Meetup Discussion article for the meetup : Fort Collins Meetup WHEN: 18 January 2012 07:00:00PM (-0700) WHERE: 144 North College Avenue, Fort Collins, CO 80524 We continue to gain in numbers and in friendship. Would you be interested in meeting accomplished, interesting people to talk about Le...
f2caa95b-2706-4669-9049-42c4da46bf87
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Will AI end everything? A guide to guessing | EAG Bay Area 23 Below is the video and transcript for my talk from EA Global, Bay Area 2023. It's about how likely AI is to cause human extinction or the like, but mostly a guide to how I think about the question and what goes into my probability estimate (though I do get ...
2b1be759-1a62-4372-a030-a4948c00f13e
trentmkelly/LessWrong-43k
LessWrong
Responding to 'Beyond Hyperanthropomorphism' TL;DR:  This is an attempt to clarify a post (or perhaps just my own confusion) from a tech commentator, that was skeptical about AGI as a possibility, and alignment research as a project.  As such, there is nothing novel in here, but it might be useful for people from outs...
7e0f72ac-db99-42ed-94d1-1388e9742e97
trentmkelly/LessWrong-43k
LessWrong
Catastrophic Regressional Goodhart: Appendix This is a more technical followup to the last post, putting precise bounds on when regressional Goodhart leads to failure or not. We'll first show conditions under which optimization for a proxy fails, and then some conditions under which it succeeds. (The second proof will...
271d6fe0-435f-4da1-8fed-61e3da99580d
trentmkelly/LessWrong-43k
LessWrong
Corrigibility or DWIM is an attractive primary goal for AGI While rereading the List of Lethalities (LoL), I was compelled by the argument against corrigibility. It's really hard to make a goal of "maximize X, except if someone tells you to shut down". I think the same argument applies to Christiano's goal of achievin...
c8376e14-e1de-4ff1-9e07-0e0645289941
trentmkelly/LessWrong-43k
LessWrong
I'm offering free math consultations! You can schedule them with me at this link: https://calendly.com/gurkenglas/consultation We can discuss whatever you're working on, such as math or code, but usually people end up having me watch their coding and giving them tips. Here's how this went last time: To my memory, a...
66c58a99-40cf-4c4d-8c37-01c9b2b9c86a
trentmkelly/LessWrong-43k
LessWrong
Meditations on Mexico City As I begin writing this, I am on a flight back from Mexico City to New York (I finished the next morning back in the city). It has been a good one-week trip. Most of the time outside of the hotel was spent walking around the city, until either me or my wife Laura could walk no further, or at...
9ae4ab7e-3e17-45c0-a6c6-e7f49485dd7a
trentmkelly/LessWrong-43k
LessWrong
Petition: Off topic area   Petition: LW should introduce a dedicated off topic area   Why?   1) I want to discuss various topics with people who are both intelligent and rationalist, and i know of no other place where to do it.   2) If find that rationality is getting boring in itself. I need to use it on some...
8073aac6-0e67-4ab9-b6af-147a46aa9db8
trentmkelly/LessWrong-43k
LessWrong
Mechanistic anomaly detection and ELK (Follow-up to Eliciting Latent Knowledge. Describing joint work with Mark Xu. This is an informal description of ARC’s current research approach; not a polished product intended to be understandable to many people.) Suppose that I have a diamond in a vault, a collection of camera...
317151cc-2550-49d2-a133-be83bcb3b03c
StampyAI/alignment-research-dataset/aisafety.info
AI Safety Info
What is reinforcement learning (RL)? [Reinforcement learning (RL)](https://en.wikipedia.org/wiki/Reinforcement_learning) trains an AI by rewarding behavior we want and punishing behavior we don't want. The AI will repeat behaviors that have helped it get higher rewards in the past, and avoid behaviors that gave nega...
8b8937cf-1e54-4c41-8b68-2c70ea7d1563
trentmkelly/LessWrong-43k
LessWrong
Improvement on MIRI's Corrigibility This post was written as a submission for the AI Alignment Award, initiated at EffiSciences' event.   This post aims to address the problem of corrigibility as identified by MIRI in 2015. We propose an extended formalism that allows us to write the desiderata of a corrigible behav...
83fe0a85-4a3f-443d-85be-03db2fac2a16
StampyAI/alignment-research-dataset/lesswrong
LessWrong
[Link/crosspost] [US] NTIA: AI Accountability Policy Request for Comment > The Department of Commerce’s National Telecommunications and Information Administration (NTIA)[[1]](https://forum.effectivealtruism.org/posts/GcrKndFY2oSKEFLub/us-ntia-ai-accountability-policy-request-for-comment#fnqz6fbua7br) has launched an i...
a32bcd31-144d-4f6b-a933-91c9181dd2ff
trentmkelly/LessWrong-43k
LessWrong
How long should political (and other) terms be? (Epistemic Status: Superficial application of complex topic to complex domain) In a representative democracy, there is always a tension between the need for elites to make unpopular decisions and the need for elites to be held responsible for mistakes by the public. In ...
bef8290c-44ed-4d91-aeee-980383b3e46f
StampyAI/alignment-research-dataset/arxiv
Arxiv
Understanding Learned Reward Functions. 1 Introduction --------------- ### 1.1 Reward Learning In the last several years, reinforcement learning techniques have produced agents surpassing human performance in tasks as varied as Atari games mnih2015human, DOTA OpenAI\_dota, Go silver2016mastering and Starcraft a...
c0dc3900-d933-4288-a75d-4c99ca6628ca
StampyAI/alignment-research-dataset/lesswrong
LessWrong
What’s going on? LLMs and IS-A sentences *This is a* [*cross-post from New Savanna*](https://new-savanna.blogspot.com/2023/11/whats-going-on-llms-and-is-sentences.html)*.* For the moment I have decided that Waddington’s classic diagram of the epigenetic landscape is a useful way of thinking about when happens when an...
d7c46446-f516-4353-9edd-631dc0e61ac9
trentmkelly/LessWrong-43k
LessWrong
Five neglected work areas that could reduce AI risk Tldr: We identify five areas of work that should be further investigated: 1. Helping information aggregators scale during advanced AI development, 2. Improving internal AI deployment in policy organizations, 3. Researching the institutional design for evaluating ...
0554d912-c2c9-4283-86cc-1abc4cc89e2d
StampyAI/alignment-research-dataset/blogs
Blogs
Import AI 325: Automated mad science; AI vs democracy; and a 12B parameter language model Welcome to Import AI, a newsletter about AI research. Import AI runs on lattes, ramen, and feedback from readers. If you’d like to support this (and comment on posts!) please subscribe. [Subscribe now](https://importai.substack....
8bcb5a1a-aca8-4ded-8de6-8e105dbcccbb
trentmkelly/LessWrong-43k
LessWrong
[SEQ RERUN] What I Think, If Not Why Today's post, What I Think, If Not Why was originally published on 11 December 2008. A summary (taken from the LW wiki):   > Yudkowsky's attempt to summarize what he thinks on the subject of Friendly AI, without providing any of the justifications for what he believes. Discuss ...
177d7eda-0f28-44d4-97b5-4e0c0eb1a293
StampyAI/alignment-research-dataset/distill
Distill Scientific Journal
Visualizing the Impact of Feature Attribution Baselines Path attribution methods are a gradient-based way of explaining deep models. These methods require choosing a hyperparameter known as the *baseline input*. What does this hyperparameter mean, and how important is it? In this article, we investigate these ques...
441ae802-3bd9-4653-aa1e-81ca6074e1df
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Levels of AI Self-Improvement [new draft for commenting, full formatted version is here: https://goo.gl/c5UfdX] **Abstract**: This article presents a model of self-improving AI in which improvement could happen on several levels: hardware, learning, changes in code, in goals system, creating virtual organization, a...
8371882e-f94f-4784-952c-8bba08774416
trentmkelly/LessWrong-43k
LessWrong
Open thread, Nov. 14 - Nov. 20, 2016 If it's worth saying, but not worth its own post, then it goes here. ---------------------------------------- Notes for future OT posters: 1. Please add the 'open_thread' tag. 2. Check if there is an active Open Thread before posting a new one. (Immediately before; refresh the ...
33be6e9c-6554-48ea-ae59-e35ef31368c6
StampyAI/alignment-research-dataset/arxiv
Arxiv
Predicting Out-of-Distribution Error with the Projection Norm 1 Introduction --------------- | | | | --- | --- | | . ( | . ( | Figure 1: How to compute Projection Norm on unlabeled OOD data. (Left) Given a classifier ˆθ, we first pseudo-label the OOD data ˜x1:m using the predictions of ˆθ. Next, we obtain a...
75b2ad26-ea71-42f9-ae1e-0b364ac3a9f0
trentmkelly/LessWrong-43k
LessWrong
I discovered LessWrong... during Good Heart Week I decided to write this because I thought my experience could be an interesting or useful data point, in particular for understanding the results of Good Heart Week: Extending the Experiment. ---------------------------------------- I happened to discover LessWrong ju...
4d6dc844-7f7b-4de8-bf53-cc45b3411c8c
trentmkelly/LessWrong-43k
LessWrong
The Futility of Emergence The failures of phlogiston and vitalism are historical hindsight. Dare I step out on a limb, and name some current theory which I deem analogously flawed? I name emergence or emergent phenomena—usually defined as the study of systems whose high-level behaviors arise or “emerge” from the inte...