text stringlengths 300 320k | source stringlengths 52 154 |
|---|---|
# What are information hazards?
_This post was written for [Convergence Analysis](https://www.convergenceanalysis.org/). Cross-posted to the [EA Forum](https://forum.effectivealtruism.org/posts/Nc5EjccDTfmcrG93j/what-are-information-hazards)._
The concept of _information hazards_ is highly relevant to many efforts to... | https://www.lesswrong.com/posts/R7szBR5H487XutfKy/what-are-information-hazards |
# Is there software for goal factoring?
Is there software for goal factoring (the CFAR technique)? I want to use it to create a directed graph, which is not necessarily a tree and may have cycles (this requirement disqualifies mind mapping software). Nodes are goals, they have attached text. There's an edge from goal ... | https://www.lesswrong.com/posts/2qukdBmdqL86WrvLn/is-there-software-for-goal-factoring |
# We Want MoR (HPMOR Discussion Podcast) Completes Book One
The _We Want MoR_ podcast, a chapter by chapter read-through and discussion of _Harry Potter and the Methods of Rationality_ by a Rationalist and a newbie, has just finished up [Book One](https://www.doofmedia.com/2020/02/18/we-want-mor-book-one-retrospective... | https://www.lesswrong.com/posts/qs5YLdMBMFijWSP8p/we-want-mor-hpmor-discussion-podcast-completes-book-one |
# Is there an intuitive way to explain how much better superforecasters are than regular forecasters?
Is there an intuitive way to explain how much better superforecasters are than regular forecasters? (I can look at the tables in [https://www.researchgate.net/publication/277087515\_Identifying\_and\_Cultivating\_Supe... | https://www.lesswrong.com/posts/yDdBLdTSAEcbjkbGj/is-there-an-intuitive-way-to-explain-how-much-better |
# Stuck Exploration
Recently, we've been considering agents that instead of just being given the problem setup have to actually learn it from experience. For example, instead of being told that an agent is a perfect predictor, they just have to play until they realise this. When an agent encounters a problem like iter... | https://www.lesswrong.com/posts/ajvvtKuNzh7aHmooT/stuck-exploration |
# Training Regime Day 5: TAPs
# Introduction
TAP stands for Trigger Action Pattern/Planning. "Pattern" is a descriptive term, i.e. it's describing what's actually happening. "Planning" is what I hope to teach you how to do, i.e. how to create certain Trigger Action Patterns in yourself.
In the literature, these are... | https://www.lesswrong.com/posts/ywWH2zRL39v3KEvHM/training-regime-day-5-taps |
# [AN #87]: What might happen as deep learning scales even further?
Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through this [spreadsheet](https://docs.google.com/spreadsheets/d/1PwWbWZ6FPqAgZWOoOcXM8N_t... | https://www.lesswrong.com/posts/69XPfonos795hD57o/an-87-what-might-happen-as-deep-learning-scales-even-further |
# On unfixably unsafe AGI architectures
There's loads of discussion on ways that things can go wrong as we enter the post-AGI world. I think an especially important one for guiding current research is:
*Maybe we'll know how to build **unfixably** unsafe AGI, but can't coordinate not to do so.*
As a special case, I w... | https://www.lesswrong.com/posts/qvyv72fCiC46sxfPt/on-unfixably-unsafe-agi-architectures |
# Tessellating Hills: a toy model for demons in imperfect search
If you haven't already, take a look at this post by johnswentworth to understand what this is all about: [https://www.lesswrong.com/posts/KnPN7ett8RszE79PH/demons-in-imperfect-search](https://www.lesswrong.com/posts/KnPN7ett8RszE79PH/demons-in-imperfect-... | https://www.lesswrong.com/posts/X7S3u5E4KktLp7gHz/tessellating-hills-a-toy-model-for-demons-in-imperfect |
# Mapping downside risks and information hazards
_This post was written for [Convergence Analysis](https://www.convergenceanalysis.org/)._
Many altruistic actions have downside risks; they might turn out to have negative effects, or to even be negative _overall_. Perhaps, for example, the biosecurity research you mig... | https://www.lesswrong.com/posts/RY9XYoqPeMc8W8zbH/mapping-downside-risks-and-information-hazards |
# Making Sense of Coronavirus Stats
_\[Update: Some have pointed out the definition of mortality rate should be that of deaths to some defined population, typically a median estimate over a period of time (week, month, year...) and not limited only to those infected. (New update here. The ratios I have been considerin... | https://www.lesswrong.com/posts/W7dcXash4KvvBngie/making-sense-of-coronavirus-stats |
# How do you survive in the humanities?
Two years ago, I fulfilled a huge part of my dream of doing what I love and started studying creative writing full time. This program is run by a humanities department, so I knew I'd have to stomach epistemic nonsense from time to time, which most of the time I can do while remi... | https://www.lesswrong.com/posts/Sjsvu4CfmKCFJ6mNz/how-do-you-survive-in-the-humanities |
# Curiosity Killed the Cat and the Asymptotically Optimal Agent
[Here's](https://e563b909-928d-4538-97f1-e473938f7515.filesusr.com/ugd/421795_b6f900ffdc194b6b88c25bbb48fb5b63.pdf) a new(ish) paper that I worked on with Marcus Hutter.
If an agent is exploring enough to be guaranteed strong performance in the limit, th... | https://www.lesswrong.com/posts/fSC98Cy3zR9GsEPnT/curiosity-killed-the-cat-and-the-asymptotically-optimal |
# Training Regime Day 6: Seeking Sense
# Introduction
Seeking sense is very easy to describe and very hard to actually do in practice. The basis of this technique is a claim is that the things that go on in other people's heads make sense from their perspective and that it's valuable to try to figure out what's happe... | https://www.lesswrong.com/posts/273Fg7Wig5ufTHpN6/training-regime-day-6-seeking-sense |
# Goal-directed = Model-based RL?
**Epistemic Status**: quick write-up, in reaction to a serendipitous encounter with an idea. I see the main value of this post as decently presenting a potentially interesting take on a concept in AI safety to the community.
While skimming my copy of [Reinforcement Learning: an intro... | https://www.lesswrong.com/posts/Tux9WH4daKcxjEetQ/goal-directed-model-based-rl |
# Training Regime Day 7: Goal Factoring
# Introduction
Humans have these things that we call goals. They also take these things called actions. Ideally, a human takes actions that move the world closer to one of the human's goals. You want your actions to be connected to your goals in this sense.
Usually, when an ac... | https://www.lesswrong.com/posts/FanNherkx3iMHc8AB/training-regime-day-7-goal-factoring |
# Theory and Data as Constraints
There’s a widely-known legend about statistician Abraham Wald’s work on planes during WWII (the veracity of the legend is examined [here](http://www.ams.org/publicoutreach/feature-column/fc-2016-06)). As the story goes, the military collected data on planes coming back from missions, m... | https://www.lesswrong.com/posts/hqSuo8ERcPw6nre2o/theory-and-data-as-constraints |
# Will AI undergo discontinuous progress?
This post grew out of conversations with several people, including [Daniel Kokotajlo](https://www.lesswrong.com/users/daniel-kokotajlo), [grue_slinky](https://www.lesswrong.com/users/grue_slinky) and [Linda Lisefors](https://www.lesswrong.com/users/linda-linsefors), and is bas... | https://www.lesswrong.com/posts/5WECpYABCT62TJrhY/will-ai-undergo-discontinuous-progress |
# Attainable Utility Preservation: Empirical Results
_Reframing Impact_ has focused on supplying the right intuitions and framing. Now we can see how these intuitions about power and the AU landscape both predict and explain AUP's empirical success thus far.
## Conservative Agency in Gridworlds
Let's start with the ... | https://www.lesswrong.com/posts/4J4TA2ZF3wmSxhxuc/attainable-utility-preservation-empirical-results |
# Gary Marcus: Four Steps Towards Robust Artificial Intelligence
Gary Marcus is a pretty controversial figure within machine learning. He's become widely known for being the most outspoken critic of the current paradigm of deep learning, arguing that it will be insufficient to yield AGI. Two years ago he came out with... | https://www.lesswrong.com/posts/CeJs4rPgPtJPNqLMt/gary-marcus-four-steps-towards-robust-artificial |
# Why SENS makes sense
**Summary**
===========
In this post, you'll find why I think SENS Research Foundation (SRF) is great to finance from an EA perspective along with the interview questions I want to ask its Chief Science Officer, Aubrey de Grey. You are welcome to contribute with your own questions in the commen... | https://www.lesswrong.com/posts/rsnNCffHa8EkkdvS2/why-sens-makes-sense |
# Training Regime Day 8: Noticing
# Introduction
In some parts of the world, people go into the forest to hunt mushrooms. The reasons why they do this are not important, but the way that they do it will serve as a guide for noticing.
The problem is that mushrooms are kind of brown and the forest floor is also kind o... | https://www.lesswrong.com/posts/2n83uRi36KWDC9LyK/training-regime-day-8-noticing |
# Continuous Improvement: Insights from 'Topology'
# Foreword
Sometimes you really like someone, but you can't for the life of you understand why. By all means, you should have tired of them long ago, but you keep coming back for more. Welcome, my friend, to [_Topology_](https://www.amazon.com/Topology-2nd-James-Munk... | https://www.lesswrong.com/posts/TSLnckszv4Tb5cHmt/continuous-improvement-insights-from-topology |
# Predictive coding and motor control
Predictive coding is a theory related to how the brain works—[here is the SSC introduction](https://slatestarcodex.com/2017/09/05/book-review-surfing-uncertainty/). As in [my earlier post](https://www.lesswrong.com/posts/cfvBm2kBtFTgxBB7s/predictive-coding-rl-sl-bayes-mpc), I'm ve... | https://www.lesswrong.com/posts/isDCEmYHsosyAjLRK/predictive-coding-and-motor-control |
# Northwest Passage Update
Singing [Stan Rogers](https://en.wikipedia.org/wiki/Stan_Rogers)' [Northwest Passage](https://en.wikipedia.org/wiki/Northwest_Passage_(song)) with friends this evening I realized these days it needs an update. People suggested lyrics, and here's a filked tag to sing after the song:
> *In th... | https://www.lesswrong.com/posts/CkaveqcuFcNDuGBaq/northwest-passage-update |
# Training Regime Day 9: Double-Crux
# Introduction
Argument A is a crux for Alice on position P if finding out that A is false causes Alice to substantially chance their view on P. If Alice thinks P, Bob thinks not P, Alice thinks A, Bob thinks not A, and A/not A are both cruxes for Alice and Bob respectively, then ... | https://www.lesswrong.com/posts/oWv9ZkCcsCATJBrdu/training-regime-day-9-double-crux |
# Information hazards: Why you should care and what you can do
_This post was written for [Convergence Analysis](https://www.convergenceanalysis.org/)._
Overview
========
We argue that many people should consider the risk that they could cause harm by developing or sharing (true) information. We think that harm from... | https://www.lesswrong.com/posts/6ur8vDX6ApAXrRN3t/information-hazards-why-you-should-care-and-what-you-can-do |
# The recent NeurIPS call for papers requires authors to include a statement about the potential broader impact of their work
[NeurIPS](https://en.wikipedia.org/wiki/Conference_on_Neural_Information_Processing_Systems) (formerly NIPS) is a top conference in machine learning and computational neuroscience. The recently... | https://www.lesswrong.com/posts/3xRyt4KnmTMcpgfdX/the-recent-neurips-call-for-papers-requires-authors-to |
# Subagents and impact measures, full and fully illustrated
# 0. Introduction: why yet another post about subagents?
I’ve recently been writing a sequence on how subagents can undermine impact penalties such as [attainable utility preservation](https://arxiv.org/pdf/1902.09725.pdf). I’m not happy with that sequence; ... | https://www.lesswrong.com/posts/mdQEraEZQLg7jtozn/subagents-and-impact-measures-full-and-fully-illustrated |
# Training Regime Day 10: Systemization
# Introduction
Imagine a person who is trying to save money but really enjoys coffee. Every morning, when they're walking to work, they pass their favorite coffee store. Every morning, they ask themselves "is it worth if for me to buy coffee today?" and they make a decision whe... | https://www.lesswrong.com/posts/zRdh8z9k4b4bqk6Bj/training-regime-day-10-systemization |
# Thoughts about Dr Stone and Mythology
I highly recommend the anime/manga Dr Stone. It's a science-backed adventure in a world where modern technology has been lost and needs to be resurrected. If you care about technology, you'll probably enjoy it quite a lot.
There's something you should know before setting off. T... | https://www.lesswrong.com/posts/GnH3qsFJKoHvtvN2G/thoughts-about-dr-stone-and-mythology |
# How Low Should Fruit Hang Before We Pick It?
_Even if we can measure how impactful an agent's actions are, how impactful do we let the agent be? This post uncovers a surprising fact: armed with just four numbers, we can set the impact level so that the agent chooses a reasonable, non-catastrophic plan on the first t... | https://www.lesswrong.com/posts/LfGzAduBWzY5gq6FE/how-low-should-fruit-hang-before-we-pick-it |
# Quarantine Preparations
A month ago I wrote about [disaster preparedness](https://www.jefftk.com/p/disasters), and while the [current coronavirus outbreak](https://en.wikipedia.org/wiki/2019%E2%80%9320_coronavirus_outbreak) had started it wasn't something I knew about yet. Now that there's a real possibility that it... | https://www.lesswrong.com/posts/mQdTCKKB8p5mGaMfC/quarantine-preparations |
# Other versions of "No free lunch in value learning"
Here are some more general results derived from "[Occam's razor is insufficient to infer the preferences of irrational agents](https://arxiv.org/abs/1712.05812)":
* Regularisation is insufficient to make inverse reinforcement learning work.
* Unsupervised learning... | https://www.lesswrong.com/posts/LRYwpq8i9ym7Wuyoc/other-versions-of-no-free-lunch-in-value-learning |
# New article from Oren Etzioni
(Cross-posted from [EA Forum](https://forum.effectivealtruism.org/posts/7x2BokkkemjnXD9B6/new-article-from-oren-etzioni).)
This just appeared in this week’s MIT Technology Review: Oren Etzioni, “[How to know if AI is about to destroy civilization](https://www.technologyreview.com/s/615... | https://www.lesswrong.com/posts/phTQMkcH9Ttc4P9LB/new-article-from-oren-etzioni |
# Training Regime Day 11: Socratic Ducking
# Introduction
As the [story](https://www.wikiwand.com/en/Rubber_duck_debugging) goes, there was once a programmer with a bug. They wanted help solving the bug, so they asked a colleague. In the process of explaining the bug to the college, they solved the bug.
The theory i... | https://www.lesswrong.com/posts/bMw8DSFGBgKD5f4Nw/training-regime-day-11-socratic-ducking |
# Hoarding and Shortages
One of the main responses to yesterday's post on [preparing for a potential quarantine](https://www.jefftk.com/p/quarantine-preparations) was something like:
> *Hoarding causes shortages. Leave masks for people that need them.*
Another commenter made a similar argument with [food](https://ww... | https://www.lesswrong.com/posts/jc4XihXLM5uqKcqmr/hoarding-and-shortages |
# Memetic downside risks: How ideas can evolve and cause harm
_This post was written for [Convergence Analysis](https://www.convergenceanalysis.org/)._
Overview
========
We introduce the concept of **_memetic downside risks (MDR)_: risks of unintended negative effects that arise from how ideas “evolve” over time (as... | https://www.lesswrong.com/posts/EdAHNdbkGR6ndAPJD/memetic-downside-risks-how-ideas-can-evolve-and-cause-harm |
# My slack budget: 3 surprise problems per week
**Previously**: [Slack](https://www.lessestwrong.com/posts/yLLkWMDbC9ZNKbjDG/slack)
In a [couple](https://www.lessestwrong.com/posts/yGycR8tFA3JJbvApp/the-relationship-between-the-village-and-the-mission) [earlier](https://www.lessestwrong.com/posts/zp5AEENssb8ZDnoZR/th... | https://www.lesswrong.com/posts/WuyNHrDXcGFgkZpBy/my-slack-budget-3-surprise-problems-per-week |
# Reviewing the Review
We just spent almost two months reviewing the best posts of 2018. It was a lot of development work, and many LW users put in a lot of work to review and vote on things.
We’ve begun work on the actual printed book, which’ll be distributed at various conferences and events as well as shipped to ... | https://www.lesswrong.com/posts/CakoLjCR9gCXhS3KG/reviewing-the-review |
# If I were a well-intentioned AI...
I: Image classifier
# Introduction: If I were a well-intentioned AI...
I've often [warned](https://intelligence.org/smarter-than-us/) people about the dangers of [anthropomorphising](https://www.lesswrong.com/posts/f4RJtHBPvDRJcCTva/when-anthropomorphism-became-stupid) AIs - how i... | https://www.lesswrong.com/posts/gzWb5kWwzhdaqmyTt/if-i-were-a-well-intentioned-ai-i-image-classifier |
# What will be the big-picture implications of the coronavirus, assuming it eventually infects >10% of the world?
Brainstorming some possibilities, not taking myself too seriously:
--2020 will be remembered as a year of death and disruption; the economy will suffer massively, perhaps enough to count as a recession.
... | https://www.lesswrong.com/posts/RukXjEvMfqDKRJaup/what-will-be-the-big-picture-implications-of-the-coronavirus |
# Where's the Turing Machine? A step towards Ontology Identification
Introduction
============
Assume you are an agent, and you have a model of the world. How do you check that you are in your model, that you exist within it?
To see why I care about this question, let's go back to [embedded agency.](https://www.alig... | https://www.lesswrong.com/posts/tDXFRfkvijTzs2Mmr/where-s-the-turing-machine-a-step-towards-ontology |
# Value of the Long Tail
Suppose we have a self-driving car which works 99% of the time. A human driver only needs to intervene to prevent an accident on one trip out of 100. How much economic value does this generate, relative to a full human-level-or-better self-driving car?
I would guess less than 10%, maybe even ... | https://www.lesswrong.com/posts/Nbcs5Fe2cxQuzje4K/value-of-the-long-tail |
# Training Regime Day 12: Focusing
Note: [Focusing for skeptics](https://medium.com/@ThingMaker/focusing-for-skeptics-6b949ef33a4f) is a far better description than I could give. I will attempt my own description anyway, for triangulation reasons.
2nd note: this technique was invented by Eugene Gendlin. The book [Foc... | https://www.lesswrong.com/posts/vHXe2f59bJHihvcBt/training-regime-day-12-focusing |
# Attainable Utility Preservation: Scaling to Superhuman
I think we're plausibly quite close to the impact measurement endgame. What do we have now, and what remains to be had?
AUP for advanced agents will basically involve restraining their power gain, per the catastrophic convergence conjecture (CCC). For simplici... | https://www.lesswrong.com/posts/S8AGyJJsdBFXmxHcb/attainable-utility-preservation-scaling-to-superhuman |
# [AN #88]: How the principal-agent literature relates to AI risk
Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through this [spreadsheet](https://docs.google.com/spreadsheets/d/1PwWbWZ6FPqAgZWOoOcXM8N_tUC... | https://www.lesswrong.com/posts/y9JeNZ2WAkR6MbBZH/an-88-how-the-principal-agent-literature-relates-to-ai-risk |
# If I were a well-intentioned AI...
II: Acting in a world
[Classifying images](https://www.lesswrong.com/posts/gzWb5kWwzhdaqmyTt/if-i-were-a-well-intentioned-ai-image-classifier) is one thing. But what if I'm an agent that is actually active in some setting?
The previous approach still applies: detecting when I'm ou... | https://www.lesswrong.com/posts/ZKzAjKSeNRtiaeJns/if-i-were-a-well-intentioned-ai-ii-acting-in-a-world |
# Training Regime Day 13: Resolve Cycles
# Introduction
Resolve cycles are a technique that channels and sharpens your innate ability to actually do things.
We're just going to jump right into doing it, then I'll explain why I think it works afterwards. This technique doesn't take very long and is potentially absur... | https://www.lesswrong.com/posts/R9uj2T9NZd4ta3hrG/training-regime-day-13-resolve-cycles |
# Response to Oren Etzioni's "How to know if artificial intelligence is about to destroy civilization"
Oren Etzioni recently wrote a [popular article titled "How to know if artificial intelligence is about to destroy civilization."](https://www.technologyreview.com/s/615264/artificial-intelligence-destroy-civilization... | https://www.lesswrong.com/posts/wiNSeNQT6jiBGZ3Pi/response-to-oren-etzioni-s-how-to-know-if-artificial |
# Resource on alcohol problems
A few people in the EA community have asked about how to help friends who seem to have an alcohol problem. I wrote this resource and am cross-posting it here.
_I'm not an expert on this. Please treat this as the writing of someone with mental health training and some internet research... | https://www.lesswrong.com/posts/dbykHPB7GYYv5GHu5/resource-on-alcohol-problems |
# Abstract Plans Lead to Failure
Epistemic Status: Probably not generalisable to everyone, applies especially strong to me
It's very easy to decide to do something in the abstract on the basis that it should give you some particular skills. Here are some personal examples where I didn't get what I wanted out of my ti... | https://www.lesswrong.com/posts/LLTtFeA3xXWqRtX7T/abstract-plans-lead-to-failure |
# Reasons for Excitement about Impact of Impact Measure Research
Can we get impact measurement _right_? Does there exist One Equation To Rule Them All?
I think there’s a decent chance there _isn’t_ a simple airtight way to implement AUP which lines up with AUP$_\text{conceptual}$, mostly because it’s just incredibly... | https://www.lesswrong.com/posts/wAAvP8RG6EwzCvHJy/reasons-for-excitement-about-impact-of-impact-measure |
# Draft: Models of Risks of Delivery Under Coronavirus
I’ve never considered prophylactically quarantining myself before, but now that I’m considering it I find it contains many more choices than I would have imagined. Let’s take my need to eat- I could go to a supermarket, but that’s full of people. I could get deliv... | https://www.lesswrong.com/posts/82rYtupeRpWXwkHG6/draft-models-of-risks-of-delivery-under-coronavirus |
# Coronavirus: Justified Practical Advice Thread
*(Added: To see the best advice in this thread,* [*read this summary*](https://www.lesswrong.com/posts/B9qzPZDcPwnX6uEpe/coronavirus-justified-practical-advice-summary)*.)*
This is a thread for practical advice for preparing for the coronavirus in places where it might... | https://www.lesswrong.com/posts/LwcKYR8bykM6vDHyo/coronavirus-justified-practical-advice-thread |
# Understandable vs justifiable vs useful
_This is a quickly written post, just sharing a rough conceptual distinction that I’ve found useful in everyday life and that I think others might find useful too._
Your boss just _expected_ you to stay late yesterday. Your partner just shot you a pretty damn sharp comment. S... | https://www.lesswrong.com/posts/kW2J2r4q2S2RZLanY/understandable-vs-justifiable-vs-useful |
# Epistea Workshop Series: Epistemics Workshop, May 2020, UK
Hi all!
We are excited to announce the **Epistea Epistemics Workshop**! This is the first out of a series of 3 workshops happening in Europe, organized and run by the Epistea team.
**Tl,dr:**
* **What:** a 3.5 day workshop focusing on improving our epis... | https://www.lesswrong.com/posts/vbGu44AreFKzC5ZN2/epistea-workshop-series-epistemics-workshop-may-2020-uk |
# If I were a well-intentioned AI...
III: Extremal Goodhart
In this post, I’ll be looking at a more extreme version of the problem in the [previous post](https://www.lesswrong.com/posts/ZKzAjKSeNRtiaeJns/if-i-were-a-well-intentioned-ai-acting-in-a-world).
The [Extremal Goodhart problem](https://www.lesswrong.com/post... | https://www.lesswrong.com/posts/NdJtfujX4sE6xLCsb/if-i-were-a-well-intentioned-ai-iii-extremal-goodhart |
# Refactoring EMH – Thoughts following the latest market crash
In [a twitter thread](https://twitter.com/ESYudkowsky/status/1233174331133284353), Eliezer Yudkowsky challenges the belief in [EMH](https://en.wikipedia.org/wiki/Efficient-market_hypothesis) due to the seemingly illogical behavior of the markets with respe... | https://www.lesswrong.com/posts/ztct6FKw58KsCGf9C/refactoring-emh-thoughts-following-the-latest-market-crash |
# Conclusion to 'Reframing Impact'
.
* * *
COVID-19 could be pretty bad for you. It could affect your travel plans as countries impose quarantines and close off borders. It could affect you materially as supply chains are disrupted and stock market... | https://www.lesswrong.com/posts/5okDRahtDewnWfFmz/seeing-the-smoke |
# What to make of Aubrey de Grey's prediction?
Aubrey de Grey (head of [SENS](https://www.sens.org/)) just had [an appearance on Joe Rogan's podcast](https://www.youtube.com/watch?v=9-z0kglwpwo). At some point during the interview, he made a fairly specific and extremely bold prediction about the near future.
See [56... | https://www.lesswrong.com/posts/oq8zzJsRbu5NQpJkC/what-to-make-of-aubrey-de-grey-s-prediction |
# Trace: Goals and Principles
In terms of research, I decided to devote the month of February mainly to foundations and tools. One project was to come up with a notation/language/framework which matches the way I’ve been thinking about computation - i.e. [DAGs with symmetry](https://www.lesswrong.com/posts/mZy6AMgCw9C... | https://www.lesswrong.com/posts/rt5X74Az3mXwTubRA/trace-goals-and-principles |
# COVID-19 - a good or bad time for extended travel?
A recent loosening of my employment and relational commitments has afforded me the opportunity to do some unencumbered international travel. My intention is to spend a month or more in a single locale so I can get to know the culture and learn from their perspective... | https://www.lesswrong.com/posts/djJ2mNBpuxkvDeH6z/covid-19-a-good-or-bad-time-for-extended-travel |
# Usable implementation of IDA available
Paul Christiano published his and Buck Shlegeris' implementation at https://github.com/paulfchristiano/amplification. It's the code behind the article [Supervising strong learners by amplifying weak experts](https://arxiv.org/abs/1810.08575).
With William Saunders' permission,... | https://www.lesswrong.com/posts/Wij9DSSAyaP9dup5f/usable-implementation-of-ida-available |
# Reminder: Blog Post Day (Unofficial)
In case you missed the [earlier post](https://www.lesswrong.com/posts/q972DHxCwZKFbEPMn/blog-post-day-unofficial), or saw it and forgot: You are invited to join us [online](https://discord.gg/j5TzpVF) today to write that blog post you've been thinking about writing but never got ... | https://www.lesswrong.com/posts/ZCof5J6jKsRyshjoe/reminder-blog-post-day-unofficial |
# Training Regime Day 15: CoZE
Note: this is going to be short because I spent some time doing [COVID-19 preparations](https://www.lesswrong.com/posts/5okDRahtDewnWfFmz/seeing-the-smoke), which I encourage you do to also. I think that it is better for my commitment to post at least one thing a day, even if it's bad. I... | https://www.lesswrong.com/posts/2qmMcGmoT2hW2Gfpa/training-regime-day-15-coze |
# What should be my triggers for initiating self quarantine re: Corona virus
\[Some answer [here](https://www.lesswrong.com/posts/LwcKYR8bykM6vDHyo/coronavirus-justified-practical-advice-thread?commentId=fWv9LSNgHWtKygcog), where I asked this question in comments.\]
Is there some reliable source on the triggers for q... | https://www.lesswrong.com/posts/k6QfEQAhJaJtzoJmJ/what-should-be-my-triggers-for-initiating-self-quarantine-re |
# Cortés, Pizarro, and Afonso as Precedents for Takeover
_Crossposted from [AI Impacts](https://aiimpacts.org/cortes-pizarro-and-afonso-as-precedents-for-ai-takeover/)._
_Epistemic status: I am not a historian, nor have I investigated these case studies in detail. I admit I am still uncertain about how the conquistad... | https://www.lesswrong.com/posts/ivpKSjM4D6FbqF4pZ/cortes-pizarro-and-afonso-as-precedents-for-takeover |
# An Analytic Perspective on AI Alignment
This is a perspective I have on how to do useful AI alignment research. Most perspectives I’m aware of are constructive: they have some blueprint for how to build an aligned AI system, and propose making it more concrete, making the concretisations more capable, and showing th... | https://www.lesswrong.com/posts/8GdPargak863xaebm/an-analytic-perspective-on-ai-alignment |
# Bayesian Inference with Partially Specified Priors
In this post, I am going to write about the practical difficulty of specifying priors that reflect internal beliefs, and about the possibility of performing probabilistic inference with partially specified or imprecise priors. For the most part, I will try to give t... | https://www.lesswrong.com/posts/3HkfkZi5q4eZRabNq/bayesian-inference-with-partially-specified-priors |
# How many people in the Bay Area are infected with Covid-19 as of the end of 2/29/2020?
I want to know your estimate for how many people you think currently have it within the San Francisco Bay Area. _Not_ how many people who have been _measured_ to have it, since our measurement is garbage. Answers can be given as a... | https://www.lesswrong.com/posts/obdmF8kWZiuBXRWgu/how-many-people-in-the-bay-area-are-infected-with-covid-19 |
# How to choose a massage therapist
*Written for Daniel Kokotajlo's unofficial blog post day! I didn't very effort here, so sorry that it's pretty rambly and full of personal anecdotes. I hope it's still at least somewhat helpful!*
Background
==========
Recently several people have asked me how to go about finding ... | https://www.lesswrong.com/posts/epc78F6BBMdoFG6Lq/how-to-choose-a-massage-therapist |
# High Variance Productivity Advice
Despite being literally inapplicable for half the population, my post on birth control and productivity was one of my most popular Facebook posts. In that spirit, here’s a roundup of my favorite weird productivity tips that aren’t for everyone.
I chose hacks that several people rat... | https://www.lesswrong.com/posts/8NFSMdgxeabhEFJpd/high-variance-productivity-advice |
# My Updating Thoughts on AI policy
*My* [*conversation on policy and AI with Richard*](https://www.lesswrong.com/posts/DJB82jKwgJE5NsWgT/some-cruxes-on-impactful-alternatives-to-ai-policy-work) *was over a year ago, so for Daniel Kokotajlo's LW writing day I thought I’d write down new thoughts I’ve thought since then... | https://www.lesswrong.com/posts/DhuPBkKA8ohyZEN8n/my-updating-thoughts-on-ai-policy |
# Blog Post Day Retrospective
Yesterday was Blog Post Day. I think it was a success! In this post I'll link to all the blog posts that were written as part of Blog Post Day, and then say some things about what I learned and when we'll do another one.
The posts (in no particular order) were:
[My Updating Thoughts... | https://www.lesswrong.com/posts/NpZsCERfyKT7cJN9a/blog-post-day-retrospective |
# UML XIII: Online Learning and Clustering
(This is the thirteenth post in a sequence on Machine Learning based on [this book](https://www.amazon.com/Understanding-Machine-Learning-Theory-Algorithms/dp/1107057132). Click [here](https://www.lesswrong.com/s/C8wgFMCmoMNbvDSMP/p/Z29etNPef6yvrPteK) for part I.)
In the pre... | https://www.lesswrong.com/posts/EcaEpLK5RL22NyRGL/uml-xiii-online-learning-and-clustering |
# Training Regime Day 16: Hamming Questions
Note: Over the past 3 days, COVID-19 preparations have taken precedence over writing. There is still more work to do, but I will likely resume on March 7th.
Edit: March 7th is a lie. Probably will be another week or two until time permits.
Later Edit: Turned out to be clos... | https://www.lesswrong.com/posts/apmQ2Qsj8mDC2CWwJ/training-regime-day-16-hamming-questions |
# Decaf vs. regular coffee self-experiment
For a while, I've been convinced that decaf coffee has roughly the same effect on me as regular coffee. However, I haven't been able to say with certainty because there's huge potential for placebo effects. Starting tomorrow, I'll be conducting a (99%) blinded experiment to t... | https://www.lesswrong.com/posts/iEfoE9hmuhzcC34k5/decaf-vs-regular-coffee-self-experiment |
# What are the long-term outcomes of a catastrophic pandemic?
Wei Dai now [writes](https://www.lesswrong.com/posts/RukXjEvMfqDKRJaup/what-will-be-the-big-picture-implications-of-the-coronavirus#2mcb949cKj2TxpiRK),
> The world is probably going to lose 5 2.5-10% of its population (380 190-760 million, see [here](https... | https://www.lesswrong.com/posts/KhFTCgprxSFgvxas6/what-are-the-long-term-outcomes-of-a-catastrophic-pandemic |
# Is Coronavirus active in Summer?
It seems like the corona virus has spread mostly in cold climates on the northern hemisphere (in winter) and there are very few (confirmed) cases in Africa, South America and Australia. On the one hand the first two probably don't have the capabilities for testing on a large scale, ... | https://www.lesswrong.com/posts/AuK2o7gZZRzWLyGu4/is-coronavirus-active-in-summer |
# "What Progress?" Gwern's 10 year retrospective
This is a thrilling read. Some favourite lines:
> And that was that. The FBI would later pay me some visits, but I was long done with the DNMs and had moved on.
and
> The key result was Rietveld et al 2013, the first truly successful IQ GWAS. Rietveld et al 2013 foun... | https://www.lesswrong.com/posts/sRchPdp6mCqY2ekJX/what-progress-gwern-s-10-year-retrospective |
# Is there any value in self-quarantine (from Coronavirus), if you live with other people who aren't taking similar precautions?
I live in a house with four other people, including two that I share a bathroom with.
Does it help me at all to self quarantine, and not leave the house, if my housemates continue to go to ... | https://www.lesswrong.com/posts/466CcCvZNKtvE5vgA/is-there-any-value-in-self-quarantine-from-coronavirus-if |
# If I were a well-intentioned AI...
IV: Mesa-optimising
Here I apply my "If I were a well-intentioned AI" filter to [mesa-optimising](https://www.lesswrong.com/posts/q2rCMHNXazALgQpGH/conditions-for-mesa-optimization)
Now, I know that a mesa-optimiser need not be a subagent (see 1.1 [here](https://arxiv.org/pdf/1906... | https://www.lesswrong.com/posts/aqhMLqaoHb7uob7fr/if-i-were-a-well-intentioned-ai-iv-mesa-optimising |
# Coronavirus is Here
Even in the best case scenarios, things are going to get a lot worse from here before they get better.
Please, please, _please, _do not rely on me here to tell you what is going on or what to do. Even more than that, please, please, _please,_ do not take this as saying that you shouldn’t do _a l... | https://www.lesswrong.com/posts/j9kxNcm2bM7A8XAc6/coronavirus-is-here |
# Epistemic standards for “Why did it take so long to invent X?”
In seeking to understand the history of progress, I keep running across intriguing cases of “[ideas behind their time](https://marginalrevolution.com/marginalrevolution/2010/11/ideas-behind-their-time.html)”—inventions that seem to have come along much l... | https://www.lesswrong.com/posts/DGoK3QeaXFu5Mvxh8/epistemic-standards-for-why-did-it-take-so-long-to-invent-x |
# How should I be thinking about the risk of air travel (re: Coronavirus)?
Is there any information about how much risk of exposure increases by air-travel?
Like **[Ŀady Jade Beacham](https://www.lesswrong.com/posts/LwcKYR8bykM6vDHyo/coronavirus-justified-practical-advice-thread?commentId=rQaZKExsqxrWu5row)**, I am c... | https://www.lesswrong.com/posts/jHTofPvPcTaWCHRgY/how-should-i-be-thinking-about-the-risk-of-air-travel-re |
# What does Solomonoff induction say about brain duplication/consciousness?
Back in 2012, in a [thread](https://www.greaterwrong.com/posts/PXoWk554FZ4Gpfvah/causal-reference/comment/5K9ewNasTGWmkfQdM) on LW, Carl Shulman wrote a couple of comments connecting Solomonoff induction to brain duplication, epiphenomenalism,... | https://www.lesswrong.com/posts/ZtYRbL5o7gLgbe5CM/what-does-solomonoff-induction-say-about-brain-duplication |
# Robustness to fundamental uncertainty in AGI alignment
Publication cycles are long, so I've talked about [some](https://www.lesswrong.com/posts/3r44dhh3uK7s9Pveq/rfc-philosophical-conservatism-in-ai-alignment-research) of the [ideas](https://www.lesswrong.com/posts/uR3znuBnaevssYDZY/rfc-meta-ethical-uncertainty-in-a... | https://www.lesswrong.com/posts/JYdGCrD55FhS4iHvY/robustness-to-fundamental-uncertainty-in-agi-alignment-1 |
# What is the appropriate way to communicate that we are experiencing a pandemic?
After spending 3 days reading all of Lesswrong's posts and comments (and sources cited) related to coronavirus, I started taking action (summarizing the information collected and planning the least harmful ways to alert my closest peers ... | https://www.lesswrong.com/posts/zTK8rRLr6RT5yWEmX/what-is-the-appropriate-way-to-communicate-that-we-are |
# Bleach
Traditional chlorine bleach, appropriately diluted, can be used for disinfecting and sanitizing, so it's potentially very handy to have in emergencies. It can also be confusing, however, because it degrades, there are many kinds, concentrations are not emphasized, and you want different dilutions for differen... | https://www.lesswrong.com/posts/QJfiKwicwTXYMzJ7q/bleach |
# Is there a better way to define groups for COVID-19 impact?
I think everyone who has posted a table of stats for COVID-19 infection or deaths seems to do a 10 year grouping. For example [here](https://www.lesswrong.com/posts/j9kxNcm2bM7A8XAc6/coronavirus-is-here). (Used only because I was looking at that post when i... | https://www.lesswrong.com/posts/ruaFhjFQg2zxQrRvZ/is-there-a-better-way-to-define-groups-for-covid-19-impact |
# Did everyone miss the big thing about your phone?
Have not read through all the comments, and probably won't, so perhaps this has already been pointed out.
I don't see that my phone is any more at risk than my hands, I can wipe the desk down with a sterilizing wipe, and do the same with my phone if I want. I tend t... | https://www.lesswrong.com/posts/LLFbjZc5wC5kTaT97/did-everyone-miss-the-big-thing-about-your-phone |
# To mask or not mask
A rather large disconnect between what I hear from medical side and how people act seem clear. The medical officials all say for the most part the general population, other than those infected, don't benefit from wearing the masks.
In part that is due to improper use. I discount that because I'v... | https://www.lesswrong.com/posts/BfvtSFq53Tv7z9jp9/to-mask-or-not-mask |
# What "Saving throws" does the world have against coronavirus? (And how plausible are they?)
Following up on [my earlier question](https://www.lesswrong.com/posts/RukXjEvMfqDKRJaup/what-will-be-the-big-picture-implications-of-the-coronavirus), which asked about the implications of coronavirus infecting >10% of the wo... | https://www.lesswrong.com/posts/5RQiZMnsMFBAHqu2C/what-saving-throws-does-the-world-have-against-coronavirus |
# [AN #89]: A unifying formalism for preference learning algorithms
Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through this [spreadsheet](https://docs.google.com/spreadsheets/d/1PwWbWZ6FPqAgZWOoOcXM8N_t... | https://www.lesswrong.com/posts/7bNXqdDPYpnfCNQhA/an-89-a-unifying-formalism-for-preference-learning |
# For viruses, is presenting with fatigue correlated with causing chronic fatigue?
Intuitively, one might expect that viruses which cause fatigue during acute infection are more likely to also cause chronic fatigue later. Is this in fact the case?
COVID-19 causes fatigue in [about 70% of cases](https://www.worldomete... | https://www.lesswrong.com/posts/uvmsoBqEDiynnJcMS/for-viruses-is-presenting-with-fatigue-correlated-with |
# Covid-19: Things I'm Doing Differently
The coronavirus is now spreading through multiple communities, and while there [isn't yet](https://www.boston.com/news/health/2020/03/03/coronavirus-covid-19-cases-massachusetts) evidence of it in Boston I expect there will be soon. Because of the [long incubation period](https... | https://www.lesswrong.com/posts/45R2Tz4Fq3dchnHm5/covid-19-things-i-m-doing-differently |
# Motte/bailey doctrine is often a byproduct of distributed argumentation
confidence: I think I'm on to something
(I'm posting this publicly because I'd like it to get passed around for corrections, both nitpicky (typos, thinkos) and important (glaring errors in logic). I need to whip up a better title for it, too.... | https://www.lesswrong.com/posts/AGZT2baXPmoEXhjeE/motte-bailey-doctrine-is-often-a-byproduct-of-distributed |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.