rational
Issue #95: Full List
19 January, 2020 // View curated list# Instrumental
Reality-Revealing and Reality-Masking Puzzles // AnnaSalamon, 15 min Favorite
Key Decision Analysis - a fundamental rationality technique // elityre, 3 min
# Epistemic
Studying Early Stage Science: Research Program Introduction // habryka, 18 min
Red Flags for Rationalization // dspeyer, 4 min
Testing for Rationalization // dspeyer, 3 min
Clarifying The Malignity of the Universal Prior: The Lexical Update // interstice, 3 min
Avoiding Rationalization // dspeyer, 2 min
Using Expert Disagreement // dspeyer, 6 min
Can we always assign probabilities? // MichaelA, 14 min
Against Rationalization II: Sequence Recap // dspeyer, 1 min
# Ai
Update on Ought's experiments on factored evaluation of arguments // Owain_Evans, 1 min
Remote AI alignment writing group seeking new members // rmoehn, 1 min
Artificial Intelligence and Life Sciences (Why Big Data is not enough to capture biological systems?) // HansNauj, 7 min
What is the relationship between Preference Learning and Value Learning? // Riccardo Volpato, 1 min
Fire Alarm for AGI // user134723, 1 min
# Meta-ethics
What do we mean by “moral uncertainty”? // MichaelA, 15 min
# Anthropic
In Defense of the Arms Races… that End Arms Races // Gentzel, 3 min
How has the cost of clothing insulation changed since 1970 in the USA? // DanielFilan, 1 min
What long term good futures are possible. (Other than FAI)? // Donald Hobson, 1 min
# Decision theory
Being a Robust, Coherent Agent (V2) // Raemon, 8 min
The Road to Mazedom // Zvi, 8 min
How to Identify an Immoral Maze // Zvi, 5 min
How to Escape From Immoral Mazes // Zvi, 23 min
ACDT: a hack-y acausal decision theory // Stuart_Armstrong, 8 min
CDT going bonkers... forever // Stuart_Armstrong, 1 min
The Alignment-Competence Trade-Off, Part 1: Coalition Size and Signaling Costs // Gentzel, 3 min
Is backwards causation necessarily absurd? // Chris_Leong, 1 min
Risk and uncertainty: A false dichotomy? // MichaelA, 23 min
# Math and cs
Malign generalization without internal search // Matthew Barnett, 4 min
In defense of deviousness // Juan Andrés Hurtado Baeza, 5 min
UML VI: Stochastic Gradient Descent // sil ver, 13 min
# Ea
Tips on how to promote effective altruism effectively? Less talk, more action. // culturechange, 1 min
# Relationships
Go F*** Someone // Jacobian, 9 min
# Community
Bay Solstice 2019 Retrospective // mingyuan, 18 min
Why a New Rationalization Sequence? // dspeyer, 3 min
Drawing Toward Power // hiblick, 2 min
What are beliefs you wouldn't want (or would feel apprehensive about being) public if you had (or have) them? // Mati_Roy, 1 min
What plausible beliefs do you think could likely get you diagnosed with a mental illness by a psychiatrist? // Mati_Roy, 1 min
Austin LW/SSC Far-comers Meetup: Feb. 8, 1:30pm // jchan, 1 min
2019 Adversarial Collaboration Winners // Scott Alexander, 8 min
# Culture war
Conversational Cultures: Combat vs Nurture (V2) // Ruby, 10 min Favorite
Repossessing Degrees // jkaufman, 2 min
The Internet of Beefs // Venkatesh Rao, 19 min
Predictable Identities 24: Anti-Identity // Jacob Falkovich, 1 min
# Misc
"human connection" as collaborative epistemics // toonalfrink, 2 min
Ascetic aesthetic // Dominik Tujmer, 3 min
Using Vickrey auctions as a price discovery mechanism // epiphi, 1 min
"How quickly can you get this done?" (estimating workload) // kerspoon, 5 min
Political Roko's basilisk // Abhimanyu Pallavi Sudhir, 1 min
How does a Living Being solve the problem of Subsystem Alignment? // Alan Givré, 1 min
What types of compute/processing could we distinguish? // MoritzG, 1 min
Is NYC Building Much Housing? // jkaufman, 1 min
Contra Contra Contra Caplan On Psych // Scott Alexander, 16 min
# Podcasts
102 – Decade in Review // The Bayesian Conspiracy, 142 min
#44 Classic episode - Paul Christiano on finding real solutions to the AI alignment problem // The 80000 Hours team, 231 min
#181 — The Illusory Self // Waking up with Sam Harris, 92 min
# Videos of the week
How civilization could destroy itself -- and 4 ways we could prevent it | Nick Bostrom // TED, 21 min