rational
Issue #123: Full List
2 August, 2020 // View curated listArguments for driverless dilemmas in the context of autonomous vehicles are "too contrived to be of practical use and an inappropriate method for making decisions" // drive.google.com
# Coronavirus
New Paper on Herd Immunity Thresholds // Zvi, 11 min
Covid 7/30: Whack a Mole // Zvi, 13 min
Food Spending During Covid // jkaufman, 1 min
# Instrumental
Attention is your scarcest resource // benkuhn, 5 min
Engaging Seriously with Short Timelines // deluks917, 4 min
Free Educational and Research Resources // wearsshoes, 7 min
How I use Google Forms in my Workflow // harri-besceli, 3 min
Jam is obsolete // jkaufman, 1 min
Basic Conversational Coordination: Micro-coordination of Intention // elityre, 2 min
Change the world a little bit // AllAmericanBreakfast, 2 min
On Procrastination - The art of shaping your future actions // neel-nanda-1, 9 min
# Epistemic
Enforcing Type Distinction // lionhearted, 2 min
Why You Might Want a New Way to Visualize Biases // dtm, 3 min
"Go west, young man!" - Preferences in (imperfect) maps // Stuart_Armstrong, 3 min
The Conceited Folly of Certainty // Deep Dives, 24 min
The Future of Science // ricraz, 9 min
# Ai
Are we in an AI overhang? // andyljones, 5 min Favorite
Developmental Stages of GPTs // orthonormal, 8 min
What specific dangers arise when asking GPT-N to write an Alignment Forum post? // matthew-barnett, 1 min
Predictions for GPT-N // hippke, 1 min
You Can Probably Amplify GPT3 Directly // zachary-robertson, 7 min
What Failure Looks Like: Distilling the Discussion // Benito, 8 min
What if memes are common in highly capable minds? // daniel-kokotajlo, 2 min
Learning the prior and generalization // evhub, 5 min
[updated] how does gpt2′s training corpus capture internet discussion? not well // nostalgebraist, 2 min
Power as Easily Exploitable Opportunities // TurnTrout, 7 min
The "best predictor is malicious optimiser" problem // donald-hobson, 2 min
Probability that other architectures will scale as well as Transformers? // daniel-kokotajlo, 1 min
To what extent are the scaling properties of Transformer networks exceptional? // abramdemski, 1 min
Is the work on AI alignment relevant to GPT? // Richard_Kennaway, 1 min
Does the lottery ticket hypothesis suggest the scaling hypothesis? // daniel-kokotajlo, 1 min
What happens to variance as neural network training is scaled? What does it imply about "lottery tickets"? // abramdemski, 1 min
Agentic Language Model Memes // FactorialCode, 2 min
AI and Efficiency // DragonGod, 1 min
A community-curated repository of interesting GPT-3 stuff // rudi-c
The Rise of Commonsense Reasoning // DragonGod
Likely AGI architectures? // Anirandis, 1 min
Interpretability in ML: A Broad Overview // lifelonglearner, 17 min
FHI Report: How Will National Security Considerations Affect Antitrust Decisions in AI? An Examination of Historical Precedents // Cullen_OKeefe, 1 min
How Will National Security Considerations Affect Antitrust Decisions in AI? An Examination of Historical Precedents // Cullen_OKeefe, 1 min
Where do people discuss doing things with GPT-3? // skybrian, 1 min
# Anthropic
Would a halfway copied brain emulation be at risk of having different values/identity? // Ghatanathoah, 2 min
# Decision theory
Generalized Efficient Markets in Political Power // johnswentworth, 10 min
The Era Of Unlimited Everything: Unlimited Materials & Unlimited Money // jdampolargo, 12 min
Asset Prices Consistently Violate Efficient Market Hypothesis // Liron, 1 min
Improving local governance in fragile states - practical lessons from the field // rockthecasbah, 6 min
Principles Behind Bottlenecks // curi, 8 min
What Future Areas Matter Most? // Robin Hanson, 1 min
# Math and cs
Pure Bayesianism vs Newcomb's Paradox // le-nguyen-hoang-1, 18 min
Godel in second-order logic? // abhimanyu-pallavi-sudhir, 1 min
Structured Tasks for Language Models // zachary-robertson, 1 min
# Ea
Billionaire Economics // Virgil Kurkjian, 1 min Favorite
# Community
Delegate a Forecast // Amandango, 1 min
Online LessWrong Community Weekend, September 11th-13th // Linda Linsefors, 2 min
Rationalist Reading Group (Online) // AspiringRationalist, 1 min
# Culture war
The Curse of Cursory Idealism // Deep Dives, 3 min
# Misc
What a 20-year-lead in military tech might look like // daniel-kokotajlo, 20 min Favorite
Let Your Mind Be Not Fixed // gworley, 3 min
Can you gain weirdness points? // NicholasKross, 1 min
Parameters of Privacy // Raemon, 7 min
The problem of Pop Analogy // thesilvermen, 1 min
Incentivized Guardians // Robin Hanson, 6 min
# Podcasts
#83 - Jennifer Doleac on preventing crime without police and prisons // The 80000 Hours team, 143 min
EP69 Rachel Haywire on Free Thinking & Expression // The Jim Rutt Show, 67 min
c10: Infinite Zebras // Constellation, 93 min
Currents 010: Tyson Yunkaporta on Humans As Custodial Species // The Jim Rutt Show, 52 min
116 – Truth, Falsity, and Censorship // The Bayesian Conspiracy, 141 min
#82 - James Forman Jr on reducing the cruelty of the US criminal legal system // The 80000 Hours team, 88 min
EP68 Mara Zepeda on Innovative Collaboration // The Jim Rutt Show, 87 min
# Videos of the week
What happens when biology becomes technology? | Christina Agapakis // TED, 10 min