rational
Issue #186: Full List
17 October, 2021 // View curated list
Abstract representations emerge naturally in neural networks trained to perform multiple tasks. // www.biorxiv.org
# Coronavirus
Covid 10/14: Less Long Cvoid // Zvi, 8 min
Long Covid Informal Study Results // pktechgirl, 6 min
Niacin as a treatment for covid? (Probably no, but I’m glad we’re checking) // pktechgirl, 14 min
Interpreting Diagnostic Tests // Tornus, 4 min
Quantifying Risk // dpandey, 2 min
Outdoor dancing is likely very safe // jkaufman, 1 min
Yet Another Covid Risk Calculator // thomas-colthurst, 1 min
# Instrumental
Without a phone for 10 days // benjamin-spiegel, 1 min
What's your plan? // jkaufman, 1 min
# Epistemic
Cup-Stacking Skills (or, Reflexive Involuntary Mental Motions) // Duncan_Sabien, 9 min
Intelligence, epistemics, and sanity, in three short parts // ozziegooen, 4 min
Is anchoring a reliable cognitive bias? // justindomke, 9 min
Forget AGI alignment, are you aligned with yourself? // samuel-shadrach, 1 min
# Ai
NVIDIA and Microsoft releases 530B parameter transformer model, Megatron-Turing NLG // Ozyrus, 1 min
On Solving Problems Before They Appear: The Weird Epistemologies of Alignment // adamShimi, 18 min
NLP Position Paper: When Combatting Hype, Proceed with Caution // sbowman, 1 min
The Extrapolation Problem // lsusr, 2 min
Modeling Risks From Learned Optimization // ben-cottier, 14 min
The evaluation function of an AI is not its aim // yair-halberstadt, 3 min
Memetic hazards of AGI architecture posts // Ozyrus, 1 min
The AGI needs to be honest // rokosbasilisk, 2 min
"Redundant" AI Alignment // mckay-jensen, 1 min
Good AI alignment online class? // alexander-1, 1 min
Compute Governance and Conclusions - Transformative AI and Compute [3/4] // lennart, 6 min
[Proposal] Method of locating useful subnets in large models // quintin-pope, 2 min
Commentary on “AGI Safety From First Principles by Richard Ngo, September 2020” // nonmali-1, 23 min
# Meta-ethics
Is nuking women and children cheaper than firebombing them? // lsusr, 1 min
Is moral duty/blame irrational because a person does only what they must? // [email protected], 6 min
# Anthropic
Consciousness, Free Will and Scientific Objectivity in Perspective-Based Reasoning // dadadarren, 4 min
Do you think you are a Boltzmann brain? If not, why not? // Jack Ryan, 1 min
# Decision theory
EDT with updating double counts // paulfchristiano, 7 min
# Math and cs
Optimization Concepts in the Game of Life // Vika, 12 min
# Books
Book Review: How To Talk So Little Kids Will Listen // weft, 9 min
Book Analysis: New Thrawn Trilogy // ozymandias, 5 min
Book Review: Feeling Great by David Burns // Fei, 4 min
Book Review: A Pattern Language by Christopher Alexander // lincolnquirk, 2 min
Book Review: Rise and Fall of the Great Powers // TheRealSlimHippo, 20 min
Books Review: "Sapiens", "Homo Deus" and "21 Lessons For The 21st Century", by Yuval Noah Harari // Luca Parodi, 18 min
Book Review: Open Borders // sam-enright, 14 min
Book Review: Denial of Death // PatrickDFarley, 44 min
What if we should use more energy, not less? // Snorkelfarsan, 7 min
Book Review: The End of Average // Vanilla_cabs, 10 min
Book Review: All I Want To Know Is Where I'm Going To Die So I'll Never Go There // Anmoljain, 55 min
Book Review: Existential Risk and Growth // jkim2, 8 min
Notes from the book ‘First Three Minutes’ by Steven Weinberg // invertedpassion, 11 min
Book Review: Philosophical Investigations by Wittgenstein // Chris_Leong, 36 min
# Community
My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage) // jessica.liu.taylor, 27 min
Zoe Curzi's Experience with Leverage Research // Ilverin
Meetup Advice #0: Sequence intro // mingyuan, 1 min
Meetup Advice #1: Choosing a venue // mingyuan, 14 min
In Wikipedia — reading about Roko's basilisk causing "nervous breakdowns" ... // ThurstonBT, 1 min
# Culture war
Choice Writings of Dominic Cummings // Connor_Flexman, 6 min Favorite
Status Explains Lots // Robin Hanson, 5 min
Highlights From The Comments On Kids And Climate Change // Scott Alexander, 20 min
Please Don't Give Up On Having Kids Because Of Climate Change // Scott Alexander, 13 min
# Misc
Creating a truly formidable Art // Valentine, 28 min
Secure homes for digital people // paulfchristiano, 11 min
Write Surprisingly About Reality // lsusr, 1 min
Framing Practicum: Semistable Equilibrium // AllAmericanBreakfast, 3 min
[linkpost] Fantasia for Two Voices // mike_hawke, 1 min
Paying people to vote for good policies. // Matteodimaio, 3 min
# Podcasts
The genetic lottery (Kathryn Paige Harden) // Rationally Speaking, 64 min
EP143 John Vervaeke on Awakening from the Meaning Crisis: Part 1 // The Jim Rutt Show, 112 min
# Rational fiction
Bayeswatch 13: Spaceship // lsusr, 1 min
Bayeswatch 12: The Singularity War // lsusr, 2 min
# Videos of the week
[ML News] Microsoft trains 530B model | ConvMixer model fits into single tweet | DeepMind profitable // Yannic Kilcher, 27 min