Daring Greatly by Bayesian Investor – Fairly positive book review, some chapters were valuable and it was an easy read. How to overcome shame and how it differs from guilt. Perfectionism vs healthy striving. If you stop caring about what others think you lose your capacity for connection
Politics and Economics:
Trust And Diver by Bryan Caplan – Robert Putnam’s work is often cited as showing the costs of diversity. However Putnam’s work shows the negative effect of diversity on trust is rather modest. On the other hand Putnam found multiple variables that are much more correlated with trust (such as home ownership).
Historical Returns Market Portfolio by Tyler Cowen – From 1960 to 2015 the global market portfolio realized a compounded real return of 4.38% with a std of 11.6%. Investors beat savers by 3.24%. Link to the original paper.
Introducing The Ea Involvement Guide by The Center for Effective Altruism (EA forum) – A huge list of concrete actions you can take to get involved. Every action has a brief description and a link to an article. Each article rates the action on time commitment, duration, familiarity and occupation. Very well put together.
Corrigibility by Paul Christiano – Paul defines the sort of AI he wants to build, he refers to such systems as “corrigible”. Paul argues that a sufficiently corrigible agent will become more corrigible over time. This implies that friendly AI is not a narrow target but a broad basin of attraction. Corrigible agents prefer to build other agents that share the overseers preferences, not their own. Predicting that the overseer wants me to turn off when he hits the off-button is not complicated relative to being deceitful. Comparison with Eliezer’s views.
Openai Baselines Dqn by Open Ai – “We’re open-sourcing OpenAI Baselines, our internal effort to reproduce reinforcement learning algorithms with performance on par with published results.” Best practices we use for correct RL algorithm implementations. First release: DQN and three of its variants, algorithms developed by DeepMind.
The End Of The World According To Isis by Waking Up with Sam Harris – Graeme Wood. His experience reporting on ISIS, the myth of online recruitment, the theology of ISIS, the quality of their propaganda, the most important American recruit to the organization, the roles of Jesus and the Anti-Christ in Islamic prophecy, free speech and the ongoing threat of jihadism.
Silinks Is Golden by Scott Alexander – Standard SSC links post.
Some Unsong Guys by Scott Alexander (Scratchpad) – Pictures of Unsong Fan Art.
Thought Experiment Coarsegrained Vr Utopia by cousin_it (lesswrong) – Assume an AI is running a Vr simulation that is hooked up to actual human brains. This means that the AI only has to simulate nature at a coarse grained level. How hard would it be to make that virtual reality a utopia?
A Call To Adventure by Robin Hanson – Meaning in life can be found by joining or starting a grand project. Two possible adventures: Promoting and implementing futarchy (decision making via prediction markets). Getting a real understanding of human motivation.
Cash is King by GiveDirectly – Eight media articles about Effective Altruism and Cash transfers.
G Reliant Skills Seem Most Susceptible To Automation by Freddie deBoer – Computers already outperform humans in g-loaded domains such as Go and Chess. Many g-loaded jobs might get automated. Jobs involving soft or people skills are resilient to automation.
Jason Khalipa by Tim Ferriss – “8-time CrossFit Games competitor, a 3-time Team USA CrossFit member, and — among other athletic feats — he has deadlifted 550 pounds, squatted 450 pounds, and performed 64 pullups at a bodyweight of 210 pounds.”
What Is Depression Anyway: The Synapse Hypothesis by Scott Alexander – Six seemingly distinct treatments for depression. How at least six can be explained by considering synapse generation rates. Skepticism that this method can be used to explain anything since the body is so inter-connected. Six points that confuse Scott and deserve more research. Very technical.
Deep Reinforcement Learning From Human Preferences by Dario Amodei, Paul Christiano & Alex Ray (Open Ai) – An algorithm learns to backflip with 900 bits of feedback from the human evaluator. “One step towards building safe AI systems is to remove the need for humans to write goal functions, since using a simple proxy for a complex goal, or getting the complex goal a bit wrong, can lead to undesirable and even dangerous behavior. In collaboration with DeepMind’s safety team, we’ve developed an algorithm which can infer what humans want by being told which of two proposed behaviors is better.”
The Rationalist-sphere And The Less Wrong Wiki by Deku-shrub (lesswrong) – What’s next for the Lesswrong wiki. A distillation of Lesswrong. Fully indexing the diaspora. A list of communities. Spreading rationalist ideas. Rationalist Research.
Persona 5 Spoiler Free Review by Zvi Moshowitz – Persona games are long but deeply worthwhile if you enjoy the gameplay and the story. Persona 5 is much more polished but Persona 3 has a more meaningful story and more interesting decisions. Tips for Maximum Enjoyment of Persona 5. Very few spoilers.
The World Desperately Needs Ai Strategists: Heres How To Become One by 80,000 Hours – 80K hours released a detailed guide to careers in AI policy. ” We discuss the main career paths; what to study; where to apply; how to get started; what topics are most in need of research; and what progress has been made in the field so far.” Transcript included.
Projects Id Like To See by William MacAskill (EA forum) – CEA is giving out £100K grants. General types of applications. EA outreach and Community, Anti-Debates, Prediction Tournaments, Shark Tank Discussions, Research Groups, Specific Skill Building, New Organizations, Writing.
Open Philanthropy Project is Now an Independent Organization by Open Philosophy – The evolution of Open Phil. Why should Open Phil split from GiveWell. LLC structure.
Separating GiveWell and the Open Philanthropy Project by The GiveWell Blog – The GiveWell perspective. Context for the sale. Effect on donors who rely on GiveWell. Organization changes at GiveWell. Steps taken to sell Open Phil assets. The new relationship between GiveWell and Open Phil.
Politics and Economics:
Is Economics Science by Noah Smith – No one knows what a Science is. Thoeries that work (4 examples). The empirical and credibility revolutions. Why we still need structural models. Ways economics could be more scientific. Data needs to kill bad theories. Slides from Noah’s talk are included and worth playing but assume familiarity with the economics profession.
Why Optimism is More Rational than Pessimism by TheMoneyIllusion – Splitting 1900-2017 into Good and Bad periods. We learn something from our mistakes. Huge areas where things have improved long term. Top 25 movies of the 21st Century. Artforms in decline.
Sea Problem by protokol2020 – A fun problem. Measuring sea level rise.
Tania Lombrozo On Why We Evolved The Urge To Explain by Rational Speaking – “Research on what purpose explanation serves — i.e., why it helps us more than our brains just running prediction algorithms. Tania and Julia also discuss whether simple explanations are more likely to be true, and why we’re drawn to teleological explanations”
Don Bourdreaux Emergent Order by EconTalk – “Why is it that people in large cities like Paris or New York City people sleep peacefully, unworried about whether there will be enough bread or other necessities available for purchase the next morning? No one is in charge–no bread czar. No flour czar.”
We Are The Athenians Not The Spartans by wubbles (lesswrong) – “Our values should be Athenian: individualistic, open, trusting, enamored of beauty. When we build social technology, it should not aim to cultivate values that stand against these. High trust, open, societies are the societies where human lives are most improved.”
The Precept Of Universalism by H i v e w i r e d – “Universality, the idea that all humans experience life in roughly the same way. Do not put things or ideas above people. Honor and protect all peoples.” Eight points expanding on how to put people first and honor everyone.
Where Do Hypotheses Come From by c0rw1n (lesswrong) – Link to a 25 page article. “Why are human inferences sometimes remarkably close to the Bayesian ideal and other times systematically biased? In particular, why do humans make near-rational inferences in some natural domains where the candidate hypotheses are explicitly available, whereas tasks in similar domains requiring the self-generation of hypotheses produce systematic deviations from rational inference. We propose that these deviations arise from algorithmic processes approximating Bayes’ rule.”
Bayes: A Kinda Sorta Masterpost by Nostalgebraist – A long and very well thought-out criticism of Bayesianism. Explanation of Bayesian methodology. Comparison with classical statistics. Arguments for Bayes. The problem of ignored hypotheses with known relations. The problem of new ideas. Where do priors come from? Regularization and insights from machine learning.
Tasting Godhood by Agenty Duck – Poetic and personal. Wine tasting. Empathizing with other people. Seeing others as whole people. How to dream about other people. Sci-fi futures. Tasting godhood is the same as tasting other people. Looking for your own godhood.
Epistemology Vs Critical Thinking by Onemorenickname (lesswrong) – Epistemies work. General approaches don’t work. Scientific approaches work. Epistemic effort vs Epistemic status. Criticisms of lesswrong Bayesianism.
The Precept Of Niceness by H i v e w i r e d – Prisoner’s Dilemma’s. Even against a truly alien opponent you should still cooperate as long as possible on the iterated prisoner’s dilemma, even with fixed round lengths, play tit-for-tat. Niceness is the best strategy.
Politics and Economics:
Summing Up My Thoughts On Macroeconomics by Noah Smith – Slides from Noah’s talk at the Norwegian Finance Ministry. Comparison of Industry, Central Bank and Academic Macroeconomics. Overview of important critiques of academic macro. The DGSE standard mode and ways to improve it. What makes a good Macro theory. Go back to the microfoundations.
81 Leaving Islam by Waking Up with Sam Harris – “Sarah Haider. Her organization Ex-Muslims of North America, how the political Left is confused about Islam, “rape culture” under Islam, honesty without bigotry, stealth theocracy, immigration, the prospects of reforming Islam”
SSC Journal Club Ai Timelines by Scott Alexander – A new paper surveying what Ai experts think about Ai progress. Contradictory results about when Ai will surpass humans at all tasks. Opinions on Ai risk, experts are taking the arguments seriously.
Summer Programming by Jacob Falkovich (Put A Number On It!) – Jacob’s Summer writing plan. Re-writing part of the lesswrong sequences. Ribbonfarm’s longform blogging course on refactored perception.
How Men And Women Perceive Relationships Differently by AellaGirl – Survey Results about Relationship quality over time. Lots of graphs and a link to the raw data. “In summary, time is not kind. Relationships show an almost universal decrease in everything good the longer they go on. Poly is hard, and you have to go all the way to make it work – especially for men. Religion is also great, if you’re a man. Women get more excited and insecure, men feel undesirable.”
Dichotomies by mindlevelup – 6 short essays about dichotomies and whats useful about noticing them. Fast vs Slow thinking. Focused vs Diffuse Mode. Clean vs Dirty Thinking. Inside vs Outside View. Object vs Meta level. Generative vs Iterative Mode. Some conclusions about the method.
Announcing Effective Altruism Grants by Maxdalton (EA forum) – CEA is goving out £100,000 grants for personal projects. “We believe that providing those people with the resources that they need to realize their potential could be a highly effective use of resources.” A list of what projects could get funded, the list is very broad. Evaluation criteria.
Review Of Ea New Zealands Doing Good Better Book by cafelow (EA forum) – New Zealand EAs gave out 250 copies of “Doing Good Better”. 80 of the recipients responded to a follow up survey. The results were extremely encouraging. Survey details and discussion. Possible flaws with the giveaway and survey.