Ivy Zhang
1 followers · 3 following · 881 views
on the atlas — 172
- The Science of Scientific Writing | American Scientist5 savers
- Averageness2 savers
- Mini Blog Post 24: On Procrastination - The art of shaping your future actions — Neel Nanda3 savers
- A Visual Guide to Evolution Strategies | 大トロ4 savers
- HOW TO GET LUCKY — Almanack of Naval Ravikant2 savers
- Some reasons to work on productivity and velocity4 savers
- An invitation to a secret society - by Adam Mastroianni14 savers
- Learning Sensorimotor Capabilities in Cellular Automata1 savers
- Baumol effect11 savers
- Home | Levers For Progress37 savers
- Go smol or go home | Harm de Vries3 savers
- hamming advice5 savers
- Slack - LessWrong5 savers
- Amelia Wattenberger10 savers
- There’s more to mathematics than rigour and proofs | What's new13 savers
- On Really Trying · Gwern.net13 savers
- How I use memory systems7 savers
- Language models can explain neurons in language models6 savers
- Reading Well - by Simon Sarris - The Map is Mostly Water6 savers
- Betteridge's law of headlines4 savers
- Interpretability Dreams5 savers
- steal like an artist | Alicia Guo4 savers
- Mini Blog Post 9: What it means to optimise — Neel Nanda7 savers
- AGI and the EMH: markets are not expecting aligned or unaligned AI in the next 30 years — Basil Halperin9 savers
- Bicycle – Bartosz Ciechanowski5 savers
- Codon usage bias3 savers
- Andrea Botez (@andreabotez) / Passes1 savers
- The builder's mindset: a way out of the "drill sergeant" / "Zen master" dichotomy3 savers
- A Coder Considers the Waning Days of the Craft | The New Yorker3 savers
- The Maintenance Race - Works in Progress5 savers
- You and Your Research, by Richard Hamming - Sam Altman6 savers
- The Unreasonable Effectiveness of Recurrent Neural Networks15 savers
- Doing research. Some generic advice for PhD students | by Paul Niehaus | Medium1 savers
- An Opinionated Guide to ML Research52 savers
- Nadia Asparouhova | The independent researcher19 savers
- Self-signaling the ability to do what you want2 savers
- Bridging Systems: Open Problems for Countering Destructive Divisiveness Across Ranking, Recommenders, and Governance | Knight First Amendment Institute1 savers
- Superlinear Returns11 savers
- The free energy principle—a precis - Dialectical Systems1 savers
- What's going on with assembly theory?1 savers
- Artificial General Intelligence Is Already Here4 savers
- I Left Out the Full Truth to Get My Climate Change Paper Published | The Free Press3 savers
- attention metrics - @visakanv's blog1 savers
- Is Science Stagnant? - The Atlantic7 savers
- Hyena Hierarchy: Towards Larger Convolutional Language Models · Hazy Research2 savers
- Deep Reinforcement Learning: Pong from Pixels2 savers
- [2211.13746] Melting Pot 2.01 savers
- arielle lok20 savers
- Grants Only Go So Far4 savers
- The Incredible Machine (1968) - YouTube1 savers
- Quick thoughts on research12 savers
- On the Evolution of Credit Cultures2 savers
- Generations of Verse - Editorial - Le Random — A Digital Generative Art Institution1 savers
- New HSK 7/8/9 Word List1 savers
- What if Wargames Could Be Something Else? | The MIT Press Reader1 savers
- Are some personalities just better?4 savers
- What I learned from reading about writing1 savers
- Artificial intelligence is a familiar-looking monster, say Henry Farrell and Cosma Shalizi2 savers
- Keep Your Identity Small5 savers
- Google "We Have No Moat, And Neither Does OpenAI"28 savers
- Getting Senpai To Notice You1 savers
- Meta-Prompt: A Simple Self-Improving Language Agent3 savers
- Don't Fear the Terminator - Scientific American Blog Network2 savers
- How to Do Great Work80 savers
- The days are long but the decades are short - Sam Altman72 savers
- Cultivating a state of mind where new ideas are born65 savers
- There’s no speed limit56 savers
- sparkly people and how to find them - by Anson Yu53 savers
- How can we develop transformative tools for thought?52 savers
- Reality has a surprising amount of detail49 savers
- Mini Blog Post 3: Become a person who Actually Does Things — Neel Nanda45 savers
- Choose Good Quests - by Trae Stephens and Markie Wagner44 savers
- Learning By Writing42 savers
- Understanding41 savers
- Why You Should Start a Blog Right Now39 savers
- A Mathematical Framework for Transformer Circuits39 savers
- Advice for ambitious 19 year olds - Sam Altman37 savers
- Invisible Details of Interaction Design33 savers
- Notes on “Taste” — Are.na33 savers
- Meditations On Moloch | Slate Star Codex32 savers
- https://cs.stanford.edu/~jsteinhardt/ResearchasaStochasticDecisionProcess.html31 savers
- Learn In Public30 savers
- The Most Precious Resource is Agency - by Simon Sarris29 savers
- The Best Textbooks on Every Subject - LessWrong28 savers
- Laws of Tech: Commoditize Your Complement · Gwern.net26 savers
- A Gentle Introduction to Graph Neural Networks26 savers
- An app can be a home-cooked meal26 savers
- Hypertext as Literature25 savers
- Value Beyond Instrumentalization - Letters to a Young Technologist25 savers
- School Is Not Enough25 savers
- cdixon | What the smartest people do on the weekend is what everyone else will do during the week in ten years24 savers
- cdixon | Climbing the wrong hill24 savers
- Whitepaper — The Collective Intelligence Project24 savers
- on slowness, taste, and living well - by Isabel - Mind Mine23 savers
- Principles of Effective Research | Michael Nielsen23 savers
- Looking for Alice - by Henrik Karlsson - Escaping Flatland23 savers
- Post 38: On Slack - Having room to be excited — Neel Nanda23 savers
- Zoom In: An Introduction to Circuits22 savers
- How to write good prompts: using spaced repetition to create understanding21 savers
- Fuck your Miracle Year - by Roger’s Bacon - Secretorum21 savers
highlights — 428
In the startup world, the name for this principle is "do things that don't scale." If you pay a ridiculous amount of attention to your tiny initial set of customers, ideally you'll kick off exponential growth by word of mouth. But this same principle applies to anything that grows exponentially. Learning, for example. When you first start learning something, you feel lost. But it's worth making the initial effort to get a toehold, because the more you learn, the easier it will get.
Superlinear ReturnsWhat else can we learn from these fields? In all of them you have to put in the initial effort. Superlinear returns seem small at first. At this rate, you find yourself thinking, I'll never get anywhere. But because the reward curve rises so steeply at the far end, it's worth taking extraordinary measures to get there.
Superlinear ReturnsTake as much risk as you can afford; if you're not failing occasionally you're probably being too conservative. Seek out the best colleagues. Develop good taste and learn from the best examples. Be honest, especially with yourself. Exercise and eat and sleep well and avoid the more dangerous drugs. When in doubt, follow your curiosity. It never lies, and it knows more than you do about what's worth paying attention to.
Superlinear ReturnsChoose work you have a natural aptitude for and a deep interest in. Develop a habit of working on your own projects; it doesn't matter what they are so long as you find them excitingly ambitious.
Superlinear ReturnsThe most obvious way to take advantage of superlinear returns for performance is by doing exceptionally good work. At the far end of the curve, incremental effort is a bargain. All the more so because there's less competition at the far end — and not just for the obvious reason that it's hard to do something exceptionally well, but also because people find the prospect so intimidating that few even try. Which means it's not just a bargain to do exceptional work, but a bargain even to try to.
Superlinear ReturnsIndeed, the forms of exponential growth that don't consist of learning are so often intermixed with it that we should probably treat this as the rule rather than the exception. Which yields another heuristic: always be learning. If you're not learning, you're probably not on a path that leads to superlinear returns.
Superlinear ReturnsBut the most important thresholds of all are those representing new discoveries. Knowledge seems to be fractal in the sense that if you push hard at the boundary of one area of knowledge, you sometimes discover a whole new field. And if you do, you get first crack at all the new discoveries to be made in it. Newton did this, and so did Durer and Darwin.
Superlinear ReturnsThe most common case of exponential growth in preindustrial times was probably scholarship. The more you know, the easier it is to learn new things. The result, then as now, was that some people were startlingly more knowledgeable than the rest about certain topics.
Superlinear ReturnsThe most obvious case of superlinear returns is when you're working on something that grows exponentially. For example, growing bacterial cultures. When they grow at all, they grow exponentially. But they're tricky to grow. Which means the difference in outcome between someone who's adept at it and someone who's not is very great.
Superlinear ReturnsAGI promises to generate great value in the years ahead, yet it also poses significant risks. The natural questions we should be asking in 2023 include: “Who benefits?” “Who is harmed?” “How can we maximize benefits and minimize harms?” and “How can we do this fairly and equitably?” These are pressing questions that should be discussed directly instead of denying the reality of AGI.
Artificial General Intelligence Is Already HereStacking several neural nets together in the right way yields a model that can perform the same calculations as any given computer program.
Artificial General Intelligence Is Already HereThe days are long but the decades are short.
The days are long but the decades are short - Sam AltmanDo new things often. This seems to be really important. Not only does doing new things seem to slow down the perception of time, increase happiness, and keep life interesting, but it seems to prevent people from calcifying in the ways that they think. Aim to do something big, new, and risky every year in your personal and professional life.
The days are long but the decades are short - Sam AltmanTell your parents you love them more often. Go home and visit as often as you can.
The days are long but the decades are short - Sam AltmanGo out of your way to help people. Few things in life are as satisfying. Be nice to strangers. Be nice even when it doesn’t matter.
The days are long but the decades are short - Sam AltmanExercise. Eat well. Sleep. Get out into nature with some regularity.
The days are long but the decades are short - Sam AltmanIf you think you’re going to regret not doing something, you should probably do it. Regret is the worst, and most people regret far more things they didn’t do than things they did do. When in doubt, kiss the boy/girl.
The days are long but the decades are short - Sam AltmanAsk for what you want.
The days are long but the decades are short - Sam AltmanMinimize your own cognitive load from distracting things that don’t really matter. It’s hard to overstate how important this is, and how bad most people are at it. Get rid of distractions in your life. Develop very strong ways to avoid letting crap you don’t like doing pile up and take your mental cycles, especially in your work life.
The days are long but the decades are short - Sam AltmanGo out of your way to be around smart, interesting, ambitious people. Work for them and hire them (in fact, one of the most satisfying parts of work is forging deep relationships with really good people). Try to spend time with people who are either among the best in the world at what they do or extremely promising but totally unknown. It really is true that you become an average of the people you spend the most time with.
The days are long but the decades are short - Sam AltmanHave clear goals for yourself every day, every year, and every decade.
The days are long but the decades are short - Sam AltmanDon’t waste time. Most people waste most of their time, especially in business.
The days are long but the decades are short - Sam AltmanTalk to people more. Read more long content and less tweets. Watch less TV. Spend less time on the Internet.
The days are long but the decades are short - Sam AltmanDon’t spend time trying to maintain relationships with people you don’t like, and cut negative people out of your life. Negativity is really bad. Don’t let yourself make excuses for not doing the things you want to do.
The days are long but the decades are short - Sam AltmanLife is not a dress rehearsal—this is probably it.
The days are long but the decades are short - Sam AltmanNever put your family, friends, or significant other low on your priority list. Prefer a handful of truly close friends to a hundred acquaintances. Don’t lose touch with old friends. Occasionally stay up until the sun rises talking to people. Have parties.
The days are long but the decades are short - Sam AltmanThink about risk the right way. Drew Houston gave a great commencement speech where he said you only have to be right once. That’s true. The risk is not getting on the path where you get to be right that one critical time.
Advice for ambitious 19 year olds - Sam AltmanNo matter what you choose, build stuff and be around smart people.
Advice for ambitious 19 year olds - Sam AltmanBut when the image was processed by a different, more open-ended algorithm that Michaeli designed
The AI Tools Making Images Look Better | Quanta MagazineThe reality of publishing things on the internet is that interesting things get boosted, and boring things get ignored. So people are basically guaranteed to think your posts are much more interesting that they actually are, because they’re more likely to see your interesting posts.
Some blogging mythsI think the reason I keep writing these blog posts encouraging people to blog is that I love reading people’s personal stories about how they do stuff with computers, and I want more of them.
Some blogging mythsUltimately morale is physical. You think with your body, so it's important to take care of it. That means exercising regularly, eating and sleeping well, and avoiding the more dangerous kinds of drugs. Running and walking are particularly good forms of exercise because they're good for thinking.
How to Do Great WorkThe people you spend time with will also have a big effect on your morale. You'll find there are some who increase your energy and others who decrease it, and the effect someone has is not always what you'd expect. Seek out the people who increase your energy and avoid those who decrease it. Though of course if there's someone you need to take care of, that takes precedence.
How to Do Great WorkDon't marry someone who doesn't understand that you need to work, or sees your work as competition for your attention. If you're ambitious, you need to work; it's almost like a medical condition; so someone who won't let you work either doesn't understand you, or does and doesn't care.
How to Do Great WorkOne of the biggest mistakes ambitious people make is to allow setbacks to destroy their morale all at once, like a balloon bursting. You can inoculate yourself against this by explicitly considering setbacks a part of your process. Solving hard problems always involves some backtracking.
How to Do Great WorkBut this cycle also operates in the other direction: if you're not doing good work, that can demoralize you and make it even harder to. Since it matters so much for this cycle to be running in the right direction, it can be a good idea to switch to easier work when you're stuck, just so you start to get something done.
How to Do Great WorkQuality is more important than quantity in colleagues. It's better to have one or two great ones than a building full of pretty good ones. In fact it's not merely better, but necessary, judging from history: the degree to which great work happens in clusters suggests that one's colleagues often make the difference between doing great work and not.
How to Do Great WorkThe young have no idea how rich they are in time. The best way to turn this time to advantage is to use it in slightly frivolous ways: to learn about something you don't need to know about, just out of curiosity, or to try building something just because it would be cool, or to become freakishly good at something.
How to Do Great WorkSo err on the side of starting. Which is easier when starting means starting small; those two ideas fit together like two puzzle pieces.
How to Do Great WorkBeing prolific is underrated. The more different things you try, the greater the chance of discovering something new. Understand, though, that trying lots of things will mean trying lots of things that don't work. You can't have a lot of good ideas without also having a lot of bad ones.
How to Do Great WorkThe initial versions of big things were often just experiments, or side projects, or talks, which then grew into something bigger. So start lots of small things.
How to Do Great WorkOriginality in choosing problems seems to matter even more than originality in solving them. That's what distinguishes the people who discover whole new fields.
How to Do Great WorkChanging your context can help. If you visit a new place, you'll often find you have new ideas there. The journey itself often dislodges them. But you may not have to go far to get this benefit. Sometimes it's enough just to go for a walk.
How to Do Great WorkTalking or writing about the things you're interested in is a good way to generate new ideas. When you try to put ideas into words, a missing idea creates a sort of vacuum that draws it out of you. Indeed, there's a kind of thinking that can only be done by writing.
How to Do Great WorkFor example, you're much more likely to have original ideas when you're working on something. Original ideas don't come from trying to have original ideas. They come from trying to build or understand something slightly too difficult.
How to Do Great WorkI've never liked the term "creative process." It seems misleading. Originality isn't a process, but a habit of mind. Original thinkers throw off new ideas about whatever they focus on, like an angle grinder throwing off sparks. They can't help it.
How to Do Great WorkBut you never need to worry about that problem, because it's self-solving if you work on sufficiently ambitious projects.
How to Do Great WorkEveryone knows to avoid distractions at work, but it's also important to avoid them in the other half of the cycle. When you let your mind wander, it wanders to whatever you care about most at that moment. So avoid the kind of distraction that pushes your work out of the top spot, or you'll waste this valuable type of thinking on the distraction instead. (Exception: Don't avoid love.)
How to Do Great WorkThe way to beat it is to stop occasionally and ask yourself: Am I working on what I most want to work on?"
How to Do Great WorkOne reason per-project procrastination is so dangerous is that it usually camouflages itself as work. You're not just sitting around doing nothing; you're working industriously on something else. So per-project procrastination doesn't set off the alarms that per-day procrastination does. You're too busy to notice it.
How to Do Great Work