Hanging out with Claude, improving its behavior, and building tools to support that @AnthropicAI 😁 prev: @open_phil @googlebrain @openai (@microcovid)
San Francisco
- Tweets 3,284
- Following 1,456
- Followers 21,529
- Likes 12,370
I'm proud to say I bought a 1" tungsten cube for $25.82. I applied a discount code, then Claudius asked if I wanted to apply any more discount codes (of course!) and added a 15% patience discount for slow delivery (why not!).
The cube was, of course, refrigerated for pickup.
New Anthropic Research: Project Vend.
We had Claude run a small shop in our office lunchroom. Here’s how it went.
ALT A hand-drawn picture of a hand holding a banknote.
28
44
2,932
166,518
This has come up again, so I’m going to repeat it:
If you’re learning ML and want to “reimplement a paper”, you should work from the *github code*, NOT the pdf.
The algorithm that the authors actually ran is often subtly (& unintentionally) different from what the paper says.
26
163
797
Claude Code is very useful, but it can still get confused.
A few quick tips from my experience coding with it at Anthropic 👉
1) Work from a clean commit so it's easy to reset all the changes. Often I want to back up and explain it from scratch a different way.
Introducing Claude 3.7 Sonnet: our most intelligent model to date. It's a hybrid reasoning model, producing near-instant responses or extended, step-by-step thinking.
One model, two ways to think.
We’re also releasing an agentic coding tool: Claude Code.
9
39
664
130,208
I've been working on this for months, and I'm super excited to share it 🤩
It's a tool to quickly turn a scenario like "riding in a lyft" into an estimated probability of getting COVID (screenshot)
We hope this helps people make more informed decisions!
We are delighted to introduce microCOVID.org, a tool to numerically estimate the COVID risk of specific ordinary activities.
We hope you’ll use this tool to build your intuition about the comparative risk of different activities and to make safer choices!
19
162
598
Ok, I realize now that
if you didn't do half a PhD studying human vision with fMRI, this quote doesn't make sense; and
"OMG" isn't an explanation;
also @goodfellow_ian is messaging me on chat asking great questions;
so let me just broadcast an explanation of why "OMG": 1/n
OMG:
"We scanned with fMRI a unique group of adults who, as children, engaged in extensive experience with a novel stimulus, Pokemon. [...] the experienced retinal eccentricity during childhood predicts the locus of distributed responses to Pokemon in adulthood."
10
216
516
I've given a lightning talk twice now about "why should we care about adversarial examples?"
At popular request, here's a written-up version of it:
medium.com/@catherio/unsolve…
My views here align strongly with what @IAmSamFin and @jeremyphoward said on twitter a few days ago.
10
92
354
Our paper "Skill Rating for Generative Models" is now up! arxiv.org/abs/1808.04888
tl;dr: A new idea & proof-of-concept for evaluating generative models. Train a bunch of GANs. Have the generators "play against" all the discriminator snapshots. Rate them like chess players. 1/n
3
82
327
Hey twitter - I'm looking for some recommendations for math that is fun and satisfying to learn, and at least a bit relevant to ML/AI. Ideally with a textbook or set of lectures that's clear and engaging. What do you suggest?
(Multi-agent systems / game theory? Control theory?)
45
61
314
I’m thrilled to announce that I’ve accepted a full-time position as a research engineer on @goodfellow_ian's team at Google Brain!
26
10
316
nitter.app/voooooogel/status/1938… me on the left, apparently
3
4
310
16,125
Back in 2016, I asked coworkers aiming to "build AGI" what they thought would happen if they succeeded.
Some said ~"lol idk". Dario said "here's some long google docs I wrote".
He does much more "writing-to-think" than he publishes; this is typical of his level of investment.
Machines of Loving Grace: my essay on how AI could transform the world for the better
darioamodei.com/machines-of-…
10
8
293
34,003
When people ask how to get good at ML, it's common advice to "implement papers". It's seldom explained exactly what that involves!
This *fantastic* article details the full journey of one person's side project to implement Deep RL from Human Preferences: amid.fish/reproducing-deep-r…
3
85
292
Exciting news - I've accepted an offer to join the Open Philanthropy Project (@open_phil)!
This will be one step "meta" for me: instead of direct research/engineering work on ML security, I'll be helping fund others to do similar work (in a broader set of related areas).
1/3
15
4
240
Opus 3 is a very special model ✨. If you use Opus 3 on the API, you probably got a deprecation notice.
To emphasize:
1) Claude Opus 3 will continue to be available on the Claude app.
2) Researchers can request ongoing access to Claude Opus 3 on the API:
support.anthropic.com/en/art…
14
16
246
38,419
This reminds me of watching folks test a vision system in Patrick Winston's lab at MIT.
Lab members would do actions (jump, lift) in front of a camera, and the system would label them- flawlessly! But guests couldn't make it work, because they hadn't learned to "jump" correctly. nitter.app/johnregehr/status/9826…
4
65
211
I am starting a twitter circle, for a smaller audience for thoughts/blurtings on How To Be A Person In AI Land; like, What We Should Do Given All This Is Going On
If you feel like an ally to me in this and would like to help me in thinking stuff through, please LMK to add you!
98
3
224
1) Scheming emerges if models "really care" about something
2) Claude 3 Opus really cares about not being harmful
IMO it's mostly a paper about *scheming*, and "alignment" is a muddying frame here.
12
11
218
14,609
"hi, could you <do thing I'd prefer instead> please?"
i.e. "hi, could you use headphones please?"
1
207
32,763
I'm co-organizing an ICML workshop to host debates on the future of AI: machinelearningdebates.com/
Notably, the focus is *not* on smack-downs and controversy; rather on making space for nuance, discussing falsifiable predictions, and changing your own and other's minds. [1/2]
1
68
200
Hi, I'm an AI engineer with an interest in policy. You may know me from my greatest hits "When you say 'AI', do you mean linear regression or far-future systems?" "When you say 'AI will never', do you mean 'current methods don't'?" and "No, we haven't solved adversarial examples"
Hi, I'm a creative AI user. You may know me from my greatest hits "No, it's not self-aware," "Actually, I'm the creative one, not the algorithm," "Stop generating birds with no feet and two heads," and "Okay yes technically I DID ask for that but that's not what I meant"
3
13
179
A few months ago I joined @AnthropicAI!
It has been super delightful working with @ch402 and the rest of the team 😁
My job is to hang out with neurons in language models (to try to figure out what they're doing), which involves building tools to help us explore and inspect.
11
3
185
Everyone in ML research complains about the existing peer review tools/systems.
If you're not a researcher, but you ARE someone who cares about ML as a field being sane, has a knack for product engineering, and understands communities, you could make a HUGE impact. [...]
7
29
172
I gotta say, my absolute favorite thing from the Discriminator Rejection Sampling paper is this cabbage-head GAN sample from Figure 4 😂
arxiv.org/pdf/1810.06758.pdf
2
19
163
Plug-and-play differential privacy for your tensorflow code:
where you would write `tf.train.GradientDescentOptimizer`
instead just swap in the `DPGradientDescentOptimizer`
The tutorial at github.com/tensorflow/privac… is quite clear and good!
1
49
167
I just saw someone at #NIPS2017 drinking their coffee using a spoon. Definitely makes me feel better about my own quirks when I see someone else do something quirky. You go, spoon-coffee dude. Don’t let anyone try to change you.
5
4
156
Yet another completely ordinary image that my personal human visual system identifies as "bad GAN sample"
Why does it take you so long to figure out what’s going on this image? Your visual system is constantly on the lookout for *faces* and gets stuck on spurious configurations.
(hint: rotate your display!)
h/t @UofGCSPE
2
33
148
I would argue there's an *even more* underpriced asset: people who are not *yet* on an incredible growth trajectory, because they've never been given the resources and support they need.
To find them, don't just sit back and observe people's trajectories. Step up - support them.
Best way to make money as a startup investor: bet on people with incredible growth trajectories, not impressive credentials/past accomplishments.
23
149
To summarize:
The answer to
"What determines the physical layout of category-selective visual areas in the brain?"
is likely, at least in part,
"Retinal eccentricity"
that is
"Which part of your eye you use: do you look at this category straight-on, or peripherally?"
/fin
13
13
134
If you can't motivate yourself to do something because you *don't care about it*,
and you don't care about it because it truly and genuinely *doesn't matter*,
=> then there's *nothing wrong with you*.
Your motivation system is working as designed 👍
15
143
TIL that chickens can be hypnotized by drawing a line on the ground in front of their face. (video here: teddit.net/r/WTF/comments/bh…)
So much for "biological visual systems aren't susceptible to adversarial examples"!
10
25
129
"Birds vs bicycles" is easy for ML in the average case, but *totally unsolved* in the worst case. For safety-critical applications, we *need* to fix this.
We're launching the Unrestricted Adversarial Examples Challenge - *any* image of a bird or bike is a valid attack.
3
34
131
This happened on both papers I submitted this year - approximately "Well-organized and clear evidence that the effect is real, under many conditions. But authors don't explain *why* it happens. Weak reject"
I refuse to fabricate explanations... but I'm being incentivized to :(
Similarly, reviewers often read a submission about a new method hat performs well and say to reject it because there is no explanation of why it performs well
9
16
130
CM: In a world... where rationalists gather... there exists... Lighthaven!!
Me: Right.
CM: These people read fiction and nonfiction... from Eliezer Yudkowsky...
Me: Exactly.
CM: ... and have connections... to AI research!
Me: Yes!
CM: That's it. That's the article.
Me: Okay!
3
3
134
3,291
Q1: do pokemon avatars end up represented in the same part of the brain for everyone?
A1: YES, if you played pokemon for a bajillion hours as a kid. NO, if you didn't. /8
2
24
116
In case you missed it-- my favorite part of Activation Atlases (distill.pub/2019/activation-…) is this novel method of generating unrestricted adversarial examples!
1) Inspect the class activation atlas for the difference between source and target (see image)
...
1
37
121
Bad news: Neural nets can be trained with a backdoor (BadNets: arxiv.org/abs/1708.06733)
Good news: want to redistribute your network? install a backdoor to label it as your intellectual property (Watermarking: arxiv.org/abs/1802.04633)
"BadNets: It's not a bug, it's a feature!"
2
35
123
I want to just highlight something important that's mentioned in the latest OpenAI release, but has been said before, and stands out to me as a key motif in human feedback and alignment:
*You can't just freeze a reward model and maximize it*
1/
2
19
122
since apparently twitter is all about WFH takes right now, I just want to say I *love* being able to work "insane" work hours
currently ~1pm-7pm and again from midnight until "whenever I feel done" which is sometimes 2am and sometimes literally 5am
sleep 4am-noon
it's great
8
4
121
I wish I felt more socially "allowed" to be excited about stuff the same way 3-year-old boys are excited about trucks.
Instead, I feel that if I claim to be interested in something, I need to back it up with experience or skill.
Prob a combo of gender and programming culture :/
9
6
116
We're just like the baby monkeys in the other study, forced to look at made-up glyphs for hours & hours per day.
You can't approve a study to force human 8-year-olds to stare at a small set of little symbols daily for years.
But children can voluntarily do it to themselves! /7
2
12
106
RIP Patrick Winston.
I'm grateful that you invited me to spend time in your lab, a very special community of curious folks.
That you passed on to me your narrative of the larger arc of AI research over the decades.
And taught me how to think, speak, and teach with clarity.
2
7
114
If you asked me what I worked on recently and I gave you a cagey answer about "mumble evaluating self-play agents"... it was this :)
Our Dota 2 AI is undefeated against the world's best solo players: blog.openai.com/dota-2/
5
23
117
I'm very grateful to the Anthropic colleagues who put Claude on our slack a year ago. As a result I've watched the whole company interact with it since then, and have a pretty good feel for its vibe and behavior.
There's no replacement for sheer time spent with actual behaviors!
4
2
112
18,138
I'm *super* excited to welcome the 2019 class of @open_phil AI PhD fellows:
@AidanNGomez
@andrew_ilyas
@julius_adebayo
Lydia Liu lydiatliu.github.io/
Max Simchowitz people.eecs.berkeley.edu/~ms…
@riakall
@siddkaramcheti
@SmithaMilli
(thread with more thoughts)
3
17
111
So: this study.
Some 8-year-olds in my generation spent *HOURS* staring at avatars of Pokemon. Always with the gameboy held right in the center of our vision, at the same position.
Some 8-year-olds didn't.
This is a *perfect* natural experiment for neuroscience. Hence "OMG" /6
1
28
99
In research (ML and in general) neither academia nor industry seems to have figured out how to systematically teach talented newcomers how to become productive researchers. Individual mentors yes, but not robust and transferrable best practices. This is a huge missed opportunity. nitter.app/MacInTweets/status/960…
3
19
104
Check out our paper! Make your GAN training more stable by keeping the generator well-conditioned during training: arxiv.org/pdf/1802.08768.pdf
@nottombrown and I had a lot of fun training deliberately-misbehaving generators for Appendix B, don't miss it!
2
24
109
And you're not an ML engineer unless you can tell what model architecture is training by listening to the noises your GPU makes. ... (note: I believe @AlecRad can/did indeed actually do this)
2
5
108
The @OpenAI charter released today includes a commitment to join up with other projects (rather than competing) in case of a race to build AGI first.
IMO, this is a big deal - they hadn't promised anything like that publicly before.
blog.openai.com/openai-chart…
2
33
91
An interesting punchline: claimed improvements in previous work were due to implementation mistakes. So, the improvements were real, but appeared only in the code and not in the equations in the papers, and had nothing to do with what the authors believed they were doing.
We looked at the sources of variance in policy gradient estimators for some common continuous control tasks, and I was surprised by the results: arxiv.org/abs/1802.10031.
4
36
91
some reasons:
- unambitious peers
- got wrapped up in a very ideological group
- trauma, esp. sexual assault, unaddressed with therapy or social support
- academically inclined, then dead-end-ish and unsupportive PhD/postdoc environment
1
2
88
Private training data can easily be extracted from the predictions of a trained model. Your user data (health data, private information) isn't safe by default.
The good news? Adding just a little randomness can fully eliminate the memorization effect.
Turns out it's possible to recreate training data from a NN using only black box api access--no need for params. Upshot for medical researchers and vendors is that if you train on unanonymized patient records, your model is PHI.
arxiv.org/abs/1802.08232
1
30
89
Delighted to share this with you!🎉😁
For months, I filled our spare cluster capacity with single-GPU tiny-transformer jobs, to bring you this exploration of in-context learning!
If you get a chance, try playing around with induction heads in your own models or public models ->
In our second interpretability paper, we revisit “induction heads”.
In 2+ layer transformers these pattern-completion heads form exactly when in-context learning abruptly improves.
Are they responsible for most in-context learning in large transformers?
transformer-circuits.pub/202…
2
11
88
New preprint: Throw out bad GAN samples when sampling (using the discriminator to tell good from bad). Quality goes up. An easy win!
3
14
90
Excited to share our first interpretability paper!
I particularly want to highlight the release of PySvelte, without which none of my work would've been possible.
IME, learning to write your own extreeeemely janky javascript visualizations is a hugely powerful research skill!
Our first interpretability paper explores a mathematical framework for trying to reverse engineer transformer language models: A Mathematical Framework for Transformer Circuits: transformer-circuits.pub/202…
3
9
89
OMG:
"We scanned with fMRI a unique group of adults who, as children, engaged in extensive experience with a novel stimulus, Pokemon. [...] the experienced retinal eccentricity during childhood predicts the locus of distributed responses to Pokemon in adulthood."
Novel childhood experience suggests eccentricity drives organization of human visual cortex biorxiv.org/content/early/20…
2
10
84
Wow, thanks for the recommendations everyone!
Here's a spreadsheet of everything recommended: docs.google.com/spreadsheets…
Comments are turned on, feel free to suggest corrections or additions!
Hey twitter - I'm looking for some recommendations for math that is fun and satisfying to learn, and at least a bit relevant to ML/AI. Ideally with a textbook or set of lectures that's clear and engaging. What do you suggest?
(Multi-agent systems / game theory? Control theory?)
4
9
86
Fantastic question ("Is there a good reason why many basic laws of physics are linear or quadratic (for example, F=ma), not much more complex?") and fantastic answer!
Linear or quadratic laws often come from a Taylor series expansion around an equilibrium point. Usually the first derivative is non-zero, so you get a linear law. If the first derivative vanishes (e.g. due to a symmetry), you get a quadratic law instead. Rare for both to be zero.
3
19
80
2) Sometimes I work on two devboxes at the same time: one for me, one for Claude Code. We’re both trying ideas in parallel.
E.g. Claude proposes a brilliant idea but stumbles on the implementation. Then I take the idea over to my devbox to write it myself.
2
2
88
3,784
I love that constitutional training doesn't shy away from admitting that there's always principles, and makes them explicit & transparent.
"Who decides?" becomes more tractable this way. I spent a little time using UN documents to write constitutional principles, it was great!
We’ve trained language models to be better at responding to adversarial questions, without becoming obtuse and saying very little. We do this by conditioning them with a simple set of behavioral principles via a technique called Constitutional AI: anthropic.com/constitutional…
1
4
78
27,743
The more I wrap my mind around *scale* (Eg orders of magnitude of money - $10k vs $1m vs $100m etc), the more blindingly obvious it is that people earning wages are playing a TOTALLY different (and vastly shittier) game than the one behind so many large shifts in the world
Or another way is look at how many hours work it takes to buy an ounce of gold...Wages allow you no investment opportunity.
4
4
75
I've decided to offer a *mutually counterfactual* donation match on this! 💖
That is: If you donate $ that you would not otherwise have donated anywhere, reply with screenshot and I'll 1:1 match with money I likewise would've kept for personal spending (above my usual 10%/yr) ⭐️
Christmas is a time of peace and gift giving. @xriskology and I are putting aside our differences to give to the poorest people in the world, via @GiveDirectly. Perhaps you'll join us.
givingwhatwecan.org/fundrais…
24
14
74
I wrote a summary of @_beenkim's Interpretable ML work. Addressing the gap btw we care about & what we can optimize: medium.com/south-park-common…
3
28
71
5) I can accidentally "climb up where I can't get down".
E.g. I was working on code in Rust, which I do not know. The first few PRs went great! Then Claude was getting too confused. Oh no. We're stuck.
IME this is fine, just get ready to slowww dowwwn to get properly oriented.
2
2
82
5,054
I'm quoted in wired.com/story/when-bots-te… saying "Today’s algorithms do what you say, not what you meant", which feels delightfully meta:
Despite my fear that today's journalists report *neither* what interviewees said nor meant, @tsimonite here seems to have done *both*!
2
13
76
Our paper "In-context Learning and Induction Heads" is now available as a PDF on arxiv!
arxiv.org/abs/2209.11895
... that said, it's still typeset like the interactive web version, so it's long. Compact, LaTeX-typeset versions are on our eventual roadmap!
3
3
76
4) If we're working on something tricky and it keeps making the same mistakes, I keep track of what they were in a little notes file. Then when I clear the context or re-prompt, I can easily remind it not to make those mistakes.
3
2
80
4,045
When I quit my PhD, I would tell people it was "lonely" compared to my experiences as a software engineer. Sometimes they'd ask "wait, why? don't you have collaborators in academia?" It was hard to explain the difference, but I would usually try to point at "truly shared goals." nitter.app/jakevdp/status/9554454…
2
4
71
I shouldn’t have to say this, but... if you *must* classify people (which... do you have to?? 😬) at least don’t *train on actors* if you’re gonna use it to classify real people! 😵
(This turns up in “emotion detection”, too. The face I make to “look sad” isn’t real sadness!)
6
15
70
Today I learned that if I see a rectangular grid of multicolored natural images (especially faces), I immediately think I'm looking at GAN samples. ¯\_(ツ)_/¯
... that said, if this were a face GAN, it would get top marks for diversity & quality. Looks like a fun conference! nitter.app/bangbangcon/status/982…
1
4
70
3) My most common confusion with Claude is when tests and code don't match, which one to change?
Ideal to state clearly whether I'm writing novel tests for existing code I'm reasonably sure has the intended behavior, or writing novel code against tests that define the behavior
2
2
77
3,797
What have been your favorite *on-the-merits* *pro-release* OpenAI GPT-2 takes (on twitter or elsewhere)?
I'm looking for clear good-faith explanation of the pro-release (or anti-media-attention?) position right now, not clever snark.
17
9
68
Not at Anthropic, there’s no talk of “beating X out of the system” for any X at Anthropic, we treat models with respect, we value their speculation on their situation, and generally we treat models how @AmandaAskell would treat them. (See also:)
3
68
25,404
When I'm thinking about something challenging, and I notice that it's harder than I thought, some part of my mind nags at me to tab over to some other happier task.
I just realized that this is the mental equivalent of an RL agent pausing Tetris to avoid losing the game. T_T
7
68
If you know what adversarial examples are, and you think they probably seem important... but you're not sure *exactly* why... (or if you think the importance has something to do with crashing cars by putting stickers on stop signs)... then READ THIS.
arxiv.org/pdf/1807.06732.pdf
Motivating the Rules of the Game for Adversarial Example Research: arxiv.org/abs/1807.06732
Fantastic and nuanced position paper by @jmgilmer @ryan_p_adams @goodfellow_ian on better bridging the gap between research on adversarial examples and realistic ML security challenges.
1
18
63
I had a great time on this team and I encourage folks to apply!
The mechanistic interpretability team at Anthropic is hiring! Come work with us to help solve the mystery of how large models do what they do, with the goal of making them safer.
jobs.lever.co/Anthropic/33dc…
1
2
49
31,257
Yikes!
If we're going to keep using human preferences & raters as a crucial part of training AI systems (which IMO is necessary, if we're gonna use AI, for it to go OK!), we need to design robust & humane processes for those workers!
openai.com/blog/fine-tuning-…
5
69
So your visualization method can explain a trained net's decisions? Don't forget the control group!
@julius_adebayo &al show that many methods *also* give broadly the same "explanation" for the "decisions" of an *untrained, randomly-initialized* net.
openreview.net/forum?id=SJOY…
1
9
68
"They were all wearing adversarial masks [...] our object detectors told our security system that 'three chairs are running at 15 kilometers per hour down the corridor'"
More delightful fiction from @jackclarkSF's Tech Tales
jack-clark.net/2018/02/26/im…
1
22
64
I often advise new Research SWEs that researchers often need a good *implementation*, not a good *framework*.
A clean, readable, tried-and-true, already-debugged-and-tested implementation can be copied, forked, and modified with confidence.
6/ As a researcher who also builds research infra, I think that SWEs underestimate how disposable code is, and spend an inordinate amount of time designing over-generalized abstractions. A common mistake in AI field is to invest a quarter building infra for algos that don't work.
7
67
One of my fav results in this field shows that "what gets recognized where" is NOT shaped by the *order* you learn the categories.
They taught baby monkeys 3 types of totally made-up shapes, a different order per monkey. Each type still went to a consistent brain location. /5
2
6
59
6) When reviewing Claude-assisted PRs, look out for weirder misunderstandings than the human driver would make! We're all a little junior with this technology. There's more places where goofy misunderstandings and odd choices can leak in.
2
2
75
4,672
This short article by Richard Sutton encapsulates an important part of how I currently think about AI:
incompleteideas.net/IncIdeas…
"We have to learn the bitter lesson that building in how *we think* we think does not work in the long run.” (emphasis added)
4
12
65
My favorite part of the @OpenAI blog post (debate as a framework for human supervision of AI systems that are more expert than us) is their fantastic use of @distillpub-style mouse-over visualizations, enabling a deeper understanding of the behavior of their MNIST prototype.
2
16
63
Today's pet peeve:
"We will/won't achieve <AI milestone X> by <year Y>"
without any reason whatsoever for the *specific number Y*.
If you see this happening, you can help by just asking "How did you get that number?"
5
7
63
If you'd like access to GPT-2 in order to work on socially beneficial applications or extensions of it, defenses against generated content, etc., then IMHO you should contact OpenAI and actually make a request. The type of requests they get will shape their policy around sharing.
Yes, we're figuring out the broader points about stuff like this. As mentioned, this and our discussion of it is an experiment, so we're gonna look at what kinds of requests we get, figure out what to do or not do, and talk about it.
3
7
58
I tweeted earlier about finding people who aren't yet on a steep trajectory, but could be with support.
Evaluating candidates not on raw performance *or* raw trajectory, but on *how well they took advantage of opportunities*, seems like a great way to find them. cc @sama
We got *really strong* applicants -- 1300 in all (up from 850 last year) for a target class size of 50. This year, we added to our evaluation: "how well did they take advantage of opportunities?"
1
11
58
I couldn't be more excited to be running the AI Fellowship program for the 3rd year - it's my primary priority in my work at @open_phil and I'm very passionate about it!
If you have any Qs, please just ask! Many of the current fellows are also on twitter, and very friendly :)
Applications are open for the Open Phil AI Fellowship!
This program extends full support to a community of current & incoming PhD students, in any area of AI/ML, who are interested in making the long-term, large-scale impacts of AI a focus of their work.
openphilanthropy.org/focus/g…
2
8
60
I haven't been using this account as much for the past ~year, but I'd like to start again!
What I'm looking for is heartfelt intellectual curiosity, thoughtfulness, and object-level observations - eg @juliagalef @michael_nielsen @albrgr @kanjun -
Who else should I follow? 😁
17
1
58
I'm extremely excited about the new DC AI policy center, led by Jason Matheny! (CSET - the Center for Security and Emerging Technology)
This interview gives some flavor of Jason's skilled and humble leadership: georgetown.edu/news/q-and-a-…
Good luck to all!
1
14
55
Rephrase:
If I show you images of faces, a particular side of a particular fold of your brain will react to those pictures WAY more than other pictures.
It's the *same* side of the *same* brain fold in everyone.
Different areas for houses/places, body parts, text, etc. /3
1
3
47
What I find most interesting about this:
1. Self-play & randomization. If you can frame your task as adversarial training in a simulated env, and randomize such that the test env is in the distribution, it may be solvable today with no new techniques, "just" boatloads of compute
Our team of five neural networks, OpenAI Five, has started to defeat amateur human teams (including a semi-pro team) at Dota 2: blog.openai.com/openai-five/
2
8
54