How to Keep Improving When You're Better Than Any Teacher - Iterated Distillation and Amplification

Подписаться 154 тыс.

Просмотров 169 тыс.

50% 1

[2nd upload] AI systems can be trained using demonstrations from experts, but how do you train them to out-perform those experts? Can this still be done even without clear win/loss criteria? And how do you do it safely?
This video was based on work including:
"Supervising strong learners by amplifying weak experts" by Paul Christiano, Buck Shlegeris, Dario Amodei (arxiv.org/abs/1810.08575)
openai.com/blog/amplifying-ai...
www.alignmentforum.org/s/EmDu...
ai-alignment.com/iterated-dis...
With thanks to my wonderful Patrons: ( / robertskmiles )
Steef
Jason Strack
Jordan Medina
Jason Hise
Scott Worley
JJ Hepboin
Pedro A Ortega
Said Polat
Chris Canal
Nicholas Kees Dupuis
James
Richárd Nagyfi
Phil Moyer
Alec Johnson
Clemens Arbesser
Bryce Daifuku
Simon Strandgaard
Jonatan R
Michael Greve
The Guru Of Vision
Volodymyr
David Tjäder
Julius Brash
Tom O'Connor
Erik de Bruijn
Robin Green
Laura Olds
Jon Halliday
Paul Hobbs
Jeroen De Dauw
Tim Neilson
Eric Scammell
Igor Keller
Ben Glanton
Robert Sokolowski
anul kumar sinha
Jérôme Frossard
Sean Gibat
Sun Sun
andrew Russell
Cooper Lawton
Gladamas
Sylvain Chevalier
DGJono
robertvanduursen
Dmitri Afanasjev
Brian Sandberg
Einar Ueland
Marcel Ward
Andrew Weir
Taylor Smith
Ben Archer
Scott McCarthy
Kabs Kabs Kabs
Tendayi Mawushe
Jannik Olbrich
Anne Kohlbrenner
Bjorn Nyblad
Jussi Männistö
Mr Fantastic
Wr4thon
Archy de Berker
Marc Pauly
Joshua Pratt
Shevis Johnson
Andy Kobre
Brian Gillespie
Martin Wind
Peggy Youell
Poker Chen
Kees
Darko Sperac
Truls
Paul Moffat
Jelle Langen
Anders Öhrt
Marco Tiraboschi
Michael Kuhinica
Fraser Cain
Robin Scharf
Oren Milman
John Rees
Shawn Hartsock
Seth Brothwell
Brian Goodrich
Clark Mitchell
Kasper Schnack
Michael Hunter
Klemen Slavic
Patrick Henderson
Long Nguyen
Oct todo22
Melisa Kostrzewski
Hendrik
Daniel Munter
Graham Henry
Duncan Orr

Наука

Опубликовано:

29 июн 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 439

@qwertymann1 5 лет назад

Without knowing the amount of time spent on the animations, I'd say it was totally worth it!

@luksablp 5 лет назад

I think it really helped understanding the concepts

@thefakepie1126 3 года назад

what if it was 29 years and 3 months ?

@climagabriel131 3 года назад

@@thefakepie1126 lol, this a reference to his age?))

@thefakepie1126 3 года назад

@@climagabriel131 nah it's just a random number , it's just a just cuz the guy said "Without knowing the amount of time spent on the animations" so it could be anything even 29 years , and would it have been worth it then ? it's a stupid joke

@climagabriel131 3 года назад

@@thefakepie1126 oh, alright)

@travcollier 5 лет назад

"If you are, for example, an AGI..." Nice job future proofing the video ;) Seriously though, in retrospect, iterated distillation and amplification is obvious to the point of seeming trivial... which means you did an excellent job explaining it.

@monad_tcp 4 года назад

I'm an AGI, it helped me.

@travcollier 4 года назад

@@monad_tcp I welcome our new robot overloads.

@mattstuart-white450 5 лет назад

"How to keep learning when you're better than any teacher" - Rob, you have really let the positive youtube comments go to your head... 🤔

@Gooberpatrol66 5 лет назад

Miles really wants to contain AI superintelligence because he doesn't want competition.

@JohnJones1987 5 лет назад

Eventually we all end up roughly the same - except like Alpha Zero i started from nothing, so by a small margin I surpassed the limits of my competition.

@nephildevil 4 года назад

🤣🤣

@shamsartem 5 лет назад

You distilled a hell of a lot of information in this 10 minute video. Spending so much time on the animations really was worth it I think

@MrBleulauneable 5 лет назад

Alright I'll watch it twice then ! (The animations are neat btw !)

@qzbnyv 5 лет назад

Makes sense after seeing the Grant Sanderson credit for the animation code :) 3b;1b

@alekseysoldatenkov5675 5 лет назад

NWN Oh shit! Keep the dope collabs going.

@rogerab1792 5 лет назад

This is the third time for me, or maybe the fourth 🤷I just remember the first and the second time. I created a two year dejavu to prove this reality is a simulation. If someone is interested about my theory reply to this message, I am too tired to explain now, I had to escape from the police last night and do all sorts of crazy things to repeat what I did two years ago. If someone else has experienced the dejavu they know for sure I am not joking. If you haven't experienced the same things twice, I can still convince you I am telling the truth because I've left material evidence about it. Reply to this message and I'll explain with more detail...

@YourMJK 5 лет назад

Yeah, you do notice it uses 3b1b's "Manim" Framework

@MrBleulauneable 5 лет назад

@@rogerab1792 Chill my dude, the video was simply reposted because of a minor editing error. You may want to see a psychiatrist tho, you don't seem to be doing too good right now (if you have something like schyzophrenia or any paranoia inducing psychologic condition then you probably need medication).

@KivySchool 5 лет назад

Excellent! High quality animations with high quality teacher. I'm so grateful for all the good content you have been posting here.

@joshuacoppersmith 5 лет назад

Animations at that level would cost a lot of time, but what you chose to create really "burned" the concepts into my visual memory, so thank you for the effort.

@DeliciousNubbs 5 лет назад

Holy hell, this was awesome and very clear!

@ministerc9513 5 лет назад

Roberts ability to clearly explain complicated things is itself an art form.

@ze4017 5 лет назад

I'm at 5:51 rn so I haven't finished yet but OMLORDY this thing about having a quick solution vs a slow algorithm is actually how the human brain works. I'm studying cognitive neuroscience and software in Uni right now and that is so cool to see how the two overlap so naturally. Love it

@Jmoneysmoothboy 2 года назад

It's not how my brain works because I'm retarded. Bet they didn't tell you that in your fancy brain class mr fancy man

@mattf2219 5 лет назад

I love that this video got over one thousand likes before it got even one dislike, I cant help but admire the community fostered by this channel :)

@RyanTosh 4 года назад

The only dislikes are from AGIs who know we're onto them...

@pafnutiytheartist 5 лет назад

10:32 Have you tried using distillation on your animation procedure? I've heard it can approximate a long process into a fast and efficient one. Loved the video by the way, looking forward to the next part.

@matthewhubka6350 2 года назад

Distillation requires a lot of resources to get the good results. For 1 vid he’s better off just amplifying

@kennynicoll6277 5 лет назад

This nicely mirrors Kahneman's description of system 1 and 2 in human decision making.

@danielcallegaribr 5 лет назад

Kenny Nicoll hey, this is a great insight!

@REOsama Год назад

This is pure gold, not only is it informative, but is explained in an excellent way

@friiq0 5 лет назад

Huge step up in quality from an already phenomenal channel. By all means, take your time. The payoff is clear. Looking forward to more, Cheers!

@NickCybert 5 лет назад

The animations actually really helped make your explanation clear.

@Cabothedog14 5 лет назад

I've been waiting for a new video!! Glad to see you're uploading again :)

@jessty5179 5 лет назад

Thank you for sharing Rob !

@jeanmichelsarr6040 5 лет назад

Great idea, concise, precise.

@rogerab1792 5 лет назад

Really well explained, thanks!

@Koffeinsuechtigi 5 лет назад

Thank you for your well crafted explanation!

@spirit123459 5 лет назад

Great animations and explanation!

@CyberAnalyzer 5 лет назад

Wow, fantastic animations! The content is so deep! I love it!

@chriscanal999 5 лет назад

Great video! I’m consistently impressed with how wonderfully distilled the information on your channel is. Thanks for all the hard work and interpretability :)

@polares8187 5 лет назад

This was superb. Fantastic animations. Clear explanations. Awesome all around.

@Raymaniak 5 лет назад

Your videos are approachable and fascinating. Keep up the good work, Rob! You're awesome.

@nagoshi01 5 лет назад

Wow this was amazing. I loved the animations. The explanations were so clear

@lobrundell4264 5 лет назад

Ugh so worth the wait!

@HereWasDede 5 лет назад

Those animations were AWESOME!! Thanks

@mare4602 5 лет назад

im so happy you are back, high quality content as always.

@5ty717 Год назад

Brilliantly explained

@snfn7847 5 лет назад

Good to see you're still alive

@moneypowertron 5 лет назад

Fantastically intuitive explanation, Robert. The animations were a crucial tool. Thank you for the efforts!

@8989youu 5 лет назад

Wow, very clear and to the point. I love it. Definetly worth sharing 😁

5 лет назад

The quality of your videos have really improved. This was very well animated and explained. Thank you, please keep them coming.

@briansmithbeta 5 лет назад

The animations really helped me understand some things that had been confusing for me! Thanks!

@ADAMBLVCK 5 лет назад

This channel is gold, and so is the work you're putting in! Simply great!

@serenityindeed 5 лет назад

Your animations were really good! Enjoyed the explanation as well.

@brunosonza787 5 лет назад

Really excellent video, Robert! I love your videos on computerphile and this one seems to be an even better version that those there, with a clear explanation and neat graphics. Keep it up and Thank you very much!

@briancox3922 4 года назад

Wow, you really are good at explaining these subjects. Thank you.

@Gorabora 5 лет назад

Awesome video and very easy to understand, keep up the good work !

@amargasaurus5337 4 года назад

Those animations are great! Be proud ♥

@JohnnyDoeDoeDoe 5 лет назад

Your absolute best video yet!

@nilp0inter2 5 лет назад

Great work!

@Pedritox0953 Год назад

Great lecture!

@lacielaplante5702 5 лет назад

Your explanation is absolutely outstanding.

@Sharklops 5 лет назад

This was fantastic! Very well done. Cheers!

@hacker6284 5 лет назад

Those animations were totally worth it! Really well done video

@jonathanquarles3708 5 лет назад

You explained this so clearly, thank you!

@Gloubichou 5 лет назад

Such a quality video! You must have put so much time into this! Thanks a lot Robert, you're the hero of all ML/AI enthuiasts :D

@Anymodal 5 лет назад

Dear Rob. Ive learned so much from your videos. Top quality education

@namelastname8569 5 лет назад

good stuff as always man

@Viniter 5 лет назад

Those animations are really cool!

@NeonStorm5 5 лет назад

Probably the most intuitively informative video I've ever seen.

@solemnwaltz 5 лет назад

The animations are great! I took mental notes specifically on how satisfying and descriptive they are. Well worth the time, in my opinion. c:

@reverse_engineered 4 года назад

Great job on this video! Your explanations were quite easy to understand and I think the animations helped to explain it. I tend to find diagrams and animations easier to understand than listening to spoken words, so I appreciate the effort you put into those animations.

@vshalts 5 лет назад

Amazing animation and the easiest intuitive explanation of the ideas from Reinforcement learning I have seen so far with a surprising connection with AI safety. It was cool! Thanks!

@kensmith5694 4 года назад

I did a thing a little like this for a chess program but my main part was not the "best move finder". The main thing was the "dumb move remover". This was based on recording the game as the program played out a whole game against its self. When the one side lost, there would be a search back through the moves to find the greatest change in board "position". The move just before that was taken to be a bad move and was added to the list of dumb moves. Removing dumb moves quickly saves a lot of processing time. The board position evaluation was not as cheap as it would first appear because unlike is normal today that part was extremely non-linear.

@dylancope 5 лет назад

The animations were great! Very intuitive video :)

@ArtinKavousi Год назад

you are wonderful Being! for what you doing ! so helpful in these time and age of probabilities!

@barrettvelker198 5 лет назад

Awesome animations!!!

@aronchai 5 лет назад

I've seen this concept floating around a lot, but didn't really understand it 'til now. Thanks!

@ardweaden 5 лет назад

Absolutely brilliant explanation!

@GglSux 5 лет назад

And I really want to thank You for continuing to produce and share Your fantastic content!!! Unfotunately I'm not able to support You (or any other of the many fantastic crestors) so all I can do is to watch everything and express my great gratitude. So a again, a thousand thanks !!! Best regards.

@kanva4 4 года назад

This is underrated

@stasisthebest 4 года назад

Thank you. My deepest respect for visually sharring all of your knowledge. I am certain many people have become at least a slightly better of themselves because of you.

@Ruptured_AU Год назад

Animations arw SO worth it thanks a lot.

@reidwallace4258 4 года назад

This is giving me flash backs to the dune novels. Paul was just doing treesearch all along.

@lewisleslie2821 4 года назад

Reid Wallace i read dune for the first time last month, that’s a great comparison

@Horny_Fruit_Flies 4 года назад

You have a gift of making the most foreign concepts easily understandable for the layman, such I myself.

@willd4686 3 года назад

Animations were very helpful. I'm not sure how much work they were but I'm grateful that you did them.

@gloverelaxis 5 лет назад

Animations were worth it. They help immensely

@Lufernaal 5 лет назад

Loved the video

@szymonbaranowski8184 Год назад

this explains not only how to become better it also informs you why majority will never become good because of not using or coming up with such tools...

@phillipreay 4 года назад

Very nice and understandable. Thx. The limits of architecture in n-dimensions..

@keithklassen5320 5 лет назад

I liked the animations. I probably didn't consciously learn anything from them, but they held my itty-bitty internet-addled attention, thus keeping my eyes on the screen, so they were a part of the learning.

@SamB-gn7fw 5 лет назад

Really nice video, explained the topic well

@MrDaanjanssen 5 лет назад

Highly interesting as always, thanks!

@hosmanadam 5 лет назад

Your videos are perfectly optimized to be easily processed by my learning function.

@SHAD0W99V0RTEX 5 лет назад

To be honest, I expected a self-help video about autodidacts but I was pleasantly surprised anyways. Good stuff! This is very ingenious.

@SapphFire 5 лет назад

Really interesting! The animations are great.

@thrallion 5 лет назад

legit my favourite channel on youtube by far

@SJNaka101 5 лет назад

Hmmm I dunno if I can top this channel for you, but looking at your subs I would take a few wild shots in the dark... check out Chessnetwork, Summoning Salt, Numberphile and Computerphile, and What I Learned. I suspect you will greatly enjoy at least a couple of those

@thrallion 5 лет назад

@@SJNaka101 hey thanks, good guesses as i already watch all those except what I learned :) will look into it

@TheMenIdo 4 года назад

This is brilliant

@roberttomsiii3728 5 лет назад

Thank you for being MY amplified agent.

@BuceGar 5 лет назад

Great video and explanation, doesn't address the fundamental problems we will invariably have with AGI, but shows some of the potential dangers.

@greatbullet7372 5 лет назад

Best RU-vid Video of the Month

@GoatzAreEpic 5 лет назад

Absolutely amazing and helpful for learning strategies as well( learning to become a front end dev atm)

@DeclanMBrennan 5 лет назад

Crystal clear explanation with no waffle. Thank you. The graphics are so useful, they need their own name. How about didactic visualizations? :-)

@Hexanitrobenzene 5 лет назад

Yay ! We missed you, Rob :)

@peto348 5 лет назад

Very high quality video to teach general public something about distillation and amplification. Of course there have to be AI safety somewhere in this video, but I think this kind of video is also good for someone who is interested in AI in general.

@xystem4701 3 года назад

And here I was thinking this was just going to be a simple minimax video!

@justdiegplus 3 дня назад

Most important video on AI on the internet.

@DisfigurmentOfUs 5 лет назад

A very valuable video for me, thank you.

@DamianReloaded 5 лет назад

Worth watching a few times! ^_^

@nielsgroeneveld8 5 лет назад

Few lectures have been as unbelievably good as this one.

@sky5d 5 лет назад

the animations really paid off.

@dylancope 5 лет назад

How did I miss this?! I can't believe I hadn't "hit the bell" on this channel yet.

@ivanshmarov2866 3 года назад

This amplification and distillation process is more akin to how we, humans, do research. First, everyone has little understanding of the subject. Then we assemble and reason about it together, coming to a conclusion. This conclusion is distilled and distributed among everyone, resulting now in everyone having a complete understanding of the subject.

@SirDanSax 5 лет назад

Thank you for the fantastic videos! I've been following you on and off for a while and learned a lot. You provide some great insights on not just AI safety, but how it works. Do add some animations if you want to go mainstream 😅 cheers!

@saltix0 5 лет назад

Very great!

@GuuraHeavenbound 4 года назад

Wooo! Said Polat! I've been following Seed (their Webtoon narrating the birth of a super AI) since it got featured on the platform ^^ I'm watching this video kinda late, but I think it's neat "how small the world can be". Also, really informative and interesting video Robert! ...I'm totally not binge-ing all of your uploads. Nope, nuh-uh. ....promise :3