Twitter | Pretraživanje | |
Will Dabney
Research scientist at DeepMind. On the critical path to AGI. Also, a persistent optimist.
76
Tweetovi
36
Pratim
577
Osobe koje vas prate
Tweetovi
Will Dabney 2. velj
Odgovor korisniku/ci @hardmaru
Self-made billionaires
Reply Retweet Označi sa "sviđa mi se"
Will Dabney proslijedio/la je tweet
Quanta Magazine 30. sij
By teaching machines to understand our true desires, one scientist hopes to avoid the potentially disastrous consequences of having them do what we command.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 31. sij
Odgovor korisniku/ci @pyoudeyer
First, a quick mea culpa, I’m definitely guilty of this too. Solutions? As problematic as p-hacking is, it would already be an improvement if we were all reporting a reasonable statistical significance vs just eye-balling the learning curves (or worse).
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 29. sij
Our paper 'A distributional code for value in dopamine-based reinforcement learning' on the cover of ! Read it here: Shout out to the amazing artists/designers at who make this possible, while we get to focus on the research.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney proslijedio/la je tweet
Sherjil Ozair 28. sij
I haven't found a single person who has used jax and said they don't like it. I've been actively priming people to criticize it, but noone does. Instead they tell me how good it feels getting off of Tensorflow. Looking forward to jaxxing myself soon.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney proslijedio/la je tweet
Pablo Samuel Castro 16. sij
Hey everyone, I'm so excited to share my recent interview on Music & AI plus "A Geometric Perspective on Reinforcement Learning" with for the podcast. Check it out! via
Reply Retweet Označi sa "sviđa mi se"
Will Dabney proslijedio/la je tweet
Eric Topol 17. sij
The reciprocal inspiration of and neuroscience. A paper this week on the mechanism of reinforcement learning by and colleagues with an excellent explainer by
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 16. sij
Odgovor korisniku/ci @wwdabney
Thank everyone! You can also read the paper for free here:
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 15. sij
Odgovor korisniku/ci @marcgbellemare
And it all started (for me) almost exactly three years ago working with and Remi on distributional RL:
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 15. sij
Odgovor korisniku/ci @zebkDotCom
It has been an incredible collaboration with my co-authors, especially working with and Matt Botvinick. Also incredibly grateful to Naoshige Uchida and Clara Starkweather from Harvard, as well as Remi Munos and Demis Hassabis for their work and constant endurance! 2/
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 15. sij
When neuroscience and AI researchers get to chatting, cool stuff happens! My first, and I hope not last, trip into neuroscience has been published in Nature. 1/
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 8. sij
Almost all of these (IMO) apply equally well to research. I most disagree with the “short 1:1, long group meetings” one, but do other research people think most of these apply to them?
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 5. sij
Odgovor korisniku/ci @pcastr
I’ve been playing with Hugo in gitlab with netlify. It’s painfully simple, but yes markdown has a little more limited formatting. I think git push for publishing a post is too beautiful to pass up.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 24. pro
Odgovor korisniku/ci @hardmaru
Building upon past work is excessively difficult without open source. I wasted tons of time reimplementing from papers in grad school. With today’s frantic pace, this time investment feels untenable. That said, I don’t think reviewer nullification is the answer.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 21. pro
Odgovor korisniku/ci @dwf @VladMnih
Work is also with Andre Barreto, , Tom Van de Wiele, and !
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 21. pro
Happy to have worked with on research combining behavioural mutual information and successor features, which has been accepted for oral presentation at ICLR. Favorite part: clean answer to where to get the “features” for successor features.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 15. pro
Odgovor korisniku/ci @doomie @JeffDean
Absolutely agree, a beautiful city and fantastic venue!
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 8. pro
So fun how non-stationary our perception is. It’s not hard to get the direction cued onto any physical change. Opening/closing hand, blinking, you can even pretend to spin it this way and that with your thumb and it will switch.
Reply Retweet Označi sa "sviđa mi se"
Will Dabney proslijedio/la je tweet
Anna Harutyunyan 5. pro
Really excited for next week and to present our spotlight on credit assignment :) tl;dr We can rewrite value functions in terms of a hindsight quantity that explicitly captures credit assignment and get a whole new family of RL algs! 🥳
Reply Retweet Označi sa "sviđa mi se"
Will Dabney 4. pro
Let’s just hope Fox News doesn’t run this, or he might just declare war on all our allies.
Reply Retweet Označi sa "sviđa mi se"