Tweetovi

	Oriol Vinyals @OriolVinyalsML	2. velj
	Odgovor korisniku/ci @svlevine I think what we as a community are missing / still iterating over is how and when to publish. Arxiv isn't always the best medium. Workshop or simple tech reports are also great instead of full blown conference or journal submissions.
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	1. velj
	Odgovor korisniku/ci @FelixHill84 @zacharylipton You have to balance quality and quantity, both as a student, faculty, well or not so well established researcher. *Precisely* due to the current climate, something of much higher quality will have higher chances of standing given the sea of papers that get published daily.
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	1. velj
	Less is more. twitter.com/zacharylipton/…
	Prikaži podatke ·

	Oriol Vinyals @OriolVinyalsML	28. sij
	seq2seq still delivers : ) Incredible how far we've gotten in ~5 years of progress in neural conversational models, with relatively small changes. More exciting is that there's still LOTS to be done! Paper: arxiv.org/abs/2001.09977 Blog: ai.googleblog.com/2020/01/toward… pic.twitter.com/HUfbgtOCZW
	Prikaz fotografije ·

	Oriol Vinyals @OriolVinyalsML	15. sij
	Odgovor korisniku/ci @AssistedEvolve I've used this in several lectures, but perhaps the NeurIPS 2017 tutorial is a good one: goo.gl/b6QBBh (context: twitter.com/OriolVinyalsML…)
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	15. sij
	Odgovor korisniku/ci @william_woof We tried both and found object based to be superior earlier in the project. You can see an interesting work on using an interesting mixture here: arxiv.org/abs/1806.01830
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	15. sij
	Odgovor korisniku/ci @AndrewTouchet Sorry for the delay. pic.twitter.com/b3me7omkGM
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	15. sij
	Odgovor korisniku/ci @__thomaswood Sure! It's a big file :)
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	15. sij
	Although neural networks usually require massive datasets to do impressive things, for me the highlight of #AlphaFold is the fact that it achieved state-of-the-art using only 30K training examples. Code: github.com/deepmind/deepm… Paper: nature.com/articles/s4158… pic.twitter.com/k533FyqD6q
	Prikaz fotografije ·

	Oriol Vinyals @OriolVinyalsML	1. sij
	Odgovor korisniku/ci @AnimaAnandkumar @sapinker @celestekidd Late to the party, but let me add that I feel incredibly proud of our community every time I say or write #neurips vs the old name. Thanks to those who fight!
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	1. sij
	Odgovor korisniku/ci @OriolVinyalsML The figure above is an (old) slide I usually show in lectures and talks. Mix and matching those components creates cool architectures such as the one used in #OpenAIFive or #AlphaStar. pic.twitter.com/6udLczZUXj
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	1. sij
	The "Deep Learning Toolbox" has greatly expanded in the last decade thanks to our wonderful research community. Also, important progress has been made to make our community more inclusive and less toxic. Still, there's LOTS to do, and I plan to keep focusing on advancing both. pic.twitter.com/C5r4wDGSN9
	Prikaz fotografije ·

	Oriol Vinyals @OriolVinyalsML	30. pro
	Odgovor korisniku/ci @Abebab Please be yourself and stand for what you believe in and feel passionate about : )
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	21. pro
	Odgovor korisniku/ci @CVC_UAB @jordiPuignero i 6 ostali Thanks for the invitation! Such a pleasure to be there with such good company, and in one of the most beautiful buildings in Barcelona.
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	18. pro
	Odgovor korisniku/ci @hannawallach Nice! The BUDS@NeurIPS meetup was also highly recommended, and a high bandwidth version of what you did : )
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	18. pro
	Odgovor korisniku/ci @vpacela @WiMLworkshop i 3 ostali Nice poster and cool work! Let me know if what we discussed works -- to input rather than output the energy to the discriminator.
	Prikaz razgovora ·

	Oriol Vinyals @OriolVinyalsML	14. pro
	Odgovor korisniku/ci @RobertTLange Nice! If you keep doing these you should train a talk2notes model : )
	Prikaz razgovora ·

	Oriol Vinyals proslijedio/la je tweet
	Robert Lange @RobertTLange	14. pro
	Many gems in @OriolVinyalsML Deep RL workshop talk at #NeurIPS2019 on AlphaStar. Including scatter connections, imitation-based regularization, the league & the unique problem decomposition. pic.twitter.com/YZIKYKrTW5
	Prikaz fotografije ·

	Oriol Vinyals @OriolVinyalsML	14. pro
	I have loved relativity theory and black holes my whole life. Super exciting to have been at #KatieBouman talk at #NeurIPS2019. Great example of the research diversity in our field! pic.twitter.com/u1PwB86469
	Prikaz fotografije ·

	Oriol Vinyals @OriolVinyalsML	14. pro
	Odgovor korisniku/ci @almuttaqiin @OpenAI youtube.com/watch?v=Oxseex…
	Prikaz razgovora ·