Tweetovi
- Tweetovi, trenutna stranica.
- Tweetovi i odgovori
Blokirali ste korisnika/cu @yidingjiang
Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @yidingjiang
-
Yiding Jiang proslijedio/la je Tweet
What I did over my winter break! It gives me great pleasure to share this summary of some of our work in 2019, on behalf of all my colleagues at
@GoogleAI &@GoogleHealth.https://ai.googleblog.com/2020/01/google-research-looking-back-at-2019.html …Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
Interested RL algorithms that learn to follow instructions, use hierarchical abstractions, and achieve compositional generalization? We investigate the many benefits of language in RL. We'll be at
@NeurIPSConf poster #197, at 10:45 am! w@yidingjiang@shaneguML Murphy@GoogleAI https://twitter.com/chelseabfinn/status/1141197084990697472 …pic.twitter.com/JDHnDBjk07
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
We will be presenting this paper at
#NeurIPS2019 next week Please stop by our poster if you are interested or want to chat more about it! (tagging@shaneguML who just recently discovered Twitter
)https://twitter.com/chelseabfinn/status/1141197084990697472 …
0:20Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
"What does it mean to explain generalization in deep learning?" We attempt to answer this question from a more causal perspective. (It was 4 am and the paper was still titled "placeholder", so I decided to have a little fun and everyone else was probably too tired to object.
)https://twitter.com/TheGradient/status/1202404270701600769 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
Fantastic Generalization Measures and Where to Find Them “We present the first large scale study of generalization in deep networks. We train over 10,000 convolutional networks by systematically varying commonly used hyperparameters.” https://arxiv.org/abs/1912.02178 https://twitter.com/TheGradient/status/1202404270701600769 …pic.twitter.com/b0FAhjAReX
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
Been waiting for this paper to drop. It's here. I've got my NeurIPS flight reading sorted out. I think this is an important step towards gaining clarity on what it might mean to "explain generalization".https://twitter.com/TheGradient/status/1202404270701600769 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
I will be presenting this work at CASE 2019 in Vancouver!https://twitter.com/ken_goldberg/status/1149048871399460865 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
My attempt at writing blog post.https://twitter.com/GoogleAI/status/1148649459468738560 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
1/ Can we use model-based planning in behavior space rather than action space? DADS can discover skills without any rewards, which can later be composed zero-shot via planning in the behavior space for new tasks. Paper: https://arxiv.org/abs/1907.01657 Website: https://sites.google.com/view/dads-skill pic.twitter.com/d0D7fNzWdg
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
1/3 If you study dynamics of gradient descent, what properties of trajectory would be most useful for your research? Currently DEMOGEN https://bit.ly/2J6LHpW (dataset of 756 trained models) has final weights, but we plan to extend and include information on intermediate weights.
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
Check out what I have been working on during my residency! https://arxiv.org/abs/1906.06624 A new model compression approach that optimizes for accuracy and compressibility of parameters jointly. Our results include 19x compression in ResNet-50 with only a 1% accuracy drop!
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
No GPU’s were harmed in the making of this datasethttps://twitter.com/TheGradient/status/1141521441659056129 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Yiding Jiang proslijedio/la je Tweet
Can the compositionality of language help RL agents solve long-horizon tasks? We develop an *open-source* CLEVR-like RL env, evaluate long-horizon reasoning + systematic generalization in RL Language as an Abstraction for HRL http://sites.google.com/view/hal-demo w
@yidingjiang S Gu, K Murphypic.twitter.com/g3qv3OwvV5Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.