Tweetovi
- Tweetovi, trenutna stranica.
- Tweetovi i odgovori
- Medijski sadržaj
Blokirali ste korisnika/cu @dpkingma
Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @dpkingma
-
Prikvačeni tweet
"Variational Inference and Deep Learning: A New Synthesis", written by yours truly, is now available for D/L here: https://goo.gl/6aGYZ1 .pic.twitter.com/ei9DJnvKFO
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
Organizers of data science and machine learning conferences (NeurIPS, ICML, AISTATS, ICLR, UAI, ...): Allow remote paper & poster presentations at conferences - Sign the Petition! http://chng.it/tTmXNsYB via
@ChangePrikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
The next decades are going to be very interesting as machine learning for robotics *really* starts to positively impact our economy. Congrats to
@pabbeel@rocky_duan@peterxichen for this launch, and robotics friends@OpenAI,@GoogleAI,@berkeley_ai for pushing the envelope.https://twitter.com/CadeMetz/status/1222511484384579586 …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Another great result demonstrating that VAEs (deep learning + amortized variational inference) make a lot of sense for data compression. Its loss function directly maximizes compressibility, and the resulting codec is fully parallelizable.https://twitter.com/TacoCohen/status/1221743084834381826 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Our paper "Variational Autoencoders and Nonlinear ICA: A Unifying Framework" has been accepted to AISTATS'20. With
@ilkhem, Ricardo Pio Monti and Aapo Hyvarinen (UCL). Surprisingly strong and general identifiability results, with rigorous proofs! https://arxiv.org/abs/1907.04809Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Clearly, symbolically defining the term "deep learning" is futile, and we can only accurately represent the concept with a deep neural net.
#connectionismHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Interested in scalable algorithms for learning energy-based models? Check out our preprint on Flow Contrastive Estimation (FCE): an elegant, adaptive extension of noise-contrastive estimation, with really promising results! Work led by brilliant
@RuiqiGao.https://twitter.com/RuiqiGao/status/1205190658144657408 …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
Our work "Flow Contrastive Estimation of Energy-Based Models" (
@RuiqiGao,@erik_nijkamp,@dpkingma, Z. Xu@iamandrewdai, YN Wu) at#NeurIPS2019: https://arxiv.org/abs/1912.00589v1 … We show (1) joint learning of EBM & Glow, (2) correction of over-dispersion, (3) competitive semi-super cls.pic.twitter.com/3S5BJl2Mgr
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
Flow Contrastive Estimation of Energy-Based Models https://deepai.org/publication/flow-contrastive-estimation-of-energy-based-models … by Ruiqi Gao et al. including
@dpkingma,@iamandrewdai#Probability#ProbabilityDistributionHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
I'd like to organize a "manifesto track" at AI conferences where speakers come up and present their controversial, slightly unhinged views on how to achieve AGI. Experimental evidence is optional. - Numenta HTM - LeCake - https://openreview.net/pdf?id=SyxQ_TEFwS … - Non-Axiomatic Reasoning System
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Joint work with
@hojonathanho and@pabbeel; paper at https://arxiv.org/abs/1905.06845Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Ever wondered, like me, if you can can turn hierarchical VAEs into a fully parallelizable lossless compression scheme? Wonder no more: new blogpost by Friso on "Bit-Swap", his recursive algorithm for doing just that: https://bair.berkeley.edu/blog/2019/09/19/bit-swap/ … Disclaimer: he's my sibling
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
https://yacht.bandcamp.com/album/chain-tripping …. Some pretty fresh tunes generated with the help of AI . I actually like these tunes a lot! Well crafted,
@YACHT and collaborators. More info:https://magenta.tensorflow.org/chain-trippingHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
... And thanks to
#Amsterdam for being such a dynamic and inspiring place for doing research! (Hint: apply for positions at, for example,@UvA_Amsterdam and the@GoogleAI lab in Amsterdam)Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
It's been 2 years since I finished my PhD thesis, but it won the ELLIS PhD award! https://ellis.eu/en/news/ellis-phd-award … All thanks to
@wellingmax for being possibly the world's best PhD advisor & thanks to great collaborators@TimSalimans,@DeepSpiker,@shakir_za, Jimmy Ba, etc.! [1/2]Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Someone is obviously really close to solving AGI: http://adamoptimizer.com
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Follow Apollo 11 in real time: https://apolloinrealtime.org/11/mobile/
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
Nonlinear ICA is known as unidentifiable with data only. They show that VAE conditioned with additional observation (e.g., time index, class label) can identify true components up to trivial transformation, showing principled disentanglement ability of VAE https://arxiv.org/abs/1907.04809
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
Dear friends, I’m extremely excited for our current research for “Latent Cinema” idea. This piece is narrated from ~300k collective photographic memories of Stockholm. Our GAN browser practically opens a camera into the space in the mind of a machine.pic.twitter.com/M6RhBibnYg
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Durk Kingma proslijedio/la je Tweet
Excited to share our latest work; it combines neural networks with stochastic processes by positing priors over the relational structure of the dataset, without relying on global latent variables!https://twitter.com/wellingmax/status/1141964606316961792 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
"In addition [...] we will give $50 million in grants through http://Google.org to nonprofits focused on the issues of homelessness and displacement. This builds on the $18 million in grants we’ve given to help address homelessness over the last five years [...]."
Prikaži ovu nitHvala. 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.