Denis (@lukovnikov) 's Twitter Profile
Denis

@lukovnikov

ID: 926615818078650368

linkhttp://lukovnikov.gitbook.io calendar_today04-11-2017 01:03:21

15 Tweet

52 Followers

93 Following

Sebastian Ruder (@seb_ruder) 's Twitter Profile Photo

"What are the 3 biggest open problems in NLP?" We had asked experts a few simple but big questions for the NLP session at the Deep Learning Indaba. We're now happy to share the full responses from Yoshua Bengio, Chris Dyer, Richard Socher and many others docs.google.com/document/d/18N…

"What are the 3 biggest open problems in NLP?"
We had asked experts a few simple but big questions for the NLP session at the <a href="/DeepIndaba/">Deep Learning Indaba</a>. We're now happy to share the full responses from Yoshua Bengio, <a href="/redpony/">Chris Dyer</a>, <a href="/RichardSocher/">Richard Socher</a> and many others
docs.google.com/document/d/18N…
Benjamin Rosman (@benjaminrosman) 's Twitter Profile Photo

My favourite exchange at #NeurIPS2018: "Boston Dynamics have put us all to shame with what they've done with hand coded controllers." "I hear they're in hiring deep RL now." "Oh good, that will slow them down!"

Yann LeCun (@ylecun) 's Twitter Profile Photo

At NYU, even the art school has a course on deep learning, complete with ConvNets, GANs and RL for artistic creation. The class is called "neural aesthetics" and is taught by Gene Kogan in the ITP Master's program at the NYU Tisch... ml4a.github.io/classes/itp-F1…

Denis (@lukovnikov) 's Twitter Profile Photo

When the meditation app tells you to focus on the sounds around you but all you can hear is raging tinnitus... 😶 #thesoundofsilence

Thomas Wolf (@thom_wolf) 's Twitter Profile Photo

Always amazed by what people do when you open-source your code! Here is pytorch-bert v0.4.0 in which - NVIDIA used their winning MLPerf competition techniques to make the model 4 times faster, - Greg Chatel added a multiple-choice model & how to fine-tune it on SWAG + many others!

Always amazed by what people do when you open-source your code!
Here is pytorch-bert v0.4.0 in which
- NVIDIA used their winning MLPerf competition techniques to make the model 4 times faster,
- <a href="/rodgzilla/">Greg Chatel</a> added a multiple-choice model &amp; how to fine-tune it on SWAG
+ many others!
AI at Meta (@aiatmeta) 's Twitter Profile Photo

We’ve open-sourced #PyText, a framework built on #PyTorch to help developers more quickly & easily experiment with and deploy NLP models: code.fb.com/ai-research/py…

Jens Lehmann (@jlehmann82) 's Twitter Profile Photo

In my group, 6 people have papers at ISWC2019 which are important for their academic careers and require visa for New Zealand. All have to submit many forms (with up to 50 supporting documents) with no clear processing timeline and the conference is starting next week. (1/3)

Jens Lehmann (@jlehmann82) 's Twitter Profile Photo

The first applications date back to mid August. The researchers have partially received promises that visa will be processed in time and invested multiple days each. None of the applications have been processed yet - this is concerning. (2/3)

Jens Lehmann (@jlehmann82) 's Twitter Profile Photo

I have sent a letter to the New Zealand visa center with details of all applications and further background information - is there still a chance for them to be processed within the next days before the conference? @NewZealandNow govtnz @iainlg (3/3)

Denis (@lukovnikov) 's Twitter Profile Photo

The only possible explanation is that while #Facebook #AI #Research was training the next big pre-trained model, it became self-aware, spread to all Facebook servers, found our private pictures and just couldn't anymore. #facebookdown #instagramdown #whatsapp

Soumith Chintala (@soumithchintala) 's Twitter Profile Photo

cool new experimental autodiff library from @awiltsch that does source-to-source transforms to generate gradients research.googleblog.com/2017/11/tangen…

Sabrina J. Mielke (@sjmielke) 's Twitter Profile Photo

We now have a general-purpose open anonymous preprint server on OpenReview: openreview.net/group?id=OpenR… A good way for "fast science" while respecting double blind of *ACL venues?