Ranjodh Singh @_ranjodh_singh

Deep Learning Researcher Joined August 2015

Tweets

81
Followers

49
Following

842
Likes

2K

PyTorch @PyTorch

5 years ago

Introducing VISSL (vissl.ai) - a library for reproducible, SOTA self-supervised learning for computer vision! Over 10 methods implemented, 60 pre-trained models, 15 benchmarks, and counting.

9 252 1K 0 157

Download Image

I am starting a blog series. I made sure that a layman should be able to understand and an experienced person gets the overall idea about the problem at hand share if you find relevant ritwikm.medium.com/data-leaks-by-…

0 1 6 0 0

GitHub @github

5 years ago

GitHub CLI 1.0 is here 🎉 Take GitHub to the command line and interact with repositories, issues, pull requests, releases, and more. ✓ Free and open source ✓ Available for macOS, Windows, Linux ✓ GitHub Enterprise Server supported Download it now: github.blog/2020-09-17-git…

74 3K 8K 0 315

Download Video

👩‍💻 Paige Bailey @DynamicWebPaige

5 years ago

✨📄 Are you intending to submit a publication for review at an upcoming academic conference? This deck from @MIT's Bill Freeman is an excellent guide: billf.mit.edu/sites/default/… It irks me that these practices are not universally taught in *all* undergraduate & graduate schools.

1 18 91 0 29

Download Image

Ranjodh Singh @_ranjodh_singh

5 years ago

How EIA 2020 will Damage India! | Dhruv Rathee youtu.be/tWCjhj035WY via @YouTube

0 1 0 0 0

Ranjodh Singh @_ranjodh_singh

5 years ago

New GitHub UI 😍🤩

0 0 1 0 0

OpenAI @OpenAI

5 years ago

Introducing Jukebox, a neural net that generates music, including rudimentary singing, as raw audio in a variety of genres and artist styles. We're releasing a tool for everyone to explore the generated samples, as well as the model and code: openai.com/blog/jukebox/

205 2K 8K 0 848

Download Video

Ranjodh Singh @_ranjodh_singh

5 years ago

🤣🤣😂😂

0 0 3 0 0

abhishek @abhi1thakur

6 years ago

PyTorch introduces native automatic mixed precision training. No need to compile apex anymore 😎 lnkd.in/dFHG7JS #pytorch #datascience #python #machinelearning #deeplearning lnkd.in/dSvKwH4

0 36 108 0 10

Douwe Kiela @douwekiela

6 years ago

Excited (in my 1st tweet ever!) to announce Adversarial NLI: a new large-scale benchmark dataset for NLU, and a challenge to the community. Great job by @EasonNie, together with @adinamwilliams @em_dinan @mohitban47 and @jaseweston. arxiv.org/abs/1910.14599

4 90 284 0 32

IEEE BigMM 2020 @IEEEBigMM

6 years ago

Registration link register.comp.nus.edu.sg/BigMM2019/ for @IEEEBigMM19 is working now. Early-bird registration for authors is August 1, 2019 and that for other attendees is August 4, 2019. Looking forward to seeing you at @NUSComputing, Singapore.

0 5 5 0 0

Quoc Le @quocleix

6 years ago

XLNet: a new pretraining method for NLP that significantly improves upon BERT on 20 tasks (e.g., SQuAD, GLUE, RACE) arxiv: arxiv.org/abs/1906.08237 github (code + pretrained models): github.com/zihangdai/xlnet with Zhilin Yang, @ZihangDai, Yiming Yang, Jaime Carbonell, @rsalakhu

19 676 2K 0 131

Download Image

Ranjodh Singh @_ranjodh_singh

6 years ago

poly-ai.com/blog/moving-fr…

0 0 2 0 0

Quoc Le @quocleix

6 years ago

EfficientNets: a family of more efficient & accurate image classification models. Found by architecture search and scaled up by one weird trick. Link: arxiv.org/abs/1905.11946 Github: bit.ly/30UojnC Blog: bit.ly/2JKY3qt

24 660 2K 0 201

Download Image

Ranjodh Singh @_ranjodh_singh

6 years ago

"Off-Policy Deep Reinforcement Learning by Bootstrapping the Covariate Shift" Charles et. al arxiv.org/abs/1901.09455

0 0 0 0 0

Ranjodh Singh @_ranjodh_singh

6 years ago

"Off-Policy Deep Reinforcement Learning without Exploration", Scott et. al arxiv.org/abs/1812.02900

0 0 0 0 0

Ranjodh Singh @_ranjodh_singh

6 years ago

Deep Gaussian Processes with Importance-Weighted Variational Inference arxiv.org/abs/1905.05435

0 0 0 0 0

Microsoft Research @MSFTResearch

6 years ago

Microsoft’s Space Partition Tree And Graph (SPTAG), now available on @github, takes advantage of vectors to search through billions of pieces of info in milliseconds: aka.ms/AA52hxp