The Ekho

@wilderrodrigues

Sr. Data Scientist & Deep Learning practitioner; creator of the Keras; Ambassador; IBM Watson AI XPRIZE contestant; Committer & PMC

Vleuten, Netherlands
Joined April 2009
584 Photos and videos Photos and videos

Tweets

You blocked @wilderrodrigues

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @wilderrodrigues

  1. Retweeted
    Sep 4

    We've used reinforcement learning from human feedback to train language models for summarization. The resulting models produce better summaries than 10x larger models trained only with supervised learning:

    25 replies 411 retweets 1,854 likes
    Undo
  2. Retweeted
    Sep 4

    🔥 Introducing... the ML Reproducibility Challenge 2020! The 4th annual edition now expands to cover papers from 7 major ML conferences: NeurIPS, EMNLP, ACL, ICML, ICLR, CVPR and ECCV. Find more here at Papers With Code:

    224 retweets 736 likes
    Undo
  3. Retweeted
    Sep 3

    Very excited about the 2020 Deep RL Workshop! Amazing line-up of invited speakers: Matt Botvinick Karen Liu Anusha Nagabandi Peter Stone Paper submission deadline: Oct 5, 2020

    2 replies 50 retweets 250 likes
    Show this thread
    Show this thread
    Undo
  4. Retweeted
    Sep 4

    Hydra 1.0 is out!

    Hydra 1.0 is out! - Config type safety via Structured Configs - More powerful command line - New plugins enabling remote launching and hyper parameter optimization And so much more. Check the blog post:
    8 retweets 42 likes
    Undo
  5. Aug 20

    Taking the rest of my holidays to recycle my Big-O, Data Structures and Algorithms knowledge. Just using frameworks makes one very rusty.

    1 like
    Undo
  6. Retweeted
    Aug 11

    New code example on , created by Amy: pneumonia x-ray slides classification, using TPUs

    25 retweets 110 likes
    Undo
  7. Aug 4

    Reading this article and finding out I have met the 3 professors whom featured on it, somehow made me feel good.

    Undo
  8. Retweeted
    Jul 28

    Happy to share EfficientDet-D7x: new SOTA 55.1 COCO AP at 153ms latency (10x faster than other multi-scale models). TF2+keras supported! Arxiv: github: paperwithcode:

    and
    7 replies 107 retweets 465 likes
    Show this thread
    Show this thread
    Undo
  9. Jul 24

    That's my way of telling a "short" story to my team. lol

    1 retweet
    Undo
  10. Retweeted
    Jul 23

    Congrats to the awesome Sepp Hochreiter for the well-deserved 2021 IEEE Neural Networks Pioneer Award! It was my great honor to be Sepp's nominator.

    12 replies 73 retweets 805 likes
    Undo
  11. Retweeted
    Jul 21

    If you can't understand a paper, don't be discouraged The current academic style *must* change: - arxiv has no page limits to drive density - ideas are more influential if more ppl can read them - true masters can explain it like I'm 5 Humble communication > self flagellation

    47 replies 214 retweets 1,650 likes
    Undo
  12. Jul 19

    How do you call your projects? I give mine related names. For instance, an object detection project I’m working on is called Cyclops, the training module. The inference module is called Cerebro; the CI/CD pipeline is Storm and the utility library is Forge.

    Undo
  13. Jul 18

    After 7 years of being a committer and PMC member of , I finally decided to start using my email for private matters. :)

    Undo
  14. Retweeted
    Jul 14

    Keras R-CNN, library for cell detection. Identifying malaria parasite development is often manual and subjective. We make it easier to apply object detection algorithms to this and other biological images. w/ et al.

    5 replies 42 retweets 136 likes
    Show this thread
    Show this thread
    Undo
  15. Retweeted
    Jul 12

    The tutorial on Bayesian Deep Learning and Probabilistic Model Construction is tomorrow! July 13, 11 am - 2 pm and 9 pm - 12 am EDT See you there!

    6 replies 143 retweets 697 likes
    Undo
  16. Retweeted
    Jul 12

    Neat package: "DeepXi is used for speech enhancement, noise estimation, for mask estimation, and as a front-end for robust ASR"

    3 replies 24 retweets 127 likes
    Undo
  17. Retweeted
    Jul 10

    🔥 Our new autocomplete gives you 𝙛𝙖𝙨𝙩 access to 25k papers with code, 2.8k leaderboards, 1.6k tasks, 760 methods. Researchers can add results & methods for papers to enhance their discoverability (). We'll be featuring the best additions next week!

    2 replies 88 retweets 407 likes
    Undo
  18. Retweeted
    Jul 9

    Want to train your Keras models on a beefy machine on GCP (or just many machines)? With TensorFlow Cloud, you can just add a single line to your script or notebook and get it running -- no config work necessary past the initial setup. Try it:

    18 replies 151 retweets 732 likes
    Show this thread
    Show this thread
    Undo
  19. Retweeted
    Jul 8

    We just open-sourced differentiable SDE solvers in PyTorch: Now you can put stochastic differential equations in your deep learning models, and neural nets in your SDEs! Credit to .

    23 replies 536 retweets 2,269 likes
    Show this thread
    Show this thread
    Undo
  20. Retweeted
    Jun 18

    A non-linearity that works much better than ReLUs. The work described in this video might also be relevant to understanding grid cells.

    43 replies 756 retweets 2,908 likes
    Undo

@wilderrodrigues hasn't Tweeted yet.

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·