FREE registration for NVIDIA GTC is now open! Excited to explore what’s next in AI, HPC, and more. nvidia.com/en-us/gtc/?nci… #GTC21

New book on the theory of deep learning coming up soon by Sho Yaida and Dan Roberts .

Learn all about self-supervised learning for vision with Ishan Misra! In this lecture, Ishan covers pretext invariant rep learning (PIRL), swapping assign. of views (SwAV), audiovisual discrimination (AVID + CMA), and Barlow Twins redundancy reduction. youtu.be/8L10w1KoOU8

There are now 100 (one hundred!) code examples at keras.io/examples, covering essential workflows across computer vision, NLP, generative learning, RL, etc. Created by the Keras community. Focused on readability/clarity, reusability, and best practices.

Pre-registration for BME courses starts today at 12:00. 200 places are available for students who want to study deep learning! Come and join us!

🙌 Our website is live! Check it out at softwareskills.eu. We are very happy to share this milestone with you. You will now be able to browse the articles, and discover our work ✅ #SoftwareSkills4EU #digitalskills

Congratulations to Peter Almasi, who achieved first place in the HTE MSc Thesis Grant with his deep reinforcement learning work for Duckietown. hte.hu/hte-diplomater…

🎯 There is a big gap between what skills the market needs and what the EU’s workforce has to offer — and it is nowhere more apparent than in software development. Our upcoming #softwareskills Needs Analysis report will shed 💡 on this bit.ly/3CCiFbW #SoftwareSkills4EU

There's no evidence that SGD plays a fundamental role in generalization. With totally deterministic full-batch gradient descent, Resnet18 still gets >95% accuracy on CIFAR10. With data augmentation, full-batch Resnet152 gets 96.76%. arxiv.org/pdf/2109.14119…

We started the deep learning class this semester in person with about 80 attendees. Looking forward to the great solutions submitted as project work!

I will hold the next NVIDIA Deep Learning Institute full-day workshop on Fundamentals of Deep Learning on 12th November 2021, online. Academia and university students and staff only, registration is free: eventbrite.com/e/nvidia-dli-w…

András Béres, one of our outstanding MSc students, has written an excellent description on "Data-efficient GANs with Adaptive Discriminator Augmentation", it is published on the Keras Code examples: keras.io/examples/gener… Congratulations! #deeplearning #keras #tensorflow #bme

The #software professionals of the future are more than just software experts. They have a full set of hard, soft, and business #skills. And that's how Europe can become resilient and competitive. Find out more: bit.ly/3p2gMlp #SoftwareSkills4EU #futureofwork

I will hold the next NVIDIA Deep Learning Institute full-day workshop on Building Transformer-based NLP Applications on 13 December 2021, online. Academia and university students and staff only, registration is free: bit.ly/dli2021nlp

1 year has passed since we first met as a consortium🎉 We’ve touched base monthly & despite the challenges of #remote work, we delivered high-quality results like our report on Europe’s most needed #software roles and #skills. You can read it here: bit.ly/3BLtus7

Last week we held an NVIDIA Deep Learning Institute full-day workshop on Building Transformer-Based Natural Language Processing Applications. We had over 70 participants from the EMEA region. Thanks for the organizers, TAs and the participants! See you in the next workshop!

Compilation thread of various Keras tips ⬇️⬇️⬇️

NVIDIA GTC returns on March 21 - 24. Join the virtual developer conference for the era of #AI. You’ll hear from visionaries who are solving the world's greatest challenges and learn how to accelerate your life’s work at #GTC22. Registration is free: nvda.ws/3ulRkty

ESSA helps businesses and learning providers to embrace #emergingtech and advancements in the #software sector. Find up-to-date resources & #skills intelligence on professional software roles 👥Subscribe to our channels 🔼More at softwareskills.eu DIGITALEUROPE

OPT-175b: Open Pre-Trained language model with 175 billion paramaters is now available to the research community. Blog post: ai.facebook.com/blog/democrati… Paper: arxiv.org/abs/2205.01068 Code + small pre-trained models: github.com/facebookresear… (using OPT-175b requires a registration)