Once again, the Transformers turn out to be amazing! Let's discover DINO from Facebook AI: towardsdatascience.com/on-dino-self-d… #transformers #ai #ml #DeepLearning #DINO

On DINO, Self-Distillation with no labels by Davide Coccomini buff.ly/3nKGXKL

DINO’s attention maps can discover and segment objects in an image or a #Video with absolutely no supervision and without being given a segmentation-targeted objective. #computervision ai.facebook.com/blog/dino-paws…

Transformers have become a standard in many NLP tasks and are poised to revolutionize the field of Computer Vision as well. Read this illustrated guide by Davide Coccomini to learn how transformers process texts, images, and videos. topbots.com/transformers-t… #AI #Transformers #ML

I just published in Towards Data Science Is attention what you really need in Transformers? towardsdatascience.com/is-attention-w… #transformers #deeplearning #MachineLearning

Is attention what you really need in Transformers? by Davide Coccomini buff.ly/3wWSlXf

Should Machine Learning Experts respond to Climate Change call to action? by Davide Coccomini buff.ly/3vxFIR6

Facebook AI has recently presented a self-supervised method called DINO, defined as a form of self-distillation with no labels, and used to train a Vision Transformer. Here's an illustrated guide to DINO prepared by Davide Coccomini. topbots.com/dino-vision-tr… #AI #ML #Transformers

Vision Transformers or Convolutional Neural Networks? Both! by Davide Coccomini buff.ly/3eWIz0O

In this article, Davide Coccomini suggests how machine learning can help to solve different problems related to climate change. For every topic, he also lists useful datasets which may be a good starting point for your project. topbots.com/machine-learni… #climatechange #ML #AI

Working with vision Transformers and CNNs shouldn't be an either/or, says Davide Coccomini, who explores ways to build hybrid architectures. buff.ly/2VapUHE

"Although Vision Transformers can achieve better results than other traditional architectures, their success depends on a considerable demand for data." Davide Coccomini looks at alternatives to training these models without extensive labelling work buff.ly/2WFrhij

Isn’t there a way to get the best out of both Convolutional Neural Networks and Vision Transformers? Davide Coccomini shows how these two architectures can be combined to obtain something capable of exploiting the positive sides of both! topbots.com/vision-transfo… #ML #Transformers

Scientists have shown us the vastness and variety of this infinite universe, artists have imagined and abstracted it since the dawn of time, but what does #artificialintelligence think about it? bit.ly/3iFrFpD #computervision #astronomy #space #astronomy #ml #ai

How does an AI Imagine the Universe? by Davide Coccomini buff.ly/3CT08J8

Can a Neural Network be creative? Can a Vision Transformer be a painter? What is creativity for a Vision Transformers? Discover in my last article on Towards Data Science! #deeplearning #transformers #ai #creativity #machinelearning #gan bit.ly/3tFqkCS

Vision Transformers are more than just a powerful technological advance, says Davide Coccomini—they're also a vehicle for creativity. buff.ly/3nzHuRt

Next AI/ML seminar features "Transformers in Vision: From Zero to Hero", by Davide Coccomini and Nicola, who will discuss: Transformer architecture and its mechanisms, transformer for vision and video understanding. Join us: aicamp.ai/event/eventdet…

If you're curious to learn about multimodal learning and the new opportunities that Transformers have opened up, here's a clear and useful explainer by Davide Coccomini. buff.ly/3ltbeNI