Will be the transformer the model leading us to artificial general intelligence? Or will be replaced? — The transformer has dominated the world of artificial intelligence for six years, achieving state-of-the-art in all subdomains of artificial intelligence. From natural language processing (NLP) to computer vision to sound and graphs, there are dedicated transformers with excellent performance. How much longer will this domain last?

Art masterpieces are a risk at any time; AI and new technologies can give a hand — In recent months there has been talk about how artificial intelligence can create images from textual prompts. Therefore, when one associates the words artificial intelligence and art, one immediately thinks of DALL-E, Stable Diffusion, and other algorithms. …

Amazonia is at risk and AI could help preserve it — Drug traffickers, gold diggers, loggers, farmers, and ranchers are invading the Amazon rainforest. The pandemic and Bolsonaro’s presidency have exacerbated an already bleak situation. How can science and artificial intelligence help save the most essential rainforest? The green lung of earth

AI models are now able to generate images from text, what if we furnish them with the words of great poets? A dreamy trip between poetry and AI. — Introduction Poetry is nothing more than the juxtaposition of skillfully chosen words. A sequence of characters, yet capable of provoking emotion. An art that is difficult to explain and codify. Listening to or reading a poem, we almost feel as if we can feel emotions, chills, or by closing our eyes…

Evolutive forces shape the brain, what if we apply the same forces to AI? — The human brain is capable of complex tasks while being energetically efficient. Throughout its development, various efforts have pushed it to solve complicated interactions with the world and at the same time optimize itself. Why not be inspired by such an efficient yet elegant system to develop the next generation…

What is the associated carbon dioxide of your favorite model? — 2023 was the year of generative AI, this explosion of it and also associated with a dark side. How much energy does it consume? How much carbon dioxide does it emit? Machine learning and climate change? Love and thunder!

What there are? Why do you care? The name is intimidating, but fear not them! — I am presenting an introduction in plain English about one of the more powerful and basic concepts in neural networks. I am providing also code (for PyTorch and Tensorflow) and an Excel file companion to better digest the concept. The physicist who states a law of nature with the aid…

Can we simplify the current architectures without losing performance? — The transformer is synonymous with self-attention. Despite their complexity, this model has changed the history of artificial intelligence, but is there no way to be able to simplify it? Do we necessarily need complex architectures to solve complex tasks? Some researchers think not, as I discuss in this article The curse of attention

From great powers derive great responsibilities: regularization allows AI to exploit its power — Why do tree-based models outperform neural networks for tabular data? For various reasons, deep learning models fail to compete with tree-based models for tabular data. Considering how widespread tabular data are this is a decisive unmet need. Tabula Rasa: Why Do Tree-Based Algorithms Outperform Neural Networks Tree-based algorithms are the winner in tabular data: Why?levelup.gitconnected.com