Does AI have to be complex?

To celebrate the start of the 21stᵉ century, the renowned journal Nature has unveiled its list of the world's ten most cited scientific papers. Four directly concern artificial intelligence, and in first place we find "Deep Residual Learning for Image Recognition", better known as ResNet. Surprisingly, this seminal paper is not technically extremely complex. And yet, it has revolutionized AI research, and continues to do so today, paving the way for countless advances.

So, does AI have to be complex to be revolutionary? At this conference/debate, we'll delve into the content of this article, look at the progress it has enabled, and then open a debate with the audience around the big questions - technical and/or philosophical - that it naturally raises.

The presentation and debate will be led by 2 experts: Clément Carré (founding president of Bionomeex - PhD specialized in statistical learning) and Jean Gabriel Ganascia.