Artificial neural networks (ANNs) have emerged as an essential tool in machine learning,
achieving remarkable success across diverse domains, including image and speech …

Modern machine learning requires system designers to specify aspects of the learning
pipeline, such as losses, architectures, and optimizers. Meta-learning, or learning-to-learn …

Optimizing functions without access to gradients is the remit of black-box methods such as
evolution strategies. While highly general, their learning dynamics are often times heuristic …

In 2020-2021, we celebrated that many of the basic ideas behind the deep learning
revolution were published three decades ago within fewer than 12 months in our" Annus …

As the capabilities of large machine learning models continue to grow, and as the autonomy
afforded to such models continues to expand, the spectre of a new adversary looms: the …

This is a point-for-point critique of ACM's justification of the ACM AM Turing Award for deep
learning, as well as a critique of the Turing Lecture given by the awardees (published by …

In 2020 we celebrated the 1/3 century anniversary of my first publication on metalearning or
learning to learn: my diploma thesis of 1987[META1](Sec. 1). For its cover I drew a robot that …

On 14 June 2022, a science tabloid that published this article[LEC22b](24 June) on LeCun's
report" A Path Towards Autonomous Machine Intelligence"[LEC22a](27 June) sent me a …

End-To-End Differentiable Fast Weights: NNs Learn to Program NNs (1991) Three decades
ago, on 26 March 1991, I described a sequence-processing slow NN that learns by …