Paper Reviews

Intuitive experimentation in the physical world
Learning the Predictability of the Future
Scaling Laws for Neural Language Models
Why does deep and cheap learning work so well?
ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
The Developing Infant Creates a Curriculum for Statistical Learning
Self-supervised learning through the eyes of a child
Are we done with ImageNet?
High Fidelity Video Prediction with Large Stochastic Recurrent Neural Networks