Math

• Simpler neural networks prefer subregular languages [pdf]
• Some thoughts on priors
• Sparsity Through Annealing of the $L_p$ (Semi-)Norm