We present ISAAC (Input-baSed ApproximAte Curvature), a novel method tha...
Recently, research has increasingly focused on developing efficient neur...
The top-k classification accuracy is one of the core metrics in machine
...
In this work, we present and study a generalized family of differentiabl...
Differentiable sorting algorithms allow training with sorting and rankin...
The integration of algorithmic components into neural architectures has
...
Sorting and ranking supervision is a method for training neural networks...