← All blueprints
1
ReLU networks and piecewise linear functions
2
The maximum of five numbers with two hidden layers
3
Improved upper bounds on the depth
4
Geometric intuition: polytopes, support functions, and subdivisions
Dependency graph
Better Neural Network Expressivity: Subdividing the Simplex
Egor Bakaev, Florestan Brunck, Christoph Hertrich, Jack Stade, and Amir Yehudayoff
1
ReLU networks and piecewise linear functions
2
The maximum of five numbers with two hidden layers
3
Improved upper bounds on the depth
4
Geometric intuition: polytopes, support functions, and subdivisions