-
Python code for the implementation of the method described in the paper "Neural Spectrahedra and Semidefinite Lifts: Global Convex Optimization of Polynomial Activation Neural Networks in Fully Polynomial-Time".
-
The paper can be downloaded at "https://arxiv.org/pdf/2101.02429.pdf".
-
The code in the notebook trains a 2-layer scalar output neural network for randomly generated data using the proposed convex SDP method.