Temporal Factorization of 3D Convolutional Kernels
3D convolutional neural networks are difficult to train be-cause they are parameter-expensive and data-hungry. To solve these problems we propose a simple technique for learning 3D convolutional kernels efficiently requiring less training data. We achieve this by factorizing the 3D kernel along the temporal dimension, reducing the number of parameters and making training from data more efficient. Additionally we introduce the Video-MNIST dataset to demonstrate the performance of our method.
Gabriëlle received her bachelor's degree in Knowledge Engineering from Maastricht University in 2015. Afterwards she graduated the Artificial Intelligence master's program cum laude at Radboud University in 2017. Now she is a doctoral candidate at the Artificial Cognitive Systems lab at Radboud University conducting research in the field of affective computing.