Gradient Origin Networks

Authors: Sam Bond-Taylor, Chris G. Willcocks

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experiments show that the proposed method converges faster, with significantly lower reconstruction error than autoencoders, while requiring half the parameters.
Researcher Affiliation Academia Sam Bond-Taylor & Chris G. Willcocks Department of Computer Science Durham University {samuel.e.bond-taylor,christopher.g.willcocks}@durham.ac.uk
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes Source code for the convolutional GON, variational GON, and implicit GON is available under the MIT license on Git Hub at: https://github.com/cwkx/GON.
Open Datasets Yes We evaluate Gradient Origin Networks on a variety of image datasets: MNIST (Le Cun et al., 1998), Fashion-MNIST (Xiao et al., 2017), Small NORB (Le Cun et al., 2004), COIL-20 (Nane et al., 1996), CIFAR-10 (Krizhevsky et al., 2009), Celeb A (Liu et al., 2015), and LSUN Bedroom (Yu et al., 2015).
Dataset Splits Yes Table 1: Validation reconstruction loss (summed squared error) over 500 epochs. Table 2: Validation ELBO in bits/dim over 1000 epochs (Celeb A is trained over 150 epochs).
Hardware Specification Yes This implementation uses Py Torch and all reported experiments use a Nvidia RTX 2080 Ti GPU.
Software Dependencies No The paper mentions 'Py Torch' but does not specify a version number for it or any other software dependency.
Experiment Setup Yes Simple models are used: for small images, implicit GONs consist of approximately 4 hidden layers of 256 units and convolutional GONs consist of 4 convolution layers with Batch Normalization (Ioffe & Szegedy, 2015) and the ELU non-linearity (Clevert et al., 2016), for larger images the same general architecture is used, scaled up with additional layers; all training is performed with the Adam optimiser (Kingma & Ba, 2015).