Generalized Tensor Models for Recurrent Neural Networks

Authors: Valentin Khrulkov, Oleksii Hrinchuk, Ivan Oseledets

ICLR 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our theoretical results are verified by a series of extensive computational experiments. In this section, we study if our theoretical findings are supported by experimental data. In particular, we investigate whether generalized tensor networks can be used in practical settings, especially in problems typically solved by RNNs (such as natural language processing problems).
Researcher Affiliation Academia 1Skolkovo Institute of Science and Technology, Moscow, Russia 2Moscow Institute of Physics and Technology, Moscow, Russia 3Institute of Numerical Mathematics, Russian Academy of Sciences, Moscow, Russia
Pseudocode No The paper does not contain any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper does not provide any statement or link indicating the release of open-source code for the described methodology.
Open Datasets Yes For the first experiment, we use two computer vision datasets MNIST (Le Cun et al., 1990) and CIFAR 10 (Krizhevsky & Hinton, 2009), and natural language processing dataset for sentiment analysis IMDB (Maas et al., 2011).
Dataset Splits No The paper mentions training and testing but does not specify validation splits or proportions.
Hardware Specification No The paper does not specify any hardware details like GPU models, CPU types, or memory used for running the experiments.
Software Dependencies No The paper mentions
Experiment Setup Yes Parameters of all networks were optimized using Adam (learning rate α = 10 4) and batch size 250.