Kernel Neural Optimal Transport

Authors: Alexander Korotin, Daniil Selikhanovych, Evgeny Burnaev

ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We test NOT with kernel costs on the unpaired image-to-image translation task.In this section, we test our algorithm on an unpaired image-to-image translation task.
Researcher Affiliation Academia Alexander Korotin Skolkovo Institute of Science and Technology Artificial Intelligence Research Institute Moscow, Russia a.korotin@skoltech.ru Daniil Selikhanovych Skolkovo Institute of Science and Technology Moscow, Russia selikhanovychdaniil@gmail.com Evgeny Burnaev Skolkovo Institute of Science and Technology Artificial Intelligence Research Institute Moscow, Russia e.burnaev@skoltech.ru
Pseudocode No The paper references 'NOT s training procedure (Korotin et al., 2023, Algorithm 1)' but does not include a pseudocode block or algorithm within this paper itself.
Open Source Code Yes The code is written in Py Torch framework and is available at https://github.com/iamalexkorotin/Kernel Neural Optimal Transport
Open Datasets Yes Image datasets. We test the following datasets as P, Q: aligned anime faces1, celebrity faces (Liu et al., 2015), shoes (Yu & Grauman, 2014), Amazon handbags, churches from LSUN dataset (Yu et al., 2015), outdoor images from the MIT places database (Zhou et al., 2014), describable textures (Cimpoi et al., 2014).
Dataset Splits No We pick 90% of each dataset for unpaired training. The rest 10% are considered as the test set. The paper explicitly mentions train and test splits but no dedicated validation split.
Hardware Specification Yes NOT with kernel costs for 128 128 images converges in 3-4 days on a 4 Tesla V100 GPUs (16 GB).
Software Dependencies No The paper mentions 'Py Torch framework' and 'Adam optimizer' but does not specify their version numbers or any other software dependencies with specific versions.
Experiment Setup Yes The learning rate is lr = 1 10 4. We use the Multi Step LR scheduler which decreases lr by 2 after [15k, 25k, 40k, 55k, 70k] (iterations of fω). The batch size is |X| = 64, |Zx| = 4. The number of inner iterations is k T = 10.