Two-Sided Wasserstein Procrustes Analysis

Authors: Kun Jin, Chaoyue Liu, Cathy Xia

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

Reproducibility Variable Result LLM Response
Research Type Experimental We experiment TWP over three applications: protein alignment, language alignment and domain adaptation and demonstrate competitive performance of TWP compared with state-of-the-art methods.
Researcher Affiliation Academia Kun Jin1 , Chaoyue Liu1 , Cathy Xia2 1Department of Computer Science and Engineering, The Ohio State University 2Department of Integrated Systems Engineering, The Ohio State University
Pseudocode Yes Algorithm 1 TWP
Open Source Code No The paper does not provide any specific repository link or explicit statement about the release of source code for the described methodology.
Open Datasets Yes We use the data from [Wang and Mahadevan, 2008], where Glutaredoxin protein PDB-1G7O composed of 215 amino acids is used.
Dataset Splits No The paper discusses the construction of datasets and their use in experiments, but it does not provide specific training/validation/test split percentages or sample counts for reproducibility.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running the experiments.
Software Dependencies No The paper mentions algorithms like Sinkhorn Knoop and Gauss-Seidel method but does not provide specific software names with version numbers for reproducibility, such as Python library versions or specialized solvers.
Experiment Setup Yes To apply TWP, we set the dimension of the common embedding space to be d = 3 (3D), d = 2 (2D) and d = 1 (1D), respectively.