Morphism-Based Learning for Structured Data

Authors: Kilho Shin, Dave Shepard5767-5775

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we propose a generic and theoretic framework to investigate similarity of structured data through structure-preserving one-to-one partial mappings, which we call morphisms.Although the effectiveness of our framework has been already proven through some experiments, we will run experiments in a larger scale with a wider variation of machine learning methods including but not limited to distance, multiple alignment, pattern extraction and kernel.
Researcher Affiliation Academia Kilho Shin,1 David Lawrence Shepard2 1Gakushuin University, Japan, 2UCLA Scholarly Innovation Lab., USA
Pseudocode Yes Abstract Center Star Algorithm Input: X1, . . . , Xn D. Output: μij MXi,Xj for distinct i, j {1, . . . , n} with μij = μ 1 ji and μij μkj μik. Procedures: 1. For each Xi, compute morphisms μij MXi,Xj for j [1,ˆi, n] with d M ϕ,c(Xi, Xj) = Ψϕ,c( μij) and let Si = j [1,ˆi,n] d M ϕ,c(Xi, Xj). 2. Pick k arg min{Si | i = 1, . . . , n}; 3. Determine μki = μki, μik = μ 1 ki and μij = μkj μ 1 ki for i = k and j = k.
Open Source Code No The paper does not provide concrete access to source code for the methodology described. It states: 'For this purpose, we have a plan to develop utility programs that analyze an input dataset exhaustively and consistently by means of the morphism distance, the morphism-based pattern extraction and the moment kernels and others derived from appropriately parameterized pairs of (M, ϕ).'
Open Datasets No The paper is theoretical and focuses on developing a framework; it does not describe or use specific datasets for empirical evaluation within its scope.
Dataset Splits No The paper is theoretical and does not conduct experiments with datasets; therefore, no dataset split information for validation is provided.
Hardware Specification No The paper is theoretical and does not report on empirical experiments; therefore, no hardware specifications are provided.
Software Dependencies No The paper is theoretical and does not report on empirical experiments; therefore, no specific software dependencies are provided.
Experiment Setup No The paper is theoretical and focuses on developing a framework; it does not describe experimental setups, hyperparameters, or training settings for empirical evaluation.