Awesome Auto Research Hub Papers · Datasets · Projects
← Back to papers

Joint discovery of governing partial differential equations from multi-source datasets by competitive optimization

arXiv 2026 47.8 method

TLDR

A competitive optimization framework (MCO-PDE) discovers shared partial differential equations from multiple datasets, validated on real-world wave-tank experiments.

Reasoning

Strengths include a novel multi-source data fusion approach and real-world experimental validation. Weaknesses: the method is narrowly focused on PDE discovery and does not address broader automated scientific discovery tasks. The methodology is sound but limited in scope.

Read-first score

Read-first score 47.8, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 36.

Recency 8%
100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Methodology quality 25%
70

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=dataset,experiment,metric

Reproducibility 25%
38

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=dataset

Topical relevance 42%
30

Uses existing LLM keyword relevance scores normalized to 0-100. AI scientist,automated scientific discovery,autonomous research agent,automated research,literature review agent,survey generation,automated experimentation,experiment design agent,AI for scientific research,paper writing agent,research automation,scientific discovery agent

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 41.

Keyword Scores

automated scientific discovery
8
scientific discovery agent
7
AI for scientific research
6
automated research
5
research automation
5
AI scientist
2
automated experimentation
2
autonomous research agent
1
literature review agent
0
survey generation
0
experiment design agent
0
paper writing agent
0

Deep Analysis

Innovations

  • Competitive optimization framework for discovering shared PDEs from multi-source datasets (MCO-PDE)
  • Soft-competitive weighting mechanism to dynamically assess dataset credibility and aggregate a consensus global coefficient
  • Integration with a genetic algorithm for simultaneous structural and parametric identification of governing equations

Methodology

The framework trains independent neural surrogates for each data source, then applies a soft-competitive weighting mechanism to dynamically evaluate dataset credibility and aggregate a consensus global coefficient. A genetic algorithm searches the functional form space, enabling joint discovery of PDE structure and parameters from heterogeneous multi-source data.

Key Results

Fusing as few as 50 observations per dataset across seven cases recovers canonical equations with high accuracy; the method handles 2D/3D domains with irregular boundaries and heterogeneous coefficients, and extracts physically meaningful laws from real-world wave-tank experiments.

Tags

LGcomp-ph