Skip to content

index.md

Latest commit

 

History

History
54 lines (42 loc) · 2.62 KB

index.md

File metadata and controls

54 lines (42 loc) · 2.62 KB
layout default
title Kolmogorov-Arnold Networks for Scientific Discovery

Kolmogorov-Arnold Networks for Scientific Discovery

Workshop Banner

Objective

This workshop will explore the use of alternative representations, such as Kolmogorov-Arnold Networks (KANs), for scientific and engineering applications, emphasizing the synergy between KAN representation, symbolic regression, and forward/inverse problem solving. We will focus on applications in molecular sciences (materials, proteins, protein-ligand interactions), geometrical representation (e.g., point-cloud prediction), PDE, and world model systems (e.g., reinforcement learning world models).

Scope

The scope includes theory, algorithms, applications, and tools. Suggested topics include, but are not limited to:

  • Neural Representations: Learning with KANs, radial basis expansion, adaptive kernel methods, smooth function approximation.
  • Symbolic and Sparse Regression: Integration of symbolic priors, sparsity-aware KAN training, differentiable symbolic discovery.
  • Molecular and Materials Applications: Protein structure-function relationships, protein-protein and protein-ligand modeling, cloud point estimation.
  • Inverse Problems and World Understanding: Solving ill-posed problems using structured priors, interpretable surrogates for simulation, knowledge-grounded modeling.
  • Comparison with Neural Operators and GNNs: Benchmarking KANs on PDEs, graph-structured data, and operator learning tasks.
  • Theory, Expressivity, and Robustness: Generalization of KAT perspectives, KAT complexity control, information bottlenecks in KANs.

Schedule

Time Activity Speaker
9:00-9:30 Registration
9:30-10:00 Keynote Speaker 1
10:00-10:30 Technical Session Speaker 2

Invited Talks

  • Talk Title 1 by Speaker Name (Affiliation)
  • Talk Title 2 by Speaker Name (Affiliation)

Important Dates

  • Paper Submission Deadline: Month Day, Year
  • Notification of Acceptance: Month Day, Year
  • Camera Ready Deadline: Month Day, Year
  • Workshop Date: Month Day, Year

Accepted Papers

Organizers

  • Jana Doppa (Washington State University)
  • Francesco Alesiani (NEC Labs Europe)
  • Yixuan (Roy) Wang (California Institute of Technology)
  • Xiaoyi Jiang (University of Münster)
  • Matthias Wolff (University of Münster)

Contact

Email