NABLA-SciML

Visit Tool

NABLA-SciML is a Research & Education tool that provides Physics Informed Machine Learning tutorials and implementations. It offers efficient and reproducible frameworks for PINNs, DeepONets, and KANs using PyTorch and JAX.

Claim this tool

2Views

At a glance

Pricing

Open Source

Free tier

Yes

API

Skill level

Technical

About

What is NABLA-SciML?

NABLA-SciML is a comprehensive collection of ongoing work in Scientific Machine Learning (SciML), developed by Juan Diego Toscano under the mentorship of Prof. George Karniadakis. This repository serves as a unified framework for efficient and reproducible implementations of various Physics-Informed Neural Networks (PINNs), DeepONets, and newer architectures like KANs. It includes tutorials for both PyTorch and JAX, along with specific modules for Residual-Based Attention (RBA), comprehensive KANs (cKANs), Kurkova-Kolmogorov-Arnold Networks (KKANs), and a Variational Framework for Residual-Based Adaptivity (vRBA). The project also features a custom, highly accurate implementation of the Self-Scaling Broyden (SSBroyden) optimizer, making it a valuable resource for researchers and students in the field.

Best used for

Ideal for students and researchers who need to learn and implement Physics-Informed Machine Learning models, including PINNs, DeepONets, and KANs. Especially valuable for those working with complex physical systems like fluid dynamics or biological flows, seeking reproducible and efficient computational frameworks.

Common actions

learn machine learning

implement neural networks

simulate physical systems

conduct scientific research

workflowslow-code/no-codedeepfakeopen-sourceautomated workflowcollaborationgithub copilot"AI Agents"face swapping

Capabilities

Key features

PINNs implementations
DeepONets implementations
KANs architectures
PyTorch tutorials
JAX tutorials
Residual-Based Attention
SSBroyden optimizer

Target Audience

studentresearcherdata scientistdeveloper

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What programming frameworks does NABLA-SciML support?

NABLA-SciML provides tutorials and implementations using both PyTorch and JAX. This allows users to work with their preferred deep learning framework for developing and experimenting with Physics-Informed Machine Learning models.

What types of neural network architectures are covered in NABLA-SciML?

The repository includes implementations and tutorials for Physics-Informed Neural Networks (PINNs), DeepONets, and newer architectures such as KANs (Kolmogorov-Arnold Networks), cKANs (comprehensive KANs), and KKANs (Kurkova-Kolmogorov-Arnold Networks).

Does NABLA-SciML include any custom optimizers?

Yes, the vRBA module within NABLA-SciML includes a custom, highly accurate implementation of the Self-Scaling Broyden (SSBroyden) optimizer. This optimizer is designed to enhance the performance of neural PDE solvers and operator learning.

Trending

Subcategories trending in Research & Education

Academic Research Study Assistants Knowledge Management Course Creation Summarization Language Learning

Trending

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce