Ziming Liu

Kolmogorov Arnold Networks

A Python library implementing Kolmogorov-Arnold Networks, a new type of neural network that is more interpretable than standard architectures, designed for scientific and mathematical problems.

Getting crystal-like representations with harmonic loss

Brain-Inspired Modular Training (BIMT), a method for making neural networks more modular and interpretable.

Omnigrok: Grokking Beyond Algorithmic Data

Understand what physics/algorithms do transformers learn internally when trained on planetary motion

We study toy models of skill learning.

Schrodinger Principal Component Analysis

Counting the number of conservation laws from trajectory data

Code for "Machine-Learning Non-Conservative Dynamics for New-Physics Detection" (arXiv: 2106.00026)

website

Quantum-inspired Hamiltonian Monte Carlo code

discovering interpretable conservation laws from differential equations

AI Poincare 2.0

No description.

Machine Learning hidden symmetries

This repository contains code examples for the Stanford's course: TensorFlow for Deep Learning Research.

NanoGPT (124M) in 3 minutes

analyze optimizer properties

Awesome paper list for physics of ai

No description.

No description.

expectation maximization

Models and examples built with TensorFlow

Implementation of spatial transformer networks in keras 2.0 using tensorflow 1.0 as backend.

No description.

LIU ZIMING's resume

2019 DATAOPEN held by Correlation One, sponsored by Citadel

Naive Bayes method

anti mapping of Hydro

Bringing up some extra Cosmo to Keras.

No description.

Deep Learning for humans

No description.

No description.

This repository is for notebooks of Mathematica.

No description.

Used for storing softwares