scratch>> Langevin-Dynamics-for-NN-Optimization>> 返回
项目作者： Unique-Divine

项目描述 ：
Bayesian Neural Network (BNN) implementations based on Langevin Dynamics and tested on real-world data
高级语言： Python
项目地址: git://github.com/Unique-Divine/Langevin-Dynamics-for-NN-Optimization.git
创建时间： 2021-04-24T19:38:46Z
项目社区：https://github.com/Unique-Divine/Langevin-Dynamics-for-NN-Optimization
开源协议：MIT License
下载


Langevin Dynamics-Based Neural Network Optimization on Real-World Data - Unique Divine

![Python 3.7+] License: MIT

This is my final project for the Applied Stochastic Analysis (APMA 4990) course at Columbia University.

Usage Instructions:

Inside optimization.py, there are PyTorch implementations for both the stochastic gradient Langevin dynamics (SGLD) optimizer and the preconditioned SGLD optimizer.

  • Li, Chen, Carlson, and Carin, 2016. Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks. [Paper link]
  • Welling and Teh, 2011. Bayesian Learning via Stochastiv Gradient Langevin Dynamics. [Paper link]

This repository works as a package. The results from the research report are collected using the model I implemented in lit_modules.py.