vrn25 / TB_Opt

Code for TB_Opt

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TB_Opt: A General Game Theoretic Recursive Framework for Turn-Based Optimization

This repository contains the PyTorch code for TB_Opt framework. The technical report can be found here.

The code is organized as follows:

.
├── README.md
├── x_y_x2y2
│   ├── optimizers_generic.py
│   ├── plots
│   │   ├── Bilinear_Bilinear
│   │   ├── Bilinear_Linear
│   │   ├── Linear_Bilinear
│   │   └── Linear_Linear
│   └── run.py   
├── x_y_xy
    ├── optimizers_xy.py
    ├── plots
    │   ├── Bilinear_Bilinear
    │   ├── Bilinear_Linear
    │   ├── Linear_Bilinear
    │   └── Linear_Linear
    └── run.py

x_y_xy contains the code for the game equation and x_y_x2y2 contains the code for the game equation.

optimizers_generic contains the the code for a generic optimizer (linear-linear, linear-bilinear, bilinear-linear, bilinear-bilinear cases) that works for any objectives equation and equation. Further details can be found here.

To run and experiment with the implementation in your browser:

Game x_y_xy: Open In Colab

Game x_y_x2y2: Open In Colab

About

Code for TB_Opt


Languages

Language:Python 100.0%