Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/facebookresearch/amortized-optimization-tutorial
Tutorial on amortized optimization for learning to optimize over continuous domains
https://github.com/facebookresearch/amortized-optimization-tutorial
Last synced: 3 months ago
JSON representation
Tutorial on amortized optimization for learning to optimize over continuous domains
- Host: GitHub
- URL: https://github.com/facebookresearch/amortized-optimization-tutorial
- Owner: facebookresearch
- License: other
- Archived: true
- Created: 2022-01-31T01:47:23.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-04-26T17:03:28.000Z (over 1 year ago)
- Last Synced: 2024-05-03T06:39:10.995Z (7 months ago)
- Language: TeX
- Size: 66.2 MB
- Stars: 225
- Watchers: 8
- Forks: 12
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
# Tutorial on Amortized Optimization
This repository contains the source code for the paper
[Tutorial on amortized optimization for
learning to optimize over continuous domains](https://arxiv.org/abs/2202.00665)
by
[Brandon Amos](http://bamos.github.io).
The main LaTeX source is in [paper](./paper)
and the source code examples are in [code](./code).
The code that generates the following plots is also
in [code/figures](./code/figures):## [main-example.py](./code/figures/main-example.py)
![](./paper/fig/opt.png?raw=true)![](./paper/fig/learning-obj.png?raw=true)
![](./paper/fig/learning-reg.png?raw=true)![](./paper/fig/learning-rl.png?raw=true)
## [maxent-animation.py](./code/figures/maxent-animation.py)
![](./paper/fig/maxent.gif?raw=true)## [maxent.py](./code/figures/maxent.py)
![](./paper/fig/maxent.png?raw=true)## [ctrl.py](./code/figures/ctrl.py)
![](./paper/fig/ctrl.png?raw=true)## [imaml.py](./code/figures/imaml.py)
![](./paper/fig/imaml.png?raw=true)## [fixed-point.py](./code/figures/fixed-point.py)
![](./paper/fig/fp.png?raw=true)## [loss-comp.py](./code/figures/loss-comp.py)
![](./paper/fig/loss-comp.png?raw=true)## [smoothed-loss.py](./code/figures/smoothed-loss.py)
![](./paper/fig/smoothed-loss.png?raw=true)# Licensing
The source code for this tutorial, plots, and
sphere experiment is licensed under the
[CC BY-NC 4.0 License](https://creativecommons.org/licenses/by-nc/4.0/).