https://github.com/isl-org/generalized-smoothing
Companion code for the ICML 2022 paper "Generalizing Gaussian Smoothing for Random Search"
https://github.com/isl-org/generalized-smoothing
Last synced: 7 months ago
JSON representation
Companion code for the ICML 2022 paper "Generalizing Gaussian Smoothing for Random Search"
- Host: GitHub
- URL: https://github.com/isl-org/generalized-smoothing
- Owner: isl-org
- License: mit
- Created: 2022-11-15T01:42:59.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2024-07-22T05:28:15.000Z (over 1 year ago)
- Last Synced: 2025-04-17T18:21:39.139Z (9 months ago)
- Language: Python
- Size: 13.7 KB
- Stars: 8
- Watchers: 4
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Generalizing Gaussian Smoothing for Random Search
This repository contains code implementing the algorithms proposed in the paper [Generalizing Gaussian Smoothing for Random Search](https://proceedings.mlr.press/v162/gao22f.html), Gao and Sener (ICML 2022).
In particular, we provide the code used to obtain the experimental results on linear regression and the Nevergrad benchmark.
For online RL, we used the [ARS](https://github.com/modestyachts/ARS) repository; our proposed algorithms may be implemented by modifying the sampling distribution of the shared noise table.
Please see the paper for additional details and the hyperparameters used.
## Requirements
The code is written in Python 3.
Aside from the standard libraries, [NumPy](https://numpy.org/) and [Matplotlib](https://matplotlib.org/) are needed.
For linear regression, you also need [SciPy](https://scipy.org/), and for Nevergrad the corresponding [package](https://facebookresearch.github.io/nevergrad/).
## Running the experiments
Please see the READMEs in the `LinearRegression` and `benchmarks` folders for further instructions.
## Citation
To cite this repository in your research, please reference the following [paper]():
> Gao, Katelyn, and Ozan Sener. "Generalizing Gaussian Smoothing for Random Search." International Conference on Machine Learning. PMLR, 2022.
```TeX
@inproceedings{gao2022generalizing,
title={Generalizing Gaussian Smoothing for Random Search},
author={Gao, Katelyn and Sener, Ozan},
booktitle={International Conference on Machine Learning},
pages={7077--7101},
year={2022},
organization={PMLR}
}
```
## Contact
If you have questions, please contact .