https://github.com/igomezv/neuralike
Using machine learning to speed-up Bayesian inference.
https://github.com/igomezv/neuralike
bayesian-inference deep-learning nested-sampling neural-networks observational-cosmology torch
Last synced: 26 days ago
JSON representation
Using machine learning to speed-up Bayesian inference.
- Host: GitHub
- URL: https://github.com/igomezv/neuralike
- Owner: igomezv
- Created: 2022-04-18T23:55:02.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2024-10-15T21:59:45.000Z (over 1 year ago)
- Last Synced: 2024-10-17T10:53:28.270Z (over 1 year ago)
- Topics: bayesian-inference, deep-learning, nested-sampling, neural-networks, observational-cosmology, torch
- Language: Python
- Homepage:
- Size: 2.83 MB
- Stars: 2
- Watchers: 4
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# neuralike
## Deep Learning and Genetic Algorithms for Cosmological Bayesian Inference Speed-up
Code of our paper *Deep Learning and genetic algorithms for cosmological Bayesian inference speed-up*, Physical Review D, 110(8) 083518. Available at https://arxiv.org/abs/2405.03293.
## Repository Structure
- neuralike/
- **NeuraLike.py**.- Main class, gathers all other classes.
- **NeuralManager.py**.- API class, Manager for neural networks to learn likelihood function over a grid.
- **NeuralNet.py**.- Class with neural net architecture in PyTorch.
- **RandomSampling.py**.- Creates random samples in the parameter space and evaluates the likelihood in them. This is used to generate the training set for a neural network.
- **pytorchtools.py**.- Methods and utilities for PyTorch.
## Usage
In the branch **neuralike** of the repository https://github.com/igomezv/simplemc_tests it is available neuralike integrated within the dynesty library for nested sampling within the SimpleMC cosmological parameter estimation code (https://igomezv.github.io/SimpleMC/).
## Acknowledgments
We based our work on the following external codes:
- Philosophy of the method
- https://arxiv.org/abs/1110.2997
- https://github.com/DarkMachines/pyBAMBI
- Nested sampling library
- https://dynesty.readthedocs.io/en/stable
- Cosmological parameter estimation
- https://igomezv.github.io/SimpleMC
- Genetic algorithms library
- https://deap.readthedocs.io/en/master
- Deep learning library
- https://pytorch.org
## Citation
If you use this work in your research, please cite:
```bibtex
@article{gomez2024neuralike,
title = {Deep learning and genetic algorithms for cosmological Bayesian inference speed-up},
author = {G\'omez-Vargas, Isidro and V\'azquez, J. Alberto},
journal = {Phys. Rev. D},
volume = {110},
issue = {8},
pages = {083518},
numpages = {15},
year = {2024},
month = {Oct},
publisher = {American Physical Society},
doi = {10.1103/PhysRevD.110.083518},
url = {https://link.aps.org/doi/10.1103/PhysRevD.110.083518}
}
```
If you find useful our [`nnogada`](https://github.com/igomezv/Nnogada) framework for hyperparameter tuning of neural networks with genetic algorithms:
```bibtex
@article{nnogada,
title={Neural networks optimized by genetic algorithms in cosmology},
author={Gómez-Vargas, I. and Andrade, J. B. and Vázquez, J. A.},
journal={Physical Review D},
volume={107},
number={4},
pages={043509},
year={2023},
publisher={American Physical Society},
doi={https://doi.org/10.1103/PhysRevD.107.043509},
url={https://doi.org/10.48550/arXiv.2209.02685}
}
```