Fetching the paper…

Reading the bibliography…

2023

Hyperparameter Optimization as a Service on INFN Cloud

Barbetti, Matteo, Anderlini, Lucio

Original
PDF
Post on X

Understand

The simplest and often most effective way of parallelizing the training of complex machine learning models is to execute several training instances on multiple machines, scanning the hyperparameter space to optimize the underlying statistical model and the learning procedure.

Often, such a meta-learning procedure is limited by the ability of accessing securely a common database organizing the knowledge of the previous and ongoing trials.
Exploiting opportunistic GPUs provided in different environments represents a further challenge when designing such optimization campaigns.
In this contribution, we discuss how a set of REST APIs can be used to access a dedicated service based on INFN Cloud to monitor and coordinate multiple training instances, with gradient-less optimization techniques, via simple HTTP requests.

Hyperparameter Optimization as a Service on INFN Cloud · Around

Built on

Akiba T et al
Original
1907
Earlier work this paper cites.
Orr G and Müller K 2003 Neural Networks: Tricks of the Trade
2003
Earlier work this paper cites.
Bergstra J et al
2011
Earlier work this paper cites.
Bergstra J, Yamins D and Cox D 2013 Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures Proceedings of the 30th International Conference on Machine Learning (PMLR)
2013
Earlier work this paper cites.
Goodfellow I et al
Original
2014
Earlier work this paper cites.
Golovin D et al
2017
Earlier work this paper cites.
Liaw R et al
Original
2018
Earlier work this paper cites.
Albertsson K et al
2018
Earlier work this paper cites.

Similar

Brown T et al
2020
Cited alongside, same era.
Richens J G, Lee C M and Johri S 2020 Nat. Comm
2020
Cited alongside, same era.
Spiga D et al
2020
Cited alongside, same era.
Ramesh A et al
2021
Cited alongside, same era.
Head T et al
2021
Cited alongside, same era.
Tani L et al
2021
Cited alongside, same era.
Mariotti M, Spiga D and Boccali T 2021 PoS
2021
Cited alongside, same era.
Docker Compose https://docs.docker.com/compose

Then

Song X et al
Original
2022
Later among the works it cites.
Barbetti M and Anderlini L 2023 Reference implementation for Hopaas REST APIs Client in Python https://doi.org/10.5281/zenodo.7528502
2023
Closest in time.
Wulff E, Girone M and Pata J 2023 J. Phys. Conf. Ser
Original
2023
Closest in time.
Anderlini L et al
2023
Closest in time.
Barbetti M 2023 Lamarr: LHCb ultra-fast simulation based on machine learning models deployed within Gauss 21st International Workshop on Advanced Computing and Analysis Techniques in Physics Research (ACAT)
Original
2023
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Cited in the paper.

Uvicorn https://www.uvicorn.org

Cited in the paper.

FastAPI https://fastapi.tiangolo.com

Cited in the paper.

Optuna https://optuna.org

Cited in the paper.

NGINX https://www.nginx.com

Cited in the paper.

Anderlini L et al

2023

Closest in time.

Ratnikov F et al

2023

Closest in time.