A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
Maximum Likelihood-based Online Adaptation of Hyper-parameters in CMA-ES
[article]
2014
arXiv
pre-print
Preserved Fulltext
Other Versions
In this paper, we propose a principled approach called self-CMA-ES to achieve the online adaptation of CMA-ES hyper-parameters in order to improve its overall performance. ...
Experimental results show that for larger-than-default population size, the default settings of hyper-parameters of CMA-ES are far from being optimal, and that self-CMA-ES allows for dynamically approaching ...
This work was supported by the grant ANR-2010-COSI-002 (SIMINOLE) of the French National Research Agency. ...
arXiv:1406.2623v2
fatcat:7wrzq43gmnd4jofbkvw5vehlbq
Maximum Likelihood-Based Online Adaptation of Hyper-Parameters in CMA-ES
[chapter]
2014
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose a principled approach called self-CMA-ES to achieve the online adaptation of CMA-ES hyper-parameters in order to improve its overall performance. ...
CMA-ES is well known to be almost parameterless, meaning that only one hyper-parameter, the population size, is proposed to be tuned by the user. ...
This work was supported by the grant ANR-2010-COSI-002 (SIMINOLE) of the French National Research Agency. ...
doi:10.1007/978-3-319-10762-2_7
fatcat:6gdaa5o325cf5b5kznel62czvy
CMA-ES for Hyperparameter Optimization of Deep Neural Networks
[article]
2016
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
As an alternative, we propose to use the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), which is known for its state-of-the-art performance in derivative-free optimization. ...
We provide a toy example comparing CMA-ES and state-of-the-art Bayesian optimization algorithms for tuning the hyperparameters of a convolutional neural network for the MNIST dataset on 30 GPUs in parallel ...
In conclusion, we propose to consider CMA-ES as one alternative in the mix of methods for hyper-
parameter optimization of DNNs. ...
arXiv:1604.07269v1
fatcat:7hdozj4bevhtrdmwysjiw3daje
Heteroscedastic Bayesian Optimisation for Stochastic Model Predictive Control
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In these scenarios, performance outcomes present noise, which is not homogeneous across the domain of possible hyper-parameter settings, but which varies in an input-dependent way. ...
To address these issues, we propose a Bayesian optimisation framework that accounts for heteroscedastic noise to tune hyper-parameters in control problems. ...
The work in [18] proposes online hyper-parameter optimisation to improve MPPI's performance. ...
arXiv:2010.00202v2
fatcat:3ecprpn3dzdtfnnr7groam7as4
Parameter Setting for Multicore CMA-ES with Large Populations
[chapter]
2016
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Section 2 rapidly introduces the problem of parameter setting, and details the hyper-parameters of CMA-ES and how Self-CMA-ES adapts them. ...
The experiments presented in this paper have first validated most of the choices made in the original Self-CMA-ES approach [17] for the online control of the usually hidden parameters c c , c 1 , c µ ...
doi:10.1007/978-3-319-31471-6_9
fatcat:5g6hwtnpmveqddl6tzyfeqa7om
Information-Geometric Optimization with Natural Selection
2020
Entropy
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The algorithm is extremely simple in implementation; it has no matrix inversion or factorization, does not require storing a covariance matrix, and may form the basis of more general model-based optimization ...
Finally, we introduce a proof-of-principle algorithm that combines natural selection, our recombination operator, and an adaptive method to increase selection and find the optimum. ...
The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. ...
doi:10.3390/e22090967
pmid:33286736
pmcid:PMC7597266
fatcat:ai4omknzbba3bixx2bf6oy242u
Information-geometric optimization with natural selection
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
The algorithm is extremely simple in implementation with no matrix inversion or factorization, does not require storing a covariance matrix, and may form the basis of more general model-based optimization ...
Our algorithm is similar to covariance matrix adaptation and natural evolutionary strategies in optimization, and has similar performance. ...
For these benchmarks, QGA requires more fine tuning of its hyper-parameter than CMA-ES. However, QGA is simpler conceptually and in implementation compared to CMA-ES and NES. ...
arXiv:1912.03395v2
fatcat:b6gmoc7f2naphm3b3cgrr74yny
Evolutionary computation for wind farm layout optimization
2018
Renewable Energy
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Online and offline APIs were implemented in Cþþ, Java, Matlab and Python for this competition to offer a common framework for the competitors. ...
During this competition, competitors were tasked with optimizing the layouts of five generated wind farms based on a simplified cost of energy evaluation function of the wind farm layouts. ...
The work of A. Kheiri and E. Keedwell was funded by EPSRC grant no. EP/K000519/1. ...
doi:10.1016/j.renene.2018.03.052
fatcat:b63j4fqq2nfu5fjcndbv6fmiyi
Automatic tuning of hyper-parameters of reinforcement learning algorithms using Bayesian optimization with behavioral cloning
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Optimal setting of several hyper-parameters in machine learning algorithms is key to make the most of available data. ...
In reinforcement learning (RL), the information content of data gathered by the learning agent while interacting with its environment is heavily dependent on the setting of many hyper-parameters. ...
optimization 4 , reinforcement learning for neural
network architecture search 5 , Covariance Matrix Adaptation Evolution Strategy (CMA-ES) 6 , to name but a few. ...
arXiv:2112.08094v1
fatcat:hd4bvvjrpzgn3oweaq2l747k6m
Multi-objective Model-based Policy Search for Data-efficient Learning with Sparse Rewards
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
The experiments show that Multi-DEX is able to solve sparse reward scenarios (with a simulated robotic arm) in much lower interaction time than VIME, TRPO, GEP-PG, CMA-ES and Black-DROPS. ...
The most data-efficient algorithms for reinforcement learning in robotics are model-based policy search algorithms, which alternate between learning a dynamical model of the robot and optimizing a policy ...
We optimize the hyper-parameters of the kernel via Maximum Likelihood Estimation using the Rprop optimizer [13, 2] . ...
arXiv:1806.09351v3
fatcat:d6m3wcr7sfauno4r4auo5xsu6e
Using Parameterized Black-Box Priors to Scale Up Model-Based Policy Search for Robotics
2018
2018 IEEE International Conference on Robotics and Automation (ICRA)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
gaits in only 16 to 30 seconds of interaction time. ...
The most data-efficient algorithms for reinforcement learning in robotics are model-based policy search algorithms, which alternate between learning a dynamical model of the robot and optimizing a policy ...
Natural Evolution Strategies (NES) [22] and Covariance Matrix Adaptation ES (CMA-ES) [23] families of algorithms are population-based blackbox optimizers that iteratively update a search distribution ...
doi:10.1109/icra.2018.8461083
dblp:conf/icra/Chatzilygeroudis18
fatcat:4rexgxn7v5cwvfdzshaqd3oe7e
Using Parameterized Black-Box Priors to Scale Up Model-Based Policy Search for Robotics
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
gaits in only 16 to 30 seconds of interaction time. ...
The most data-efficient algorithms for reinforcement learning in robotics are model-based policy search algorithms, which alternate between learning a dynamical model of the robot and optimizing a policy ...
Natural Evolution Strategies (NES) [22] and Covariance Matrix Adaptation ES (CMA-ES) [23] families of algorithms are population-based blackbox optimizers that iteratively update a search distribution ...
arXiv:1709.06917v2
fatcat:7uo7o27esffyvhvrgbkcawrnv4
Adaptive Informative Path Planning Using Deep Reinforcement Learning for UAV-based Active Sensing
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
To address this, we propose a new approach for informative path planning based on deep reinforcement learning (RL). ...
However, a key challenge is efficiently planning paths to maximize the information value of acquired data as an initially unknown environment is explored. ...
We set CMA-ES parameters to 45 iterations, 12 offsprings, and (4, 4, 3) m coordinate-wise step size to trade-off between performance and runtime. ...
arXiv:2109.13570v2
fatcat:2tkvura4xfcdjjokgrj6zwgsyy
Evolution by Adapting Surrogates
2013
Evolutionary Computation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
methodology that brings about fitness improvement in the evolutionary search is introduced as the basis for adaptation. ...
The backbone of the proposed EvoLS is a statistical learning scheme to determine the evolvability of each approximation methodology while the search progresses online. ...
Le is grateful to the financial support of Honda Research Institute Europe. The authors would like to thank the editor and reviewers for their thoughtful suggestions and constructive comments. ...
doi:10.1162/evco_a_00079
pmid:22564044
fatcat:rcrwznblnvaovhtqamodmh4nse
A review on probabilistic graphical models in evolutionary computation
2012
Journal of Heuristics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Specifically, we give a survey of probabilistic model building-based evolutionary algorithms, called estimation of distribution algorithms, and compare different methods for probabilistic modeling in these ...
Thanks to their inherent properties, probabilistic graphical models are one of the prime candidates for machine learning and decision making tasks especially in uncertain domains. ...
Scoring metrics Most of the popular scoring metrics are based on one of the following approaches: (i) penalized maximum likelihood, and (ii) marginal likelihood. ...
doi:10.1007/s10732-012-9208-4
fatcat:54ipbzsryfbt5nqmaczgurb2he
« Previous
Showing results 1 — 15 out of 129 results