dorsal/arxiv
View SchemaImproving CMA-ES Convergence Speed, Efficiency, and Reliability in Noisy Robot Optimization Problems
| Authors | Russell M. Martin, Steven H. Collins |
|---|---|
| Categories | |
| ArXiv ID | 2601.09594vv1 |
| URL | https://arxiv.org/abs/2601.09594 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Experimental robot optimization often requires evaluating each candidate policy for seconds to minutes. The chosen evaluation time influences optimization because of a speed-accuracy tradeoff: shorter evaluations enable faster iteration, but are also more subject to noise. Here, we introduce a supplement to the CMA-ES optimization algorithm, named Adaptive Sampling CMA-ES (AS-CMA), which assigns sampling time to candidates based on predicted sorting difficulty, aiming to achieve consistent precision. We compared AS-CMA to CMA-ES and Bayesian optimization using a range of static sampling times in four simulated cost landscapes. AS-CMA converged on 98% of all runs without adjustment to its tunable parameter, and converged 24-65% faster and with 29-76% lower total cost than each landscape's best CMA-ES static sampling time. As compared to Bayesian optimization, AS-CMA converged more efficiently and reliably in complex landscapes, while in simpler landscapes, AS-CMA was less efficient but equally reliable. We deployed AS-CMA in an exoskeleton optimization experiment and found the optimizer's behavior was consistent with expectations. These results indicate that AS-CMA can improve optimization efficiency in the presence of noise while minimally affecting optimization setup complexity and tuning requirements.
{
"annotation_id": "aa0084c9-592f-4f75-ac14-837b1e7b3d78",
"date_created": "2026-02-17T05:53:19.558000Z",
"date_modified": "2026-02-17T05:53:19.558000Z",
"file_hash": "642b5a25738a77a71537dd46477030f35b64a8390a519da3bf314047ec025798",
"private": false,
"record": {
"abstract": "Experimental robot optimization often requires evaluating each candidate policy for seconds to minutes. The chosen evaluation time influences optimization because of a speed-accuracy tradeoff: shorter evaluations enable faster iteration, but are also more subject to noise. Here, we introduce a supplement to the CMA-ES optimization algorithm, named Adaptive Sampling CMA-ES (AS-CMA), which assigns sampling time to candidates based on predicted sorting difficulty, aiming to achieve consistent precision. We compared AS-CMA to CMA-ES and Bayesian optimization using a range of static sampling times in four simulated cost landscapes. AS-CMA converged on 98% of all runs without adjustment to its tunable parameter, and converged 24-65% faster and with 29-76% lower total cost than each landscape\u0027s best CMA-ES static sampling time. As compared to Bayesian optimization, AS-CMA converged more efficiently and reliably in complex landscapes, while in simpler landscapes, AS-CMA was less efficient but equally reliable. We deployed AS-CMA in an exoskeleton optimization experiment and found the optimizer\u0027s behavior was consistent with expectations. These results indicate that AS-CMA can improve optimization efficiency in the presence of noise while minimally affecting optimization setup complexity and tuning requirements.",
"arxiv_id": "2601.09594",
"authors": [
"Russell M. Martin",
"Steven H. Collins"
],
"categories": [
"cs.NE"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Improving CMA-ES Convergence Speed, Efficiency, and Reliability in Noisy Robot Optimization Problems",
"url": "https://arxiv.org/abs/2601.09594",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "0d76713d-f697-4104-ba93-85da1bac6478",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}