{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/adaptive-system-optimization-using-random","title":"Adaptive system optimization using random directions stochastic approximation","arxiv_id":"1502.05577","date":"2015-02-19","proceeding":null,"authors":["Prashanth L. A.","Shalabh Bhatnagar","Michael Fu","Steve Marcus"],"abstract":"We present novel algorithms for simulation optimization using random\ndirections stochastic approximation (RDSA). These include first-order\n(gradient) as well as second-order (Newton) schemes. We incorporate both\ncontinuous-valued as well as discrete-valued perturbations into both our\nalgorithms. The former are chosen to be independent and identically distributed\n(i.i.d.) symmetric, uniformly distributed random variables (r.v.), while the\nlatter are i.i.d., asymmetric, Bernoulli r.v.s. Our Newton algorithm, with a\nnovel Hessian estimation scheme, requires N-dimensional perturbations and three\nloss measurements per iteration, whereas the simultaneous perturbation Newton\nsearch algorithm of [1] requires 2N-dimensional perturbations and four loss\nmeasurements per iteration. We prove the unbiasedness of both gradient and\nHessian estimates and asymptotic (strong) convergence for both first-order and\nsecond-order schemes. We also provide asymptotic normality results, which in\nparticular establish that the asymmetric Bernoulli variant of Newton RDSA\nmethod is better than 2SPSA of [1]. Numerical experiments are used to validate\nthe theoretical results.","url_abs":"http://arxiv.org/abs/1502.05577v2","url_pdf":"http://arxiv.org/pdf/1502.05577v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"adaptive-system-optimization-using-random","repo_url":"https://github.com/prashla/RDSA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}