{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,28]],"date-time":"2026-06-28T01:05:13Z","timestamp":1782608713510,"version":"3.54.5"},"publisher-location":"California","reference-count":0,"publisher":"International Joint Conferences on Artificial Intelligence Organization","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,7]]},"abstract":"<jats:p>In many practical uses of reinforcement learning (RL) the set of actions available at a given state is a random variable, with realizations governed by an exogenous stochastic process. Somewhat surprisingly, the foundations for such sequential decision processes have been unaddressed. In this work, we formalize and investigate MDPs with stochastic action sets (SAS-MDPs) to provide these foundations. We show that optimal policies and value functions in this model have a structure that admits a compact representation. From an RL perspective, we show that Q-learning with sampled action sets is sound. In model-based settings, we consider two important special cases: when individual actions are available with independent probabilities, and a sampling-based model for unknown distributions. We develop polynomial-time value and policy iteration methods for both cases, and provide a polynomial-time linear programming solution for the first case.<\/jats:p>","DOI":"10.24963\/ijcai.2018\/650","type":"proceedings-article","created":{"date-parts":[[2018,7,5]],"date-time":"2018-07-05T01:49:10Z","timestamp":1530755350000},"page":"4674-4682","source":"Crossref","is-referenced-by-count":7,"title":["Planning and Learning with Stochastic Action Sets"],"prefix":"10.24963","author":[{"given":"Craig","family":"Boutilier","sequence":"first","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alon","family":"Cohen","sequence":"additional","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Avinatan","family":"Hassidim","sequence":"additional","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yishay","family":"Mansour","sequence":"additional","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ofer","family":"Meshi","sequence":"additional","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin","family":"Mladenov","sequence":"additional","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dale","family":"Schuurmans","sequence":"additional","affiliation":[{"name":"Google Research"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"10584","event":{"name":"Twenty-Seventh International Joint Conference on Artificial Intelligence {IJCAI-18}","theme":"Artificial Intelligence","location":"Stockholm, Sweden","acronym":"IJCAI-2018","number":"27","sponsor":["International Joint Conferences on Artificial Intelligence Organization (IJCAI)"],"start":{"date-parts":[[2018,7,13]]},"end":{"date-parts":[[2018,7,19]]}},"container-title":["Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence"],"original-title":[],"deposited":{"date-parts":[[2018,7,5]],"date-time":"2018-07-05T01:54:54Z","timestamp":1530755694000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ijcai.org\/proceedings\/2018\/650"}},"subtitle":[],"proceedings-subject":"Artificial Intelligence Research Articles","short-title":[],"issued":{"date-parts":[[2018,7]]},"references-count":0,"URL":"https:\/\/doi.org\/10.24963\/ijcai.2018\/650","relation":{},"subject":[],"published":{"date-parts":[[2018,7]]}}}