{"id":223684,"date":"2024-01-23T16:25:51","date_gmt":"2024-01-23T16:25:51","guid":{"rendered":"https:\/\/datascientest.com\/?p=223684"},"modified":"2024-01-24T14:04:29","modified_gmt":"2024-01-24T14:04:29","slug":"sarsa-ou-apprentissage-par-renforcement","status":"publish","type":"post","link":"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement","title":{"rendered":"SARSA : Comment \u00e7a fonctionne en Machine Learning ?"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"223684\" class=\"elementor elementor-223684\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3c40103 e-flex e-con-boxed e-con\" data-id=\"3c40103\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-e713be0 elementor-widget elementor-widget-heading\" data-id=\"e713be0\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-heading-title{padding:0;margin:0;line-height:1}.elementor-widget-heading .elementor-heading-title[class*=elementor-size-]>a{color:inherit;font-size:inherit;line-height:inherit}.elementor-widget-heading .elementor-heading-title.elementor-size-small{font-size:15px}.elementor-widget-heading .elementor-heading-title.elementor-size-medium{font-size:19px}.elementor-widget-heading .elementor-heading-title.elementor-size-large{font-size:29px}.elementor-widget-heading .elementor-heading-title.elementor-size-xl{font-size:39px}.elementor-widget-heading .elementor-heading-title.elementor-size-xxl{font-size:59px}<\/style><h2 class=\"elementor-heading-title elementor-size-default\">Le Reinforcement Learning, ou apprentissage par renforcement en fran\u00e7ais, est, avec les apprentissages supervis\u00e9 et non supervis\u00e9, l\u2019une des trois grandes techniques d\u2019apprentissage automatique.<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-7f30dc9 elementor-widget elementor-widget-text-editor\" data-id=\"7f30dc9\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-text-editor.elementor-drop-cap-view-stacked .elementor-drop-cap{background-color:#69727d;color:#fff}.elementor-widget-text-editor.elementor-drop-cap-view-framed .elementor-drop-cap{color:#69727d;border:3px solid;background-color:transparent}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap{margin-top:8px}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap-letter{width:1em;height:1em}.elementor-widget-text-editor .elementor-drop-cap{float:left;text-align:center;line-height:1;font-size:50px}.elementor-widget-text-editor .elementor-drop-cap-letter{display:inline-block}<\/style>\t\t\t\t<p><span style=\"font-weight: 400;\">Cette famille d\u2019algorithme cr\u00e9e beaucoup d\u2019engouement depuis quelques ann\u00e9es, avec notamment les produits innovants de <a href=\"https:\/\/datascientest.com\/openai-lentreprise-qui-revolutionne-lia\">l\u2019entreprise OpenAI<\/a> tels que OpenAI Five, une IA qui a r\u00e9ussi \u00e0 battre une \u00e9quipe de joueurs professionnels sur le jeu vid\u00e9o Dota 2, ou le c\u00e9l\u00e8bre <a href=\"https:\/\/datascientest.com\/formation-chatgpt\">ChatGPT<\/a> qui utilise cette technique pour ajuster ses param\u00e8tres.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-26521ff elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"26521ff\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formation-data-scientist\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Tout apprendre du reinforcement learning<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-dbe63af e-flex e-con-boxed e-con\" data-id=\"dbe63af\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-3c5ca89 elementor-widget elementor-widget-heading\" data-id=\"3c5ca89\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Qu\u2019est-ce que le reinforcement learning ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-0c24a22 elementor-widget elementor-widget-text-editor\" data-id=\"0c24a22\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<a href=\"https:\/\/datascientest.com\/reinforcement-learning\"><span style=\"font-weight: 400;\">L\u2019apprentissage par renforcement<\/a>, est un domaine de l\u2019apprentissage automatique ou un <\/span><b>agent <\/b><span style=\"font-weight: 400;\">(entit\u00e9 virtuelle: robot, programme\u2026) est plac\u00e9 dans un<\/span><b> environnement interactif <\/b><span style=\"font-weight: 400;\">dans lequel il doit apprendre \u00e0 faire des <\/span><b>actions <\/b><span style=\"font-weight: 400;\">maximisant des <\/span><b>r\u00e9compenses <\/b><span style=\"font-weight: 400;\">quantitatives.<\/span>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-7952116 e-flex e-con-boxed e-con\" data-id=\"7952116\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-dfda5c4 elementor-widget elementor-widget-heading\" data-id=\"dfda5c4\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Qu\u2019est-ce que l\u2019algorithme SARSA ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-9acd062 elementor-widget elementor-widget-text-editor\" data-id=\"9acd062\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">SARSA est un algorithme d\u2019apprentissage dont le nom vient de l\u2019anglais <strong>State-Action-Reward-State-Action<\/strong>, qui signifie \u00c9tat-Action-R\u00e9compense-\u00c9tat-Action et qui d\u00e9signe la suite d\u2019\u00e9l\u00e9ments qui constituent cet algorithme. Il s&rsquo;agit d&rsquo;un <a href=\"https:\/\/datascientest.com\/algorithme-tout-savoir\">algorithme<\/a> bas\u00e9 sur un tableau de valeurs d&rsquo;action (ou <strong>Q-table<\/strong>, Q repr\u00e9sentant la mesure de la qualit\u00e9 d\u2019une action effectu\u00e9e) qui attribue \u00e0 chaque paire \u00e9tat-action une valeur repr\u00e9sentant la r\u00e9compense attendue.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-492a3d4 e-flex e-con-boxed e-con\" data-id=\"492a3d4\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-f8fa83b elementor-widget elementor-widget-heading\" data-id=\"f8fa83b\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">D\u00e9roulement de l\u2019algorithme<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3dd6257 elementor-widget elementor-widget-text-editor\" data-id=\"3dd6257\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">On peut illustrer les diff\u00e9rentes \u00e9tapes qui constituent cet algorithme en prenant pour exemple un livreur qui doit amener un colis d\u2019un point A \u00e0 un point B.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3974aa3 elementor-widget elementor-widget-image\" data-id=\"3974aa3\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-image{text-align:center}.elementor-widget-image a{display:inline-block}.elementor-widget-image a img[src$=\".svg\"]{width:48px}.elementor-widget-image img{vertical-align:middle;display:inline-block}<\/style>\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"750\" height=\"500\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/SARSA.jpg\" class=\"attachment-large size-large wp-image-223693\" alt=\"\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/SARSA.jpg 750w, https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/SARSA-300x200.jpg 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-a009c77 elementor-widget elementor-widget-text-editor\" data-id=\"a009c77\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Tout d\u2019abord la Q-table est initialis\u00e9e avec des valeurs nulles. Les diff\u00e9rentes \u00e9tapes de SARSA peuvent ensuite commencer :<\/span><\/p><ul><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">L\u2019algorithme est dans un \u00e9tat <\/span><b>S<\/b><span style=\"font-weight: 400;\"> (Le livreur conna\u00eet sa distance par rapport au point B)<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Vous choisissez une action <\/span><b>A<\/b><span style=\"font-weight: 400;\"> \u00e0 effectuer, soit en exploitant vos connaissances soit en explorant des nouvelles possibilit\u00e9s (le livreur arrive \u00e0 un croisement depuis le nord et conna\u00eet la route qui va vers le sud, il suit donc cette route)<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Vous recevez une r\u00e9compense <\/span><b>R<\/b><span style=\"font-weight: 400;\"> (le livreur gagne 1 minute si son choix le rapproche du point B mais perd 5 minutes si il s\u2019en \u00e9loigne). A ce moment la Q-table est mise \u00e0 jour \u00e0 l\u2019aide de la formule Q[\u00e9tat_1, action_1] = (1-\u03b1) * Q[\u00e9tat_1, action_1] + \u03b1 * (r + \u03b3 * Q[\u00e9tat_2, action_2]) avec<\/span><ul><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Le <\/span><b>taux d&rsquo;apprentissage<\/b><span style=\"font-weight: 400;\"> alpha ou \u03b1 qui contr\u00f4le l&rsquo;ampleur de la mise \u00e0 jour. Si \u03b1 = 0 on n\u2019apprend rien, si \u03b1 = 1 on apprend en oubliant ce qui a \u00e9t\u00e9 appris pr\u00e9c\u00e9demment.<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">La <\/span><b>r\u00e9compense re\u00e7ue<\/b><span style=\"font-weight: 400;\"> r apr\u00e8s avoir fait une action \u00e0 un instant donn\u00e9.<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Le<\/span><b> facteur d&rsquo;actualisation<\/b><span style=\"font-weight: 400;\"> gamma ou \u03b3 qui pond\u00e8re les r\u00e9compenses futures.<\/span><\/li><\/ul><\/li><\/ul><ul><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">L\u2019algorithme est dans un nouvel \u00e9tat <\/span><b>S<\/b><span style=\"font-weight: 400;\"> (le livreur conna\u00eet sa nouvelle distance par rapport au point B)<\/span><\/li><li><span style=\"font-weight: 400;\">Vous choisissez d\u2019effectuer une nouvelle action\u00a0<\/span><b>A<\/b><\/li><\/ul>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ad3de5c elementor-widget elementor-widget-text-editor\" data-id=\"ad3de5c\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Ces \u00e9tapes se r\u00e9p\u00e8tent jusqu\u2019\u00e0 ce que l\u2019algorithme converge (jusqu\u2019\u00e0 ce que le livreur arrive au point B).<\/span><\/p><p><span style=\"font-weight: 400;\">Cet algorithme est appel\u00e9 \u00ab\u00a0on-policy\u00a0\u00bb, ce qui signifie que la politique utilis\u00e9e pour d\u00e9terminer les actions est la m\u00eame que celle qui met \u00e0 jour les valeurs de la table Q. SARSA utilise g\u00e9n\u00e9ralement une politique \u03b5-greedy, o\u00f9 \u03b5 est un param\u00e8tre qui d\u00e9termine la probabilit\u00e9 d&rsquo;exploration (s\u00e9lection al\u00e9atoire d&rsquo;actions) par rapport \u00e0 l&rsquo;exploitation (s\u00e9lection de la meilleure action selon la Q-table). Cette caract\u00e9ristique le diff\u00e9rencie des algorithmes \u00ab\u00a0off-policy\u00a0\u00bb comme le <a href=\"https:\/\/datascientest.com\/q-learning-le-machine-learning-avec-apprentissage-par-renforcement\">Q-Learning<\/a>.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-940d9d0 e-flex e-con-boxed e-con\" data-id=\"940d9d0\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-467ede0 elementor-widget elementor-widget-heading\" data-id=\"467ede0\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Conclusion<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-dab07cb elementor-widget elementor-widget-text-editor\" data-id=\"dab07cb\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">En r\u00e9sum\u00e9, SARSA est un algorithme d&rsquo;apprentissage par renforcement qui vise \u00e0 apprendre \u00e0 un agent les d\u00e9cisions \u00e0 prendre dans un environnement gr\u00e2ce \u00e0 une <strong>Q-table<\/strong> mise \u00e0 jour it\u00e9rativement. Il suit une politique d&rsquo;exploration et d&rsquo;exploitation tout en interagissant avec l&rsquo;environnement, et il est utilis\u00e9 dans divers domaines comme les jeux vid\u00e9o,<\/span><span style=\"font-weight: 400;\"> la prise de d\u00e9cisions dans la robotique, ou la r\u00e9solution de probl\u00e8mes de planification de trajets.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-cfe2f75 elementor-widget elementor-widget-image\" data-id=\"cfe2f75\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"762\" height=\"600\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/SARSA-2.jpg\" class=\"attachment-large size-large wp-image-223697\" alt=\"\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/SARSA-2.jpg 762w, https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/SARSA-2-300x236.jpg 300w\" sizes=\"(max-width: 762px) 100vw, 762px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d0567af elementor-widget elementor-widget-text-editor\" data-id=\"d0567af\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Si vous souhaitez vous former \u00e0 ce domaine, n\u2019h\u00e9sitez pas \u00e0 consulter notre <a href=\"https:\/\/datascientest.com\/formation-data-scientist-belgique\">formation de Data Scientist<\/a>.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-6ece749 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"6ece749\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formation-data-scientist\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">D\u00e9couvrir notre formation Data Scientist<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-346b65b elementor-widget elementor-widget-html\" data-id=\"346b65b\" data-element_type=\"widget\" data-widget_type=\"html.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<script type=\"application\/ld+json\">\r\n{\r\n  \"@context\": \"https:\/\/schema.org\",\r\n  \"@type\": \"FAQPage\",\r\n  \"mainEntity\": [{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Qu\u2019est-ce que l\u2019algorithme SARSA ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"SARSA est un algorithme d\u2019apprentissage dont le nom vient de l\u2019anglais State-Action-Reward-State-Action, qui signifie \u00c9tat-Action-R\u00e9compense-\u00c9tat-Action et qui d\u00e9signe la suite d\u2019\u00e9l\u00e9ments qui constituent cet algorithme. Il s'agit d'un algorithme bas\u00e9 sur un tableau de valeurs d'action (ou Q-table, Q repr\u00e9sentant la mesure de la qualit\u00e9 d\u2019une action effectu\u00e9e) qui attribue \u00e0 chaque paire \u00e9tat-action une valeur repr\u00e9sentant la r\u00e9compense attendue.\"\r\n    }\r\n  },{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Comment fonctionne l'algorithme SARSA ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"Cet algorithme est appel\u00e9 \u201con-policy\u201d, ce qui signifie que la politique utilis\u00e9e pour d\u00e9terminer les actions est la m\u00eame que celle qui met \u00e0 jour les valeurs de la table Q. SARSA utilise g\u00e9n\u00e9ralement une politique \u03b5-greedy, o\u00f9 \u03b5 est un param\u00e8tre qui d\u00e9termine la probabilit\u00e9 d'exploration (s\u00e9lection al\u00e9atoire d'actions) par rapport \u00e0 l'exploitation (s\u00e9lection de la meilleure action selon la Q-table). Cette caract\u00e9ristique le diff\u00e9rencie des algorithmes \u201coff-policy\u201d comme le Q-Learning.\"\r\n    }\r\n  }]\r\n}\r\n<\/script>\r\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Le Reinforcement Learning, ou apprentissage par renforcement en fran\u00e7ais, est, avec les apprentissages supervis\u00e9 et non supervis\u00e9, l\u2019une des trois grandes techniques d\u2019apprentissage automatique. Cette famille d\u2019algorithme cr\u00e9e beaucoup d\u2019engouement depuis quelques ann\u00e9es, avec notamment les produits innovants de l\u2019entreprise OpenAI tels que OpenAI Five, une IA qui a r\u00e9ussi \u00e0 battre une \u00e9quipe de [&hellip;]<\/p>\n","protected":false},"author":78,"featured_media":223686,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"content-type":"","inline_featured_image":false},"categories":[2497],"tags":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.7 (Yoast SEO v21.7) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>SARSA : Comment \u00e7a fonctionne en Machine Learning ?<\/title>\n<meta name=\"description\" content=\"D\u00e9couvrez tout ce qu&#039;il y a \u00e0 savoir sur le SARSA ou State-Action-Reward-State-Action, qui est un algorithme d\u2019apprentissage par renforcement.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"SARSA : Comment \u00e7a fonctionne en Machine Learning ?\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement\" \/>\n<meta property=\"og:site_name\" content=\"Formation Data Science | DataScientest.com\" \/>\n<meta property=\"article:published_time\" content=\"2024-01-23T16:25:51+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-01-24T14:04:29+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/Illu_BLOG__SARSA.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"562\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"J\u00e9r\u00e9my Robert\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"J\u00e9r\u00e9my Robert\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement\",\"url\":\"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement\",\"name\":\"SARSA : Comment \u00e7a fonctionne en Machine Learning ?\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/#website\"},\"datePublished\":\"2024-01-23T16:25:51+00:00\",\"dateModified\":\"2024-01-24T14:04:29+00:00\",\"author\":{\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925\"},\"description\":\"D\u00e9couvrez tout ce qu'il y a \u00e0 savoir sur le SARSA ou State-Action-Reward-State-Action, qui est un algorithme d\u2019apprentissage par renforcement.\",\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/datascientest.com\/#website\",\"url\":\"https:\/\/datascientest.com\/\",\"name\":\"Formation Data Science | DataScientest.com\",\"description\":\"Leader de la formation en Data Science\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/datascientest.com\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925\",\"name\":\"J\u00e9r\u00e9my Robert\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg\",\"contentUrl\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg\",\"caption\":\"J\u00e9r\u00e9my Robert\"},\"description\":\"J\u00e9r\u00e9my Robert est actuellement Directeur p\u00e9dagogique B2B chez DataScientest.com. Avant de rejoindre DataScientest, il a travaill\u00e9 en tant que Consultant Business Intelligence chez DEAPDATA, o\u00f9 il a g\u00e9r\u00e9 des projets dans les domaines de l'assurance, bancaire et industrie. Il a \u00e9galement \u00e9t\u00e9 Co-fondateur et pr\u00e9sident de Tripzer. En tant que consultant ind\u00e9pendant, J\u00e9r\u00e9my a propos\u00e9 des formations sur MicroStrategy pour divers niveaux professionnels. Il est dipl\u00f4m\u00e9 de l'Universit\u00e9 Paris Cit\u00e9 avec un Master MIAGE, combinant comp\u00e9tences en informatique et gestion.\",\"sameAs\":[\"https:\/\/www.linkedin.com\/in\/jeremyrobert1\/\",\"Data Science\"],\"url\":\"https:\/\/datascientest.com\/author\/jeremyrobert\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"SARSA : Comment \u00e7a fonctionne en Machine Learning ?","description":"D\u00e9couvrez tout ce qu'il y a \u00e0 savoir sur le SARSA ou State-Action-Reward-State-Action, qui est un algorithme d\u2019apprentissage par renforcement.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement","og_locale":"fr_FR","og_type":"article","og_title":"SARSA : Comment \u00e7a fonctionne en Machine Learning ?","og_url":"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement","og_site_name":"Formation Data Science | DataScientest.com","article_published_time":"2024-01-23T16:25:51+00:00","article_modified_time":"2024-01-24T14:04:29+00:00","og_image":[{"width":1024,"height":562,"url":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/Illu_BLOG__SARSA.png","type":"image\/png"}],"author":"J\u00e9r\u00e9my Robert","twitter_card":"summary_large_image","twitter_misc":{"\u00c9crit par":"J\u00e9r\u00e9my Robert","Dur\u00e9e de lecture estim\u00e9e":"4 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement","url":"https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement","name":"SARSA : Comment \u00e7a fonctionne en Machine Learning ?","isPartOf":{"@id":"https:\/\/datascientest.com\/#website"},"datePublished":"2024-01-23T16:25:51+00:00","dateModified":"2024-01-24T14:04:29+00:00","author":{"@id":"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925"},"description":"D\u00e9couvrez tout ce qu'il y a \u00e0 savoir sur le SARSA ou State-Action-Reward-State-Action, qui est un algorithme d\u2019apprentissage par renforcement.","inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datascientest.com\/sarsa-ou-apprentissage-par-renforcement"]}]},{"@type":"WebSite","@id":"https:\/\/datascientest.com\/#website","url":"https:\/\/datascientest.com\/","name":"Formation Data Science | DataScientest.com","description":"Leader de la formation en Data Science","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datascientest.com\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"fr-FR"},{"@type":"Person","@id":"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925","name":"J\u00e9r\u00e9my Robert","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/datascientest.com\/#\/schema\/person\/image\/","url":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg","contentUrl":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg","caption":"J\u00e9r\u00e9my Robert"},"description":"J\u00e9r\u00e9my Robert est actuellement Directeur p\u00e9dagogique B2B chez DataScientest.com. Avant de rejoindre DataScientest, il a travaill\u00e9 en tant que Consultant Business Intelligence chez DEAPDATA, o\u00f9 il a g\u00e9r\u00e9 des projets dans les domaines de l'assurance, bancaire et industrie. Il a \u00e9galement \u00e9t\u00e9 Co-fondateur et pr\u00e9sident de Tripzer. En tant que consultant ind\u00e9pendant, J\u00e9r\u00e9my a propos\u00e9 des formations sur MicroStrategy pour divers niveaux professionnels. Il est dipl\u00f4m\u00e9 de l'Universit\u00e9 Paris Cit\u00e9 avec un Master MIAGE, combinant comp\u00e9tences en informatique et gestion.","sameAs":["https:\/\/www.linkedin.com\/in\/jeremyrobert1\/","Data Science"],"url":"https:\/\/datascientest.com\/author\/jeremyrobert"}]}},"_links":{"self":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/223684"}],"collection":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/users\/78"}],"replies":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/comments?post=223684"}],"version-history":[{"count":0,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/223684\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media\/223686"}],"wp:attachment":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media?parent=223684"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/categories?post=223684"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/tags?post=223684"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}