{"id":550,"date":"2019-03-10T22:01:15","date_gmt":"2019-03-10T20:01:15","guid":{"rendered":"http:\/\/www.noxis-studio.com\/blog\/?page_id=550"},"modified":"2019-09-17T23:21:27","modified_gmt":"2019-09-17T21:21:27","slug":"cart-pole","status":"publish","type":"page","link":"http:\/\/www.noxis-studio.com\/blog\/cart-pole\/","title":{"rendered":"Cart pole"},"content":{"rendered":"<p>[vc_row][vc_column][vc_column_text]<\/p>\n<h3>Cart Pole experiment<\/h3>\n<p>This experiment uses <strong>Tensorflow.js<\/strong> in order to solve the cartpole problem using <em>Policy Gradient Reinforcement learning<\/em>.<\/p>\n<ul>\n<li>The Source code is available on <a href=\"https:\/\/github.com\/NoxisStyle\/AI_JS_experiments\/tree\/master\/cart-pole-02\" target=\"_blank\">Github<\/a><\/li>\n<li>Try the experiment in your <a href=\"http:\/\/www.noxis-studio.com\/ai\/cart-pole\/\" target=\"_blank\">browser<\/a><\/li>\n<\/ul>\n<h4>Algorithms<\/h4>\n<p>You can choose between several algorithms :<\/p>\n<ul>\n<li>REINFORCE<\/li>\n<li>REINFORCE with baseline<\/li>\n<li>Actor Critic (A2C)<\/li>\n<\/ul>\n<h4>Technologies<\/h4>\n<ul>\n<li><strong>Tensorflow.js<\/strong> is used for the neural network<\/li>\n<li>The Experiment is rendered using <strong>Phaser 3<\/strong> game engine<\/li>\n<li><strong>Tfjs-vis<\/strong> is used for learning graph visualization<\/li>\n<\/ul>\n<p>&nbsp;[\/vc_column_text][\/vc_column][\/vc_row][vc_row][vc_column][vc_single_image image=&#8221;554&#8243; img_size=&#8221;full&#8221; alignment=&#8221;center&#8221;][\/vc_column][\/vc_row][vc_row][vc_column][vc_column_text]The Settings page allows to:<\/p>\n<ul>\n<li>Configure the Hypermarameters of the Policy and Value models<\/li>\n<li>Store the trained models in the Browser<\/li>\n<li>Retrieve and use stored models<\/li>\n<\/ul>\n<p>[\/vc_column_text][vc_single_image image=&#8221;646&#8243; img_size=&#8221;full&#8221; alignment=&#8221;center&#8221;][\/vc_column][\/vc_row][vc_row][vc_column][vc_video link=&#8221;https:\/\/www.youtube.com\/watch?v=q8lftqQH7O8&#038;&#8221;][\/vc_column][\/vc_row]<\/p>\n","protected":false},"excerpt":{"rendered":"<p>[vc_row][vc_column][vc_column_text] Cart Pole experiment This experiment uses Tensorflow.js in order to solve the cartpole problem using Policy Gradient Reinforcement learning. The Source code is [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"class_list":["post-550","page","type-page","status-publish","hentry"],"acf":{"full_width":false,"header_transparent":false},"aioseo_notices":[],"_links":{"self":[{"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/pages\/550","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/comments?post=550"}],"version-history":[{"count":4,"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/pages\/550\/revisions"}],"predecessor-version":[{"id":647,"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/pages\/550\/revisions\/647"}],"wp:attachment":[{"href":"http:\/\/www.noxis-studio.com\/blog\/wp-json\/wp\/v2\/media?parent=550"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}