{"id":8450,"date":"2015-03-07T20:35:17","date_gmt":"2015-03-07T20:35:17","guid":{"rendered":"https:\/\/voxelvid.com\/?p=8450"},"modified":"2025-11-19T18:40:40","modified_gmt":"2025-11-19T18:40:40","slug":"google-deepminds-deep-q-learning-playing-atari-breakout","status":"publish","type":"post","link":"https:\/\/voxelvid.com\/?p=8450","title":{"rendered":"Google DeepMind&#8217;s Deep Q-learning playing Atari Breakout!"},"content":{"rendered":"<p>Google DeepMind created an artificial intelligence program using deep reinforcement learning that plays Atari games and improves itself to a superhuman level. It is capable of playing many Atari games and uses a combination of deep artificial neural networks and reinforcement learning. After presenting their initial results with the algorithm, Google almost immediately acquired the company for several hundred million dollars, hence the name Google DeepMind. Please enjoy the footage and let me know if you have any questions regarding deep learning!<\/p>\n<p>______________________<\/p>\n<p>Recommended for you:<br \/>\n1. How DeepMind&#8217;s AlphaGo Defeated Lee Sedol &#8211; https:\/\/www.youtube.com\/watch?v=a-ovvd_ZrmA&amp;index=58&amp;list=PLujxSBD-JXgnqDD1n-V30pKtp6Q886x7e<br \/>\n2. How DeepMind Conquered Go With Deep Learning (AlphaGo) &#8211; https:\/\/www.youtube.com\/watch?v=IFmj5M5Q5jg&amp;index=42&amp;list=PLujxSBD-JXgnqDD1n-V30pKtp6Q886x7e<br \/>\n3. Google DeepMind&#8217;s Deep Q-Learning &amp; Superhuman Atari Gameplays &#8211; <\/p>\n<div class=\"video-ratio-16-9 whatever-embed-responsive\"><iframe title=\"Google DeepMind&#039;s Deep Q-Learning &amp; Superhuman Atari Gameplays | Two Minute Papers #27\" width=\"640\" height=\"360\" src=\"https:\/\/www.youtube.com\/embed\/Ih8EfvOzBOY?list=PLujxSBD-JXgnqDD1n-V30pKtp6Q886x7e\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe><\/div>\n<p>Subscribe if you would like to see more content like this: http:\/\/www.youtube.com\/subscription_center?add_user=keeroyz<\/p>\n<p>&#8211; Original DeepMind code: https:\/\/sites.google.com\/a\/deepmind.com\/dqn\/<\/p>\n<p>&#8211; Ilya Kuzovkin&#8217;s fork with visualization:<br \/>\nhttps:\/\/github.com\/kuz\/DeepMind-Atari-Deep-Q-Learner<\/p>\n<p>&#8211; This patch fixes the visualization when reloading a pre-trained network. The window will appear after the first evaluation batch is done (typically a few minutes):<br \/>\nhttp:\/\/cg.tuwien.ac.at\/~zsolnai\/wp\/wp-content\/uploads\/2015\/03\/train_agent.patch<\/p>\n<p>&#8211; This configuration file will run Ilya Kuzovkin&#8217;s version with less than 1GB of VRAM:<br \/>\nhttp:\/\/cg.tuwien.ac.at\/~zsolnai\/wp\/wp-content\/uploads\/2015\/03\/run_gpu<\/p>\n<p>&#8211; The original Nature paper on this deep learning technique is available here:<br \/>\nhttp:\/\/www.nature.com\/nature\/journal\/v518\/n7540\/full\/nature14236.html<\/p>\n<p>&#8211; And some mirrors that are not behind a paywall:<br \/>\nhttp:\/\/www.cs.swarthmore.edu\/~meeden\/cs63\/s15\/nature15b.pdf<br \/>\nhttp:\/\/diyhpl.us\/~nmz787\/pdf\/Human-level_control_through_deep_reinforcement_learning.pdf<\/p>\n<p>Web \u2192 https:\/\/cg.tuwien.ac.at\/~zsolnai\/<br \/>\nTwitter \u2192 https:\/\/twitter.com\/karoly_zsolnai<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Google DeepMind created an artificial intelligence program using deep reinforcement learning that plays Atari games and improves itself to a superhuman level. It is capable of playing many Atari games and uses a combination of deep artificial neural networks and reinforcement learning. After presenting their initial results with the algorithm, Google almost immediately acquired the [&hellip;]<\/p>\n","protected":false},"author":0,"featured_media":8451,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"video","meta":{"yasr_overall_rating":0,"yasr_post_is_review":"","yasr_auto_insert_disabled":"","yasr_review_type":"","footnotes":""},"categories":[23],"tags":[2796,2797,2791,2792,2798,2789,2794,2777,2774,2786,2776,2788,2778,2783,2782,2779,2795,2781,2780,2785,2784,2775,2790,2793,2787],"class_list":["post-8450","post","type-post","status-publish","format-video","has-post-thumbnail","hentry","category-ai_tech","tag-ai","tag-ai-games","tag-ai-plays-atari","tag-ai-plays-breakout","tag-ai-plays-game","tag-atari-ai","tag-atari-breakout","tag-breakout","tag-deep-learning","tag-deep-neural-network","tag-deep-q-learning","tag-deep-reinforcement-learning","tag-deepmind","tag-deepmind-ai","tag-deepmind-artificial-intelligence","tag-deepmind-atari","tag-deepmind-breakout","tag-deepmind-google","tag-google-deepmind","tag-google-deepmind-atari","tag-google-deepminds-deep-qlearning-playing-atari-breakout","tag-q-learning","tag-superhuman-atari","tag-two-minute-papers","tag-what-is-deep-q-learning","post_format-post-format-video"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Google DeepMind&#039;s Deep Q-learning playing Atari Breakout! - voxelvid<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/voxelvid.com\/?p=8450\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Google DeepMind&#039;s Deep Q-learning playing Atari Breakout! - voxelvid\" \/>\n<meta property=\"og:description\" content=\"Google DeepMind created an artificial intelligence program using deep reinforcement learning that plays Atari games and improves itself to a superhuman level. It is capable of playing many Atari games and uses a combination of deep artificial neural networks and reinforcement learning. After presenting their initial results with the algorithm, Google almost immediately acquired the [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/voxelvid.com\/?p=8450\" \/>\n<meta property=\"og:site_name\" content=\"voxelvid\" \/>\n<meta property=\"article:published_time\" content=\"2015-03-07T20:35:17+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-11-19T18:40:40+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/voxelvid.com\/wp-content\/uploads\/2025\/11\/private-id-v1eynij0rnk-youtube-a.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1280\" \/>\n\t<meta property=\"og:image:height\" content=\"720\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450\"},\"author\":{\"name\":\"\",\"@id\":\"\"},\"headline\":\"Google DeepMind&#8217;s Deep Q-learning playing Atari Breakout!\",\"datePublished\":\"2015-03-07T20:35:17+00:00\",\"dateModified\":\"2025-11-19T18:40:40+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450\"},\"wordCount\":355,\"commentCount\":22,\"image\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/voxelvid.com\\\/wp-content\\\/uploads\\\/2025\\\/11\\\/private-id-v1eynij0rnk-youtube-a.jpg\",\"keywords\":[\"ai\",\"ai games\",\"ai plays atari\",\"ai plays breakout\",\"ai plays game\",\"atari ai\",\"atari breakout\",\"breakout\",\"deep learning\",\"deep neural network\",\"deep q learning\",\"deep reinforcement learning\",\"deepmind\",\"deepmind ai\",\"deepmind artificial intelligence\",\"deepmind atari\",\"deepmind breakout\",\"deepmind google\",\"google deepmind\",\"google deepmind atari\",\"google deepminds deep qlearning playing atari breakout\",\"q learning\",\"superhuman atari\",\"two minute papers\",\"what is deep q learning\"],\"articleSection\":[\"AI &amp; Tech\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/voxelvid.com\\\/?p=8450#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450\",\"url\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450\",\"name\":\"Google DeepMind's Deep Q-learning playing Atari Breakout! - voxelvid\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/voxelvid.com\\\/wp-content\\\/uploads\\\/2025\\\/11\\\/private-id-v1eynij0rnk-youtube-a.jpg\",\"datePublished\":\"2015-03-07T20:35:17+00:00\",\"dateModified\":\"2025-11-19T18:40:40+00:00\",\"author\":{\"@id\":\"\"},\"breadcrumb\":{\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/voxelvid.com\\\/?p=8450\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#primaryimage\",\"url\":\"https:\\\/\\\/voxelvid.com\\\/wp-content\\\/uploads\\\/2025\\\/11\\\/private-id-v1eynij0rnk-youtube-a.jpg\",\"contentUrl\":\"https:\\\/\\\/voxelvid.com\\\/wp-content\\\/uploads\\\/2025\\\/11\\\/private-id-v1eynij0rnk-youtube-a.jpg\",\"width\":1280,\"height\":720},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/voxelvid.com\\\/?p=8450#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/voxelvid.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Google DeepMind&#8217;s Deep Q-learning playing Atari Breakout!\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/voxelvid.com\\\/#website\",\"url\":\"https:\\\/\\\/voxelvid.com\\\/\",\"name\":\"voxelvid\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/voxelvid.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},false]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Google DeepMind's Deep Q-learning playing Atari Breakout! - voxelvid","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:\/\/voxelvid.com\/?p=8450","og_locale":"en_US","og_type":"article","og_title":"Google DeepMind's Deep Q-learning playing Atari Breakout! - voxelvid","og_description":"Google DeepMind created an artificial intelligence program using deep reinforcement learning that plays Atari games and improves itself to a superhuman level. It is capable of playing many Atari games and uses a combination of deep artificial neural networks and reinforcement learning. After presenting their initial results with the algorithm, Google almost immediately acquired the [&hellip;]","og_url":"https:\/\/voxelvid.com\/?p=8450","og_site_name":"voxelvid","article_published_time":"2015-03-07T20:35:17+00:00","article_modified_time":"2025-11-19T18:40:40+00:00","og_image":[{"width":1280,"height":720,"url":"https:\/\/voxelvid.com\/wp-content\/uploads\/2025\/11\/private-id-v1eynij0rnk-youtube-a.jpg","type":"image\/jpeg"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/voxelvid.com\/?p=8450#article","isPartOf":{"@id":"https:\/\/voxelvid.com\/?p=8450"},"author":{"name":"","@id":""},"headline":"Google DeepMind&#8217;s Deep Q-learning playing Atari Breakout!","datePublished":"2015-03-07T20:35:17+00:00","dateModified":"2025-11-19T18:40:40+00:00","mainEntityOfPage":{"@id":"https:\/\/voxelvid.com\/?p=8450"},"wordCount":355,"commentCount":22,"image":{"@id":"https:\/\/voxelvid.com\/?p=8450#primaryimage"},"thumbnailUrl":"https:\/\/voxelvid.com\/wp-content\/uploads\/2025\/11\/private-id-v1eynij0rnk-youtube-a.jpg","keywords":["ai","ai games","ai plays atari","ai plays breakout","ai plays game","atari ai","atari breakout","breakout","deep learning","deep neural network","deep q learning","deep reinforcement learning","deepmind","deepmind ai","deepmind artificial intelligence","deepmind atari","deepmind breakout","deepmind google","google deepmind","google deepmind atari","google deepminds deep qlearning playing atari breakout","q learning","superhuman atari","two minute papers","what is deep q learning"],"articleSection":["AI &amp; Tech"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/voxelvid.com\/?p=8450#respond"]}]},{"@type":"WebPage","@id":"https:\/\/voxelvid.com\/?p=8450","url":"https:\/\/voxelvid.com\/?p=8450","name":"Google DeepMind's Deep Q-learning playing Atari Breakout! - voxelvid","isPartOf":{"@id":"https:\/\/voxelvid.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/voxelvid.com\/?p=8450#primaryimage"},"image":{"@id":"https:\/\/voxelvid.com\/?p=8450#primaryimage"},"thumbnailUrl":"https:\/\/voxelvid.com\/wp-content\/uploads\/2025\/11\/private-id-v1eynij0rnk-youtube-a.jpg","datePublished":"2015-03-07T20:35:17+00:00","dateModified":"2025-11-19T18:40:40+00:00","author":{"@id":""},"breadcrumb":{"@id":"https:\/\/voxelvid.com\/?p=8450#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/voxelvid.com\/?p=8450"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/voxelvid.com\/?p=8450#primaryimage","url":"https:\/\/voxelvid.com\/wp-content\/uploads\/2025\/11\/private-id-v1eynij0rnk-youtube-a.jpg","contentUrl":"https:\/\/voxelvid.com\/wp-content\/uploads\/2025\/11\/private-id-v1eynij0rnk-youtube-a.jpg","width":1280,"height":720},{"@type":"BreadcrumbList","@id":"https:\/\/voxelvid.com\/?p=8450#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/voxelvid.com\/"},{"@type":"ListItem","position":2,"name":"Google DeepMind&#8217;s Deep Q-learning playing Atari Breakout!"}]},{"@type":"WebSite","@id":"https:\/\/voxelvid.com\/#website","url":"https:\/\/voxelvid.com\/","name":"voxelvid","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/voxelvid.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},false]}},"yasr_visitor_votes":{"stars_attributes":{"read_only":true,"span_bottom":"<div class='yasr-small-block-bold'><span class='yasr-visitor-votes-must-sign-in'>You must sign in to vote<\/span><\/div>"},"number_of_votes":0,"sum_votes":0},"cmb2":{"poster_feature_image":{"vid_m_player_background":"","vid_m_poster_image":"","vid_m_poster_image_id":""},"gif_feature_image":{"vid_m_gif_image":"","vid_m_gif_image_id":""},"video_movie_settings":{"vid_m_director":"","vid_m_actor":"","vid_m_imdb_ratings":"","vid_m_amazon_associates":"","vid_m_vid_tmdb_movie":"","vid_m_vid_tmdb_tv_shows":"","vid_m_vid_tmdb_thumb":""},"vid_ads_m_videoads_settings_post":{"vid_ads_m_video_ads":"","vid_ads_m_video_ads_type":"","vid_ads_m_group_google_ima":"","vid_ads_m_group_image":"","vid_ads_m_group_html5_video":"","vid_ads_m_group_html":"","vid_ads_m_vpaid_mode":"","vid_ads_m_vast_preroll":"","vid_ads_m_vast_postroll":"","vid_ads_m_vast_pauseroll":"","vid_ads_m_vast_midroll":"","vid_ads_m_group_dynamic":"","vid_ads_m_time_to_show_ads":"","vid_ads_m_time_skip_ads":"","vid_ads_m_time_to_hide_ads":""},"feature_image_settings":{"feature_image_position":""},"video_player_settings":{"single_video_style":"","video_player_position":"","vm_video_url_btn_choose":"","vm_video_url_btn_choose_id":"","vm_video_url":"https:\/\/www.youtube.com\/watch?v=V1eYniJ0Rnk","vm_video_ratio":"","vm_video_ratio_mobile":"","video_player_library":"","plyr_player":"","video_streaming":"","vm_video_shortcode":"","vm_video_timelapse":"","vm_media_subtitles":"","vm_media_sources":"","vid_vtt_preview":"","vid_vtt_preview_vtt_file":"","vid_vtt_preview_vtt_file_id":"","vid_vtt_preview_sprite_image":"","vid_vtt_preview_sprite_image_id":"","vid_auto_play":"","vid_preview_mode":"","vm_video_trailer_manl_url":"","vm_screenshot_preview":"","vid_tags":"","vid_p_sync_playlist":"","vid_c_sync_channel":"","get_post_format":"","multiple_links_structure":"","multi_links_title":"","original_video_title":"","multi_links_series":"","vm_video_multi_links":"","vm_duration":"01:43","vid_social_locker":"","vid_locker_mode":"","vid_social_locker_id":"","vid_social_email_locker_id":"","vid_social_signin_locker_id":"","sv_login_to_watch":"","vid_membership_action":"","vm_video_trailer_url":"","vid_download_type":"","vid_download_target":"","vid_woo_product":"","vid_download_mode":"","vid_download_mode_trailer":"","vm_media_download":"","user_rating":"","user_rating_position":"","user_rating_mode":"","user_rating_multi_sets":"","ads_above_single_player":""},"video_auto_fetch_settings":{"vm_video_fetch":""},"video_fake_views_likes":{"vm_fake_views":"","vm_fake_likes":"","vm_fake_dislikes":""},"gallery_settings":{"gallery_position":""},"quote_settings":{"quote_position":""},"post_options":{"main_layout":"","main_logo":"","main_logo_id":"","main_logo_retina":"","main_logo_retina_id":"","main_logo_mobile":"","main_logo_mobile_id":"","main_logo_mobile_retina":"","main_logo_mobile_retina_id":"","sticky_logo":"","sticky_logo_id":"","sticky_logo_retina":"","sticky_logo_retina_id":"","main_nav_layout":"","header_background":"","theme_sidebar":"","single_post_comment_type":"","theme_background":""}},"_links":{"self":[{"href":"https:\/\/voxelvid.com\/index.php?rest_route=\/wp\/v2\/posts\/8450","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/voxelvid.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/voxelvid.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/voxelvid.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=8450"}],"version-history":[{"count":1,"href":"https:\/\/voxelvid.com\/index.php?rest_route=\/wp\/v2\/posts\/8450\/revisions"}],"predecessor-version":[{"id":8452,"href":"https:\/\/voxelvid.com\/index.php?rest_route=\/wp\/v2\/posts\/8450\/revisions\/8452"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/voxelvid.com\/index.php?rest_route=\/wp\/v2\/media\/8451"}],"wp:attachment":[{"href":"https:\/\/voxelvid.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=8450"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/voxelvid.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=8450"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/voxelvid.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=8450"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}