{"id":12544,"date":"2023-07-06T19:58:17","date_gmt":"2023-07-06T19:58:17","guid":{"rendered":"https:\/\/www.cryptoheretostay.com\/?p=12544"},"modified":"2023-07-06T19:58:17","modified_gmt":"2023-07-06T19:58:17","slug":"scientists-created-a-crypto-portfolio-management-ai-trained-with-on-chain-data","status":"publish","type":"post","link":"https:\/\/cryptoheretostay.com\/?p=12544","title":{"rendered":"Scientists created a crypto portfolio management AI trained with on-chain data"},"content":{"rendered":"<script type=\"text\/javascript\">\r\namzn_assoc_placement = \"adunit0\";\r\namzn_assoc_tracking_id = \"totafreearti-20\";\r\namzn_assoc_ad_mode = \"search\";\r\namzn_assoc_ad_type = \"smart\";\r\namzn_assoc_marketplace = \"amazon\";\r\namzn_assoc_region = \"US\";\r\namzn_assoc_default_search_phrase = \"crypto\";\r\namzn_assoc_default_category = \"All\";\r\namzn_assoc_search_bar = \"false\";\r\namzn_assoc_title = \"\";\r\namzn_assoc_rows =\"1\";\r\n<\/script>\r\n<script src=\"\/\/z-na.amazon-adsystem.com\/widgets\/onejs?MarketPlace=US\"><\/script>\n<p>A pair of researchers from the University of Tsukuba in Japan recently built an AI-powered cryptocurrency portfolio management system that utilizes on-chain data for training, the first of its kind according to the scientists.\u00a0<\/p>\n<p>Called CryptoRLPM, short for \u201cCryptocurrency reinforcement learning portfolio manager,\u201d the AI system utilizes a training technique called \u201creinforcement learning&#8221; to implement on-chain data into its model. <\/p>\n<p>Reinforcement learning (RL) is an optimization paradigm wherein an AI system interacts with its environment \u2014 in this case, a cryptocurrency portfolio \u2014 and updates its training based on reward signals.<\/p>\n<p>CryptoRLPM applies feedback from RL throughout its architecture. The system is structured into five primary units which work together to process information and manage structured portfolios. <\/p>\n<p>These modules include a Data Feed Unit, Data Refinement Unit, Portfolio Agent Unit, Live Trading Unit, and an Agent Updating Unit. <\/p>\n<p><em>Screenshot of pre-print research, 2023 Huang, Tanaka, &#8220;A Scalable Reinforcement Learning-based System Using On-Chain Data for Cryptocurrency Portfolio Management&#8221;<\/em><\/p>\n<p>Once developed, the scientists tested CryptoRLPM by assigning it three portfolios. The first contained only Bitcoin (BTC) and Storj (STORJ), the second kept BTC and STORJ while adding Bluzelle (BLZ), and the third kept all three alongside Chainlink (LINK). <\/p>\n<p>The experiments were conducted over a period lasting from October of 2020 to September of 2022 with three distinct phases (training, validation, backtesting.)<\/p>\n<p>The researchers measured the success of CryptoRLPM against a baseline evaluation of standard market performance through three metrics: \u201caccumulated rate of return\u201d (AAR), \u201cdaily rate of return\u201d (DRR), and \u201cSortino ratio\u201d (SR). <\/p>\n<p>AAR and DRR are at-a-glance measures of how much an asset has lost or gained in a given time period and the SR measures an asset\u2019s risk-adjusted return.<\/p>\n<p><em>Screenshot of pre-print research, 2023 Huang, Tanaka, &#8220;A Scalable Reinforcement Learning-based System Using On-Chain Data for Cryptocurrency Portfolio Management&#8221;<\/em><\/p>\n<p>According to the scientists\u2019 pre-print research paper, CryptoRLPM demonstrates significant improvements over baseline performance:<\/p>\n<p>\u201cSpecifically, CryptoRLPM shows at least a 83.14% improvement in ARR, at least a 0.5603% improvement in DRR, and at least a 2.1767 improvement in SR, compared to the baseline Bitcoin.\u201d<\/p>\n<p><strong><em>Related: <\/em><\/strong><strong><em>DeFi meets AI: Can this synergy be the new focus of tech acquisitions?<\/em><\/strong><\/p>\n<p><script type=\"text\/javascript\">\r\namzn_assoc_placement = \"adunit0\";\r\namzn_assoc_tracking_id = \"totafreearti-20\";\r\namzn_assoc_ad_mode = \"search\";\r\namzn_assoc_ad_type = \"smart\";\r\namzn_assoc_marketplace = \"amazon\";\r\namzn_assoc_region = \"US\";\r\namzn_assoc_default_search_phrase = \"bitcoin\";\r\namzn_assoc_default_category = \"All\";\r\namzn_assoc_search_bar = \"false\";\r\namzn_assoc_title = \"\";\r\namzn_assoc_rows =\"1\";\r\n<\/script>\r\n<script src=\"\/\/z-na.amazon-adsystem.com\/widgets\/onejs?MarketPlace=US\"><\/script><br \/>\n<br \/><a href=\"https:\/\/cointelegraph.com\/news\/scientists-created-crypto-portfolio-management-ai-trained-with-on-chain-data\" target=\"_blank\" rel=\"noopener\">Source<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>A pair of researchers from the University of Tsukuba in Japan recently built an AI-powered cryptocurrency portfolio management system that utilizes on-chain data for training, the first of its kind according to the scientists.\u00a0 Called CryptoRLPM, short for \u201cCryptocurrency reinforcement learning portfolio manager,\u201d the AI system utilizes a training technique called \u201creinforcement learning&#8221; to implement [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":12545,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","_wp_rev_ctl_limit":""},"categories":[2],"tags":[],"class_list":["post-12544","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-bitcoin-news"],"_links":{"self":[{"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/posts\/12544","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=12544"}],"version-history":[{"count":1,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/posts\/12544\/revisions"}],"predecessor-version":[{"id":12546,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/posts\/12544\/revisions\/12546"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=\/wp\/v2\/media\/12545"}],"wp:attachment":[{"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=12544"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=12544"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cryptoheretostay.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=12544"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}