{"paper_id":"2307.01599","title":"A Scalable Reinforcement Learning-based System Using On-Chain Data for Cryptocurrency Portfolio Management","authors":["Zhenhan Huang","Fumihide Tanaka"],"paper_date":"2023-07-04","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2307.01599v1","page_url":"https://thequant.space/flowcharts/a-scalable-reinforcement-learning-based-system-using-on-chai/","scores":{"empirical_rigor":8,"math_complexity":3.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes CryptoRLPM, a novel reinforcement learning system that incorporates on-chain data for end-to-end cryptocurrency portfolio management.","quote":"Therefore, in this study, we propose CryptoRLPM, a novel RL-based system incorporating on-chain data for end-to-end crypto PM.","summary":"On-chain data (metrics) of blockchain networks, akin to company fundamentals, provide crucial and comprehensive insights into the networks. Despite their informative nature, on-chain data have not been utilized in reinforcement learning (RL)-based systems for cryptocurrency (crypto) portfolio…","strategy":"CryptoRLPM"},"assumptions":{"status":"available","items":[{"assumption":"The Efficient Market Hypothesis holds for blockchain crypto valuations, i.e., prices reflect all accessible information including on-chain metrics.","type":"market","stated":"explicit","quote":"As per the Efficient Market Hypothesis (EMH) [4], a blockchain’s crypto valuation presumably reflects all accessible information, including on-chain metrics."},{"assumption":"On-chain metrics are effective for each cryptocurrency when tested and specified individually.","type":"data","stated":"explicit","quote":"On-chain metrics are tested and specified for each cryptocurrency, overcoming the issue of metric ineffectiveness."}]},"datasets":{"status":"available","items":[{"name":"Historical price data","provider":"Binance","kind":"market-prices","access":"public","period":null,"universe":"cryptocurrencies underlying the portfolios (BTC, STORJ, BLZ, LINK)","quote":"(DFU) retrieves historical price data and on-chain metrics using Binance REST API and Santiment’s SanAPI (SanAPI Basic Subscription) respectively","known_id":"binance"},{"name":"On-chain metrics","provider":"Santiment (SanAPI Basic Subscription)","kind":"crypto-onchain","access":"commercial","period":null,"universe":"cryptocurrencies underlying the portfolios (BTC, STORJ, BLZ, LINK)","quote":"(DFU) retrieves historical price data and on-chain metrics using Binance REST API and Santiment’s SanAPI (SanAPI Basic Subscription) respectively"},{"name":"CM(training) dataset","provider":null,"kind":"synthetic","access":"unknown","period":"2020 Oct–2021 Dec","universe":null,"quote":"CM(training) from October 2020 to December 2021"},{"name":"CM(validation) dataset","provider":null,"kind":"synthetic","access":"unknown","period":"2022 Jan–2022 Feb","universe":null,"quote":"CM(validation) from January 2022 to February 2022"},{"name":"CM(backtesting) dataset","provider":null,"kind":"synthetic","access":"unknown","period":"2022 Mar–2022 Sept","universe":null,"quote":"CM(backtesting) from March 2022 to September 2022"}],"known":[{"id":"binance","name":"Binance","provider":"Binance","kind":"market-prices","access":"public","quote":"n weights obtained from all 𝑚CMs (cryptos) of the portfolio at time step 𝑡: 𝑃𝑡= n\u0010 𝑝1 𝑡, . . . , 𝑝𝑚 𝑡 \u0011 | 𝑝𝑖 𝑡∈R2 for every 𝑖∈{1, . . . ,𝑚} o , (1) and the voted weight 𝑤𝑡will be formalized as the reallocation weight of the portfolio at time step 𝑡 𝑤𝑡= Í𝑑𝑖 𝑡 𝑚, for 𝑖∈{1, . . . ,𝑚} (2) The formalized reallocation weight 𝑤𝑡of the portfolio is transformed into the format required by the designated exchange’s API (e.g., ","mentions":2,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":"crypto","frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"none_found","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"none","rule":"R0b","candidates":{"front_matter":"unclear","evidence_card":"no","evidence_card_quoted":false},"meaning":"no cost treatment was found in the passages the extractor read; that is not evidence that costs are not modeled"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"CryptoRLPM outperforms all baselines across three portfolios, achieving at least an 83.14% improvement in accumulated rate of return, a 0.5603% improvement in daily rate of return, and a 2.1767 improvement in Sortino ratio compared to Bitcoin.","kind":"performance","quote":"CryptoRLPM achieves at least a 83.14% improvement on ARR, a 0.5603% improvement on DRR, and a 2.1767 improvement on SR."},{"finding":"Backtesting shows CryptoRLPM yields positive ARR, DRR, and SR values for all portfolios, while baseline cryptos produce negative ARR.","kind":"performance","quote":"CryptoRLPM consistently outperforms the baselines across all three portfolios, achieving positive values in terms of ARR, DRR, and SR, while the baselines yield negative values on ARR."}],"reported_performance":{"strategy":"CryptoRLPM","sharpe":null,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":null},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":18,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:25:06Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}