{"paper_id":"2406.16505","title":"Alpha^2: Discovering Logical Formulaic Alphas using Deep Reinforcement Learning","authors":["Feng Xu","Yan Yin","Xinyu Zhang","Tianyuan Liu","Shengyi Jiang","Zongzhang Zhang"],"paper_date":"2024-06-24","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2406.16505v2","page_url":"https://thequant.space/flowcharts/alpha-2-discovering-logical-formulaic-alphas-using/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes Alpha2, a novel deep‑reinforcement‑learning framework that discovers formulaic alphas by treating alpha generation as a program‑construction task.","quote":"In this work, we propose a novel framework for alpha discovery using DRL by formulating the alpha discovery process as program construction.","summary":"Alphas are pivotal in providing signals for quantitative trading. The industry highly values the discovery of formulaic alphas for their interpretability and ease of analysis, compared with the expressive yet overfitting-prone black-box alphas. In this work, we focus on discovering formulaic…","strategy":"Alpha^2"},"assumptions":{"status":"available","items":[{"assumption":"The study assumes a day‑frequency trading strategy over a stock market with n distinct stocks and T trading days.","type":"market","stated":"explicit","quote":"In this study, we focus on finding a day-frequency trading strategy in a stock market consisting of n distinct stocks spanning T trading days."},{"assumption":"Alphas are evaluated primarily by Information Correlation (IC) with the 20‑day return.","type":"statistical","stated":"explicit","quote":"method is to find alphas that have high IC with the 20-day return of the stocks."},{"assumption":"The data is split into training (2009/01/01‑2018/12/31), validation (2019/01/01‑2020/12/31) and test (2021/01/01‑2023/12/31) periods.","type":"data","stated":"explicit","quote":"The dataset is split into a training set (2009/01/01 to 2018/12/31), a validation set (2019/01/01 to 2020/12/31), and a test set (2021/01/01 to 2023/12/31)."},{"assumption":"The universe of stocks is limited to constituents of the CSI300 and CSI500 indices of China A‑shares.","type":"market","stated":"explicit","quote":"We use the constituents of the CSI300 and CSI500 indices of China A-shares as the stock set."},{"assumption":"Alpha combination is performed by training an XGBoost model on the top 20 generated alphas ranked by IC.","type":"model","stated":"explicit","quote":"The XGBoost model is trained to fit the alpha signals to the return signals on the training dataset, using the top 20 generated alphas ranked by IC."},{"assumption":"The back‑test uses a top‑k/drop‑n strategy with k=50 and at most n=5 stocks traded per day, investing evenly across selected stocks.","type":"execution","stated":"explicit","quote":"On each trading day, stocks are first sorted according to the alpha values, then the top‑k stocks are selected to trade. With at most n stocks traded every day, we try to invest evenly across the k stocks. In out experiment, k = 50 and n = 5."}]},"datasets":{"status":"available","items":[{"name":"Chinese A‑shares market via baostock","provider":"baostock","kind":"market-prices","access":"public","period":"2009‑2023","universe":"CSI300 and CSI500 constituents","quote":"... The dataset is split into a training set (2009/01/01 to 2018/12/31), a validation set (2019/01/01 to 2020/12/31), and a test set (2021/01/01 to 2023/12/31). We use the constituents of the CSI300 and CSI500 indices of China A-shares as the stock set.","known_id":"baostock"}],"known":[],"sample":{"start":"2021-01","end":"2023-12","market":"Chinese A-shares (CSI300, CSI500)","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":false,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"no","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Alpha2 identifies a diverse set of logical and effective alphas that significantly improve the performance of the final trading strategy.","kind":"performance","quote":"Empirical experiments on real-world stock markets demonstrates Alpha2’s capability to identify a diverse set of logical and effective alphas, which significantly improves the performance of the final trading strategy."},{"finding":"Alpha2 shows superior back‑test performance on the CSI300 stock market compared to baselines.","kind":"performance","quote":"Our method demonstrates superior performance on the CSI300 stock market."},{"finding":"Alpha2 generates alphas with lower mean correlation than gplearn, leading to a more robust trading strategy.","kind":"performance","quote":"With the more diverse and better alpha set, Alpha2 has greater potential to generate a more robust trading strategy."}],"reported_performance":{"strategy":"Alpha^2","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":16,"citations":{"cited_by_in_corpus":2,"by_independent_authors":2,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":true,"code_url":"https://github.com/x35f/alpha2","note":null},"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":"2026-11-14T18:07:15Z"},"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/"}