{"paper_id":"2502.02619","title":"Regret-Optimized Portfolio Enhancement through Deep Reinforcement Learning and Future Looking Rewards","authors":["Daniil Karzanov","Rubén Garzón","Mikhail Terekhov","Caglar Gulcehre","Thomas Raffinot","Marcin Detyniecki"],"paper_date":"2025-02-04","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2502.02619v1","page_url":"https://thequant.space/flowcharts/regret-optimized-portfolio-enhancement-through-deep-reinforc/","scores":{"empirical_rigor":6.5,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a novel agent‑based PPO approach to dynamically rebalance and enhance an existing high‑performing 60/40 portfolio using a regret‑based Sharpe reward and a transaction‑cost scheduler.","quote":"Our target is to enhance the traditional 60/40 benchmark (60% stocks, 40% bonds) by employing the Regret‑based Sharpe reward function. To address the impact of transaction fee frictions and prevent signal loss, we develop a transaction cost scheduler.","summary":"This paper introduces a novel agent-based approach for enhancing existing portfolio strategies using Proximal Policy Optimization (PPO). Rather than focusing solely on traditional portfolio construction, our approach aims to improve an already high-performing strategy through dynamic rebalancing…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The environment has no price impact from the agent’s allocations.","type":"execution","stated":"implied","quote":"our environment lacks price impact and remains unaffected by allocations."},{"assumption":"Future return information is unavailable at inference time to avoid data leakage.","type":"data","stated":"implied","quote":"Since we do not have access to future return information during testing, we set the reward to zero in the environment when deploying the trained model."},{"assumption":"Transaction costs are gradually introduced during training via a scheduler.","type":"execution","stated":"explicit","quote":"we introduce a transaction cost scheduler that incrementally raises the training transaction cost at each step according to 𝑇𝐶train(𝑥) = 𝑇𝐶max 𝑆𝑎·𝑥𝑎 if 0 ≤𝑥≤𝑆."},{"assumption":"Synthetic training data generated by circular block bootstrap approximates the real data distribution.","type":"data","stated":"explicit","quote":"we opted for a circular block bootstrap [37] of the underlying training data, applied every 10 episodes."},{"assumption":"The agent can observe a vector of current returns, averages, volatilities, previous weights and transaction costs.","type":"model","stated":"implied","quote":"At each timestep, the agent observes a vector 𝑜𝑡= [𝜇𝑡, 𝛼𝑡, ¯𝜇𝑡−40𝑡, ¯𝜎𝑡−60𝑡, ¯𝑞𝑡−60𝑡, 𝑤𝑡−1, 𝑇𝐶𝑡𝑟𝑎𝑖𝑛(𝑡)] ∈R19."}]},"datasets":{"status":"available","items":[{"name":"Daily price data for assets and index tickers","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"We transform the price data for both asset and index tickers into daily strategy returns, 𝜇𝑡∈R3 and 𝛼𝑡∈R3 respectively..."},{"name":"Synthetic data via circular block bootstrap","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"we opted for a circular block bootstrap [37] of the underlying training data, applied every 10 episodes."},{"name":"60/40 benchmark (60% stocks, 40% bonds)","provider":null,"kind":"benchmark","access":"unknown","period":null,"universe":null,"quote":"Our target is to enhance the traditional 60/40 benchmark (60% stocks, 40% bonds)..."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Costs are ramped up during training and fixed at 0.0025 (0.25%) for evaluation.","quote":"After the ramp limit, the maximal costs, 𝑇𝐶max = 𝑇𝐶eval = 0.0025 (a fair value for traditional brokers and large institutional traders), are applied.","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Costs are ramped up during training and fixed at 0.0025 (0.25%) for evaluation.","card_quote":"After the ramp limit, the maximal costs, 𝑇𝐶max = 𝑇𝐶eval = 0.0025 (a fair value for traditional brokers and large institutional traders), are applied.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The PPO agent outperforms the 60/40 portfolio in return across all periods except the validation set in the pre‑pandemic phase.","kind":"performance","quote":"Our PPO agent outperforms the 60/40 portfolio in terms of return across all periods, except for the validation set in the pre‑pandemic phase."},{"finding":"Regret‑PPO achieves a significantly higher Calmar ratio in most cases, though not in Phase 2.","kind":"performance","quote":"The Calmar ratio is significantly higher for regret PPO in most cases, with the exception of Phase 2."},{"finding":"Embedded drawdown and differential Sharpe reward functions are more conservative, performing well on risk‑adjusted metrics such as Sharpe, Sortino, and MDD.","kind":"performance","quote":"Both the embedded drawdown and differential Sharpe reward functions are more conservative, consistently performing well in risk‑accounting measures such as Sharpe, Sortino, and MDD."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":22,"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":"2026-11-14T17:46:04Z"},"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/"}