{"paper_id":"2511.07434","title":"RL-Exec: Impact-Aware Reinforcement Learning for Opportunistic Optimal Liquidation, Outperforms TWAP and a Book-Liquidity VWAP on BTC-USD Replays","authors":["Enzo Duflot","Stanislas Robineau"],"paper_date":"2025-10-30","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2511.07434v1","page_url":"https://thequant.space/flowcharts/rl-exec-impact-aware-reinforcement-learning-for-opportunist/","scores":{"empirical_rigor":8.5,"math_complexity":6,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes RL-Exec, a PPO reinforcement‑learning agent that learns impact‑aware opportunistic liquidation policies using depth‑20 LOB features and microstructure indicators.","quote":"We present RL-Exec, a PPO agent trained on historical replays augmented with endogenous transient impact (resilience), partial fills, maker/taker fees, and latency.","summary":"We study opportunistic optimal liquidation over fixed deadlines on BTC-USD limit-order books (LOB). We present RL-Exec, a PPO agent trained on historical replays augmented with endogenous transient impact (resilience), partial fills, maker/taker fees, and latency. The policy observes depth-20 LOB…","strategy":"RL-Exec (PPO agent for opportunistic liquidation)"},"assumptions":{"status":"available","items":[{"assumption":"The liquidation task is sell‑only with a fixed residual target inventory.","type":"execution","stated":"explicit","quote":"acts under a sell-only inventory constraint to reach a residual target."},{"assumption":"Evaluation is performed on a single asset (BTC‑USD) and a single venue.","type":"market","stated":"explicit","quote":"Our study is intentionally narrow to make inference clean, but that comes with caveats. First, the evaluation is single‑asset (BTC‑USD), single‑venue, sell‑only liquidation..."},{"assumption":"Impact is modeled as a parametric transient (resilience) process in a replay‑with‑impact simulator.","type":"model","stated":"explicit","quote":"augmented with endogenous transient impact (resilience), partial fills, maker/taker fees and latency."},{"assumption":"Maker/taker fees and latency are fixed and known.","type":"execution","stated":"explicit","quote":"we assume fixed maker/taker fees and latency, and a constrained action space consistent with the sell‑only task."},{"assumption":"Training and testing periods are strictly time‑split (Jan‑2020 train, Feb‑2020 test).","type":"statistical","stated":"explicit","quote":"Training covers January 2020 (31 days) and testing covers February 2020 (28 days)."},{"assumption":"Baseline schedules (TWAP and VWAP‑like) are executed on exactly the same timestamps and cost model as the RL agent.","type":"other","stated":"explicit","quote":"proxy are executed on exactly the same timestamps and under the same fee and transient‑impact/resilience settings as the agent, ensuring cost and path parity."}]},"datasets":{"status":"available","items":[{"name":"BTC‑USD limit‑order‑book data","provider":"Jonathan Sadighian (SESAMm)","kind":"order-book","access":"unknown","period":"January 2020 – February 2020","universe":"BTC‑USD","quote":"We use BTC‑USD limit‑order‑book data at 1 s resolution with 20 levels per side. Training covers January 2020 (31 days) and testing covers February 2020 (28 days)."}],"known":[],"sample":{"start":"2020-01","end":"2020-02","market":"BTC-USD","asset_class":"crypto","frequency":"intraday"},"data_resolution":"l2-book"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":true,"detail":"Costs include maker/taker fees, latency, and a parametric transient impact (resilience) model; both RL‑Exec and baselines use the same cost model.","quote":"augmented with endogenous transient impact (resilience), partial fills, maker/taker fees and latency.","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Costs include maker/taker fees, latency, and a parametric transient impact (resilience) model; both RL‑Exec and baselines use the same cost model.","card_quote":"augmented with endogenous transient impact (resilience), partial fills, maker/taker fees and latency.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"execution-cost","findings":[{"finding":"RL‑Exec outperforms TWAP and the VWAP‑like proxy on all horizons, with mean daily gaps increasing from +2.25 bps at 1,800 s to +22.96 bps at 7,200 s.","kind":"performance","quote":"RL‑Exec beats both baselines on every horizon, with one‑sided Wilcoxon tests (paired by day) significant after Benjamini–Hochberg correction and 95% bootstrap CIs strictly positive:"},{"finding":"The advantage of RL‑Exec grows monotonically with execution horizon (≈+2–3 bps at 30 min, ≈+7–8 bps at 60 min, ≈+23 bps at 120 min).","kind":"performance","quote":"The advantage monotonically increases with the horizon (≈+2–3 bps at 30 min →≈+8 bps at 60 min →≈+23 bps at 120 min)."},{"finding":"A +2–3 bps improvement corresponds to roughly $200–$300 reduction in implementation shortfall per $1 M sell program, after accounting for spreads, fees, latency, and transient impact.","kind":"other","quote":"A +2–3 bps improvement over a $1,000,000 sell program corresponds to roughly $200–$300 of implementation‑shortfall reduction per program, with spreads, fees, latency, and transient impact already included in our cost model."}],"reported_performance":{"strategy":"RL-Exec (PPO agent for opportunistic liquidation)","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":20,"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":true,"code_url":"https://github.com/Giafferri/RL-Exec","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":"2026-11-14T17:23:13Z"},"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/"}