{"paper_id":"2406.08742","title":"DeepUnifiedMom: Unified Time-series Momentum Portfolio Construction via Multi-Task Learning with Multi-Gate Mixture of Experts","authors":["Joel Ong","Dorien Herremans"],"paper_date":"2024-06-13","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2406.08742v1","page_url":"https://thequant.space/flowcharts/deepunifiedmom-unified-time-series-momentum-portfolio-const/","scores":{"empirical_rigor":8.5,"math_complexity":8,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes DeepUnifiedMom, a deep multi‑task learning framework with a multi‑gate mixture‑of‑experts to construct unified time‑series momentum portfolios across multiple speeds.","quote":"We propose an innovative approach to bridge the research gaps identified earlier inspired by recent advancements in deep learning for portfolio construction","summary":"This paper introduces DeepUnifiedMom, a deep learning framework that enhances portfolio management through a multi-task learning approach and a multi-gate mixture of experts. The essence of DeepUnifiedMom lies in its ability to create unified momentum portfolios that incorporate the dynamics of…","strategy":"DeepUnifiedMom(CAN)"},"assumptions":{"status":"available","items":[{"assumption":"The model is trained to maximize the Sharpe Ratio as the objective function.","type":"model","stated":"explicit","quote":"Since the objective of the CAN differs from those of the task-specific networks, we use the Sharpe Ratio (Sharpe, 1994) as the objective function for training."},{"assumption":"A soft‑capped Sharpe Ratio is used to prevent over‑fitting to noisy batches.","type":"model","stated":"explicit","quote":"We introduce a modified objective function called the Sharpe Ratio with a Soft Capping Mechanism to mitigate this risk."},{"assumption":"Transaction costs of 3 basis points are applied to all backtest results.","type":"execution","stated":"explicit","quote":"Table 2: Backtest results (net) for the period from January 2000 to December 2023, with transaction costs set at 3 basis points."},{"assumption":"Continuous futures contracts are constructed using the backward‑ratio method to create a continuous price history.","type":"data","stated":"explicit","quote":"We leverage the continuous contract history of each asset, constructed through end‑to‑end concatenation and price adjustment using the backward‑ratio method"},{"assumption":"An expanding‑window cross‑validation with a 10‑year initial training window and 20% validation split is used for model training.","type":"other","stated":"explicit","quote":"Throughout the training phase, 20% of the data was allocated for validation purposes."}]},"datasets":{"status":"available","items":[{"name":"Pinnacle Data Corp CLC database","provider":"Pinnacle Data Corp","kind":"market-prices","access":"commercial","period":"January 1990 to December 2023","universe":"49 futures contracts across commodities, currencies, fixed income, and equity index future","quote":"The data we used in our experimentation spans from January 1990 to December 2023, providing daily data and encompassing over three decades of historical information."}],"known":[],"sample":{"start":"2000","end":"2023","market":"global equities, fixed income, FX, commodities","asset_class":"multi-asset","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":true,"detail":"transaction costs set at 3 basis points","quote":"Table 2: Backtest results (net) for the period from January 2000 to December 2023, with transaction costs set at 3 basis points.","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"transaction costs set at 3 basis points","card_quote":"Table 2: Backtest results (net) for the period from January 2000 to December 2023, with transaction costs set at 3 basis points.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"DeepUnifiedMom(CAN) achieves a Sharpe Ratio of 1.92 and maximum drawdown of -1.02%, outperforming benchmark TSMOM strategies.","kind":"performance","quote":"DeepUnifiedMom(CAN) 1.92 0.82 2.33 3.81 -1.02"},{"finding":"The unified portfolio weights assigned by DeepUnifiedMom(CAN) remain remarkably consistent over time, diverging from traditional equal‑weighting approaches.","kind":"performance","quote":"Figure 5 shows that the weights assigned by the DeepUnifiedMom(CAN) to the DeepUnifiedMom portfolios —Fast, Medium, and Slow— maintain remarkable consistency throughout the year."},{"finding":"Backtest results using the Sharpe Ratio with Soft Capping mechanism show superior risk‑adjusted performance across all asset classes.","kind":"performance","quote":"Our extensive backtesting, spanning various asset classes such as equity indexes, bonds, currencies, and commodities, has consistently shown that DeepUnifiedMom surpasses benchmarks, maintaining its superior performance even after accounting for transaction costs."}],"reported_performance":{"strategy":"DeepUnifiedMom(CAN)","sharpe":2.33,"sharpe_benchmark":2.31,"annual_return_pct":1.92,"total_return_pct":null,"max_drawdown_pct":1.02},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":43,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"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/joelowj/unified_mom_mmoe","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":"2026-11-14T18:07:45Z"},"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/"}