{"paper_id":"2412.17293","title":"Multimodal Deep Reinforcement Learning for Portfolio Optimization","authors":["Sumit Nawathe","Ravi Panguluri","James Zhang","Sashwat Venkatesh"],"paper_date":"2024-12-23","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2412.17293v1","page_url":"https://thequant.space/flowcharts/multimodal-deep-reinforcement-learning-for-portfolio-optimiz/","scores":{"empirical_rigor":6,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a reinforcement learning framework that uses multimodal data—historical prices, sentiment, and news embeddings—to optimize trading strategies for S&P 100 stocks.","quote":"We propose a reinforcement learning (RL) framework that leverages multimodal data—including historical stock prices, sentiment analysis, and topic embeddings from news articles—to optimize trading strategies for S&P100 stocks.","summary":"We propose a reinforcement learning (RL) framework that leverages multimodal data including historical stock prices, sentiment analysis, and topic embeddings from news articles, to optimize trading strategies for SP100 stocks. Building upon recent advancements in financial reinforcement learning…","strategy":"SEC+News CNN (Profit)"},"assumptions":{"status":"available","items":[{"assumption":"There are m tradable stocks in the universe and one risk‑free asset, and portfolio weights sum to one.","type":"model","stated":"explicit","quote":"Assume that there are m tradable stocks in our universe and one risk-free asset. The action space A = a ∈Rm+1 |"},{"assumption":"State transitions are deterministic (delta function).","type":"model","stated":"explicit","quote":"The transition function T is a delta function since state transitions are deterministic."},{"assumption":"Sentiment scores decay exponentially over days since the last report.","type":"data","stated":"explicit","quote":"where a represents the company’s sentiment score on the most recent reporting date, t represents the time (in days) between the last report date and the current day, and γ is a constant between 0 and 1 representing the daily decay factor."},{"assumption":"The reward function can be pure profit or a differential Sharpe ratio.","type":"model","stated":"explicit","quote":"One simple reward function is pure profit: R(st, at) is how much profit is gained"}]},"datasets":{"status":"available","items":[{"name":"NewsHeadlines dataset","provider":null,"kind":"text-news","access":"unknown","period":null,"universe":"84 S&P 100 tickers","quote":"Our dataset contains data for 84 out of the total 100 tickers in the S&P 100, and it contains 70,872 entries containing the sentiment embedding of news for a company on a given day."},{"name":"Historical stock price data","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":"S&P 100 stocks","quote":"We propose a reinforcement learning (RL) framework that leverages multimodal data—including historical stock prices, sentiment analysis, and topic embeddings from news articles—to optimize trading strategies for S&P100 stocks."}],"known":[{"id":"crsp","name":"CRSP","provider":"Center for Research in Security Prices","kind":"market-prices","access":"commercial","quote":"11 • OLMAR and WMAMR: Described in Appendix A. • S&P500: The returns of the index, as provided by CRSP (see Section 2).","mentions":2,"used_in":2},{"id":"wrds","name":"WRDS","provider":"Wharton Research Data Services","kind":"other","access":"commercial","quote":"The BUFN Computational Finance Minor program provided us access to Wharton Research Data Services (WRDS), specifically data from the Center for Research in Security Prices (CRSP).","mentions":2,"used_in":1},{"id":"edgar","name":"SEC EDGAR","provider":"U.S. Securities and Exchange Commission","kind":"text-news","access":"public","quote":"We used the EDGAR database [9] to download 10-K and 10-Q SEC filings for S&P100 for the last 30 years.","mentions":1,"used_in":1},{"id":"kaggle","name":"Kaggle","provider":"Kaggle","kind":"other","access":"public","quote":"2.2.1 Daily Financial Headlines Dataset The dataset we use in this project is Daily Financial News for 6000+ Stocks which was downloaded via Kaggle [10].","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":"S&P100 stocks","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"partial","headline_net_of_costs":false,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"partial","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"costs appear only as a robustness check, a flat fee without spread or impact, or outside the headline numbers"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Agents trained with the profit reward consistently achieve superior portfolio performance compared to those using the differential Sharpe ratio.","kind":"performance","quote":"Our principal observations indicate a disparity in learning complexity between the profit reward function and the differential Sharpe ratio for our agent, resulting in consistently superior portfolio performance when optimizing with the former."},{"finding":"CNN‑EIIE and RNN‑EIIE policies outperform MLP under the profit reward, while MLP tends to overfit.","kind":"performance","quote":"agents trained on CNN EIIE and RNN EIIE, exhibit enhanced performance under the profit reward, while the MLP policy network seems to overfit significantly."},{"finding":"Integrating SEC filings data improves performance across both reward functions relative to baseline models.","kind":"performance","quote":"Upon integrating SEC filings data, notable performance enhancements are observed across both reward functions compared to baseline models."},{"finding":"Combining news and SEC data yields the best performance, but news data alone hampers learning of the differential Sharpe ratio due to sparsity.","kind":"performance","quote":"Optimal performance is achieved when simultaneously integrating news and SEC data into the agent’s environment."}],"reported_performance":{"strategy":"SEC+News CNN (Profit)","sharpe":0.5536,"sharpe_benchmark":0.5522,"annual_return_pct":null,"total_return_pct":16.6081,"max_drawdown_pct":19.5472},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":32,"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":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":"2026-11-14T17:48:57Z"},"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/"}