Filter before You Solve: A Deterministic-First/Learned-Second Architecture for AI-Driven Portfolio Management with Real-Money Training-Investment Calibration
This study investigates a deterministic-first/learned-second AI/ML framework that is deployed in regulated retail brokerage accounts. A two-stage calibration system is employed that combines historical back-testing with live recalibration via continuous position snapshots. The framework includes a unique explainability layer that employs the global sensitivity analysis method, SimDec, to identify the most influential components. SimDec renders every admissible decision fully attributable to a joint state of the input space. Because the attribution is an input to the trigger rather than a commentary on it, no generative step exists in the decision path and hallucination is structurally unreachable and, therefore, absent by construction. An intraday options trading financial framework is illustrated through a live training-investment cycle on long-call positions using “real money”. The major contribution of this research is the overall architectural and methodological framework. The filter-before-you-solve approach enables contributions to be evaluated independently of specific implementations. Beyond financial applications, it is described how the complete architectural pattern can actually generalize to many AI/ML deployment contexts that require auditable deterministic gating prior to learned inference.
References
[1]
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., et al. (2023) Survey of Hallucination in Natural Language Generation. ACMComputingSurveys, 55, Article No. 248. https://doi.org/10.1145/3571730
[2]
Rudin, C. (2019) Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead. NatureMachineIntelligence, 1, 206-215. https://doi.org/10.1038/s42256-019-0048-x
[3]
Melville, G. and Yeomans, J. (2026) Decomposing the Theta Cliff: A SIMDEC Filtering of Asymptotic Time-Decay in Long-Call Options with a Real-Money Intraday Illustration. AI, 7, Article No. 257. https://doi.org/10.3390/ai7070257
[4]
Doshi-Velez, F. and Kim, B. (2017) Towards a Rigorous Science of Interpretable Machine Learning. https://arxiv.org/abs/1702.08608
[5]
Jiang, M., Huang, T., Guo, B., Lu, Y. and Zhang, F. (2025) Enhancing Robustness in Large Language Models: Prompting for Mitigating the Impact of Irrelevant Information. In: Mahmud, M., Doborjeh, M., Wong, K., Leung, A.C.S., Doborjeh, Z. and Tanveer, M., Eds., Neural Information Processing. ICONIP 2024. Communications in Computer and Information Science, Vol. 2295, Springer, 207-222. https://doi.org/10.1007/978-981-96-7030-7_15
[6]
Zhu, D.H., Xiong, Y.J., Zhang, J.C., Xijiong, X. and Xia, C.M. (2025) Understanding before Reasoning: Enhancing Chain-of-Thought with Iterative Summarization Pre-Prompting. https://doi.org/10.48550/arXiv.2501.04341
[7]
Kozlova, M., Ahola, A., Roy, P.T. and Yeomans, J.S. (2025) Simple Binning Algorithm and SimDec Visualization for Comprehensive Sensitivity Analysis of Complex Computational Models. Journal of Environmental Informatics Letters, 13, 38-56.
[8]
Kozlova, M., Moss, R.J., Yeomans, J.S. and Caers, J. (2024) Uncovering Heterogeneous Effects in Computational Models for Sustainable Decision-Making. EnvironmentalModelling&Software, 171, Article ID: 105898. https://doi.org/10.1016/j.envsoft.2023.105898
[9]
Marzban, S. and Lahmer, T. (2016) Conceptual Implementation of the Variance-Based Sensitivity Analysis for the Calculation of the First-Order Effects. JournalofStatisticalTheoryandPractice, 10, 589-611. https://doi.org/10.1080/15598608.2016.1207578
[10]
Sobol, I.M. (2001) Global Sensitivity Indices for Nonlinear Mathematical Models and Their Monte Carlo Estimates. MathematicsandComputersinSimulation, 55, 271-280. https://doi.org/10.1016/s0378-4754(00)00270-6
[11]
Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., et al. (2008) Global Sensitivity Analysis: The Primer. Wiley. https://doi.org/10.1002/9780470725184
[12]
Helo, J., Kozlova, M., Roy, P. and Yeomans, J.S. (2026) Regional Variance-Based Sensitivity Analysis for Complex Models. Risk Analysis.
[13]
Black, F. and Scholes, M. (1973) The Pricing of Options and Corporate Liabilities. JournalofPoliticalEconomy, 81, 637-654. https://doi.org/10.1086/260062
[14]
Hull, J.C. (2018) Options, Futures, and Other Derivatives. 10th Edition, Pearson.
[15]
Carr, P. and Wu, L. (2009) Variance Risk Premiums. ReviewofFinancialStudies, 22, 1311-1341. https://doi.org/10.1093/rfs/hhn038
[16]
Bakshi, G. and Kapadia, N. (2003) Delta-Hedged Gains and the Negative Market Volatility Risk Premium. ReviewofFinancialStudies, 16, 527-566. https://doi.org/10.1093/rfs/hhg002
[17]
Merton, R.C. (1973) Theory of Rational Option Pricing. TheBellJournalofEconomicsandManagementScience, 4, 141-183. https://doi.org/10.2307/3003143
[18]
Schwab Center for Financial Research (2026) Theta Decay in Options Trading: Three Strategies. Charles Schwab. https://www.schwab.com/learn/story/theta-decay-options-trading
[19]
Cao, J. and Han, B. (2013) Cross Section of Option Returns and Idiosyncratic Stock Volatility. JournalofFinancialEconomics, 108, 231-249. https://doi.org/10.1016/j.jfineco.2012.11.010
[20]
Natenberg, S. (2015) Option Volatility and Pricing: Advanced Trading Strategies and Techniques. 2nd Edition, McGraw-Hill.
[21]
Gu, S., Kelly, B. and Xiu, D. (2020) Empirical Asset Pricing via Machine Learning. TheReviewofFinancialStudies, 33, 2223-2273. https://doi.org/10.1093/rfs/hhaa009
[22]
Kolm, P.N., Tütüncü, R. and Fabozzi, F.J. (2014) 60 Years of Portfolio Optimization: Practical Challenges and Current Trends. EuropeanJournalofOperationalResearch, 234, 356-371. https://doi.org/10.1016/j.ejor.2013.10.060
[23]
López de Prado, M. (2018) Advances in Financial Machine Learning. John Wiley & Sons.
[24]
Taleb, N.N. (2007) The Black Swan. The Impact of the Highly Improbable. Random House.
[25]
Markowitz, H. (1952) Portfolio Selection. TheJournalofFinance, 7, 77-91. https://doi.org/10.1111/j.1540-6261.1952.tb01525.x
[26]
Ignizio, J.P. (1976) Goal Programming and Extensions. Lexington Books.
[27]
Huang, L., Yu, W., Ma, W., Zhong, W., Feng, Z., Wang, H., et al. (2025) A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions. ACMTransactionsonInformationSystems, 43, Article No. 42. https://doi.org/10.1145/3703155
[28]
Courant, R. (1943) Variational Methods for the Solution of Problems of Equilibrium and Vibrations. BulletinoftheAmericanMathematicalSociety, 49, 1-23. https://doi.org/10.1090/s0002-9904-1943-07818-4
[29]
Fiacco, A.V. and McCormick, G.P. (1968) Nonlinear Programming: Sequential Unconstrained Minimization Techniques. Wiley. (Reprinted as SIAM Classics in Applied Mathematics 4, 1990)
[30]
Coello, C.A. (2002) Theoretical and Numerical Constraint-Handling Techniques Used with Evolutionary Algorithms: A Survey of the State of the Art. ComputerMethodsinAppliedMechanicsandEngineering, 191, 1245-1287. https://doi.org/10.1016/s0045-7825(01)00323-1
[31]
Asness, C.S., Frazzini, A. and Pedersen, L.H. (2019) Quality Minus Junk. ReviewofAccountingStudies, 24, 34-112. https://doi.org/10.1007/s11142-018-9470-2
Krauss, C. (2016) Statistical Arbitrage Pairs Trading Strategies: Review and Outlook. JournalofEconomicSurveys, 31, 513-545. https://doi.org/10.1111/joes.12153
[34]
Apple Inc. (2026) Form 8-K, Current Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934, Filed with the U.S. Securities and Exchange Com-mission on 20 April 2026, Available via SEC EDGAR.
[35]
Melville, G. and Yeomans, J.S. (2026) Architectural Transferability in Bounded AI: Five Conditions for Transfer from Options Trading to Regulated Decision Domains.
