A Three-Phase Foundation Model for Tax-Aware Personalized Portfolio Management
Quick Answer
This paper introduces a three-phase deep reinforcement learning model for personalized portfolio management, addressing ticker lock-in, monolithic objectives, and static user models.
Quick Take
It employs a T5-based time series model for asset encoding, a Mixture of Experts architecture for diverse investment goals, and a personalized inference layer using transaction history, marking a significant advancement in financial AI applications.
Key Points
- Phase 1 utilizes a ticker-identity-free encoder with a T5-based time series model.
- Phase 2 employs a Mixture of Experts architecture for six distinct investment goals.
- Phase 3 personalizes investment strategies using a 76-parameter LoRA module.
- The model eliminates cross-objective gradient conflicts during training.
- Investment objectives are inferred from actual trading behavior, not questionnaires.
Paper Resources
📖 Reader Mode
~2 min readAbstract:We present a three-phase deep reinforcement learning system for personalized portfolio management that addresses three limitations shared by all prior financial RL work: 1) ticker lock-in, 2) monolithic objectives , and 3) static user models. Phase 1 pretrains a ticker-identity-free cross asset encoder via self-supervised learning on a multi-asset corpus, augmented by a frozen parallel branch using Chronos, a T5-based time series foundation model, fused via a learned gating mechanism. To our knowledge, this is the first application of a time series foundation model to portfolio management RL. The encoder generalizes to any publicly traded asset via a 50-dimensional observable metadata vector that requires no retraining for new tickers. Phase 2 fine-tunes a MoE (Mixture of Experts) portfolio actor critic with PPO under an objective-conditioned reward that simultaneously serves six distinct investment goals sampled per episode: short-term alpha, short-term gain, long-term gain, capital preservation, tax-loss harvesting, and long-term-gains-only. A MoE architecture assigns each objective to a specialized expert head (momentum, growth, defensive, tax-aware), and a learned intent router blends experts based on the active objective and current market regime, which eliminates cross-objective gradient conflict. Phase 3 adds a lightweight personalization layer further adapted at inference time to each individual via a 76-parameter LoRA module fine-tuned on real brokerage transaction history, inferring investment objectives from revealed trading behavior rather than questionnaires. A natural language intent parser converts free-form goals directly into structured investment objective parameters.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.30997 [cs.AI] |
| (or arXiv:2606.30997v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2606.30997 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ramin Pishehvar [view email]
[v1]
Tue, 30 Jun 2026 00:19:59 UTC (42 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
HOBA (Hierarchical On-policy Bidding Agents) is a novel hierarchical reinforcement learning framework that enhances online advertising bidding systems by improving adaptability and reducing hyperparameter tuning costs. It utilizes a for hyperparameter inference, a SARSA agent for expert model selection, and a dynamic expert pool for bid execution, achieving a +3.6% increase in target cost during large-scale deployment and outperforming state-of-the-art baselines on AuctionNet.