English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Portfolio management (PM) is a fundamental financial trading task, whichexplores the optimal periodical reallocation of capitals into different stocksto pursue long-term profits. Reinforcement learning (RL) has recently shown itspotential to train profitable agents for PM through interacting with financialmarkets. However, existing work mostly focuses on fixed stock pools, which isinconsistent with investors' practical demand. Specifically, the target stockpool of different investors varies dramatically due to their discrepancy onmarket states and individual investors may temporally adjust stocks they desireto trade (e.g., adding one popular stocks), which lead to customizable stockpools (CSPs). Existing RL methods require to retrain RL agents even with a tinychange of the stock pool, which leads to high computational cost and unstableperformance. To tackle this challenge, we propose EarnMore, a rEinforcementleARNing framework with Maskable stOck REpresentation to handle PM with CSPsthrough one-shot training in a global stock pool (GSP). Specifically, we firstintroduce a mechanism to mask out the representation of the stocks outside thetarget pool. Second, we learn meaningful stock representations through aself-supervised masking and reconstruction process. Third, a re-weightingmechanism is designed to make the portfolio concentrate on favorable stocks andneglect the stocks outside the target pool. Through extensive experiments on 8subset stock pools of the US stock market, we demonstrate that EarnMoresignificantly outperforms 14 state-of-the-art baselines in terms of 6 popularfinancial metrics with over 40% improvement on profit.

Reinforcement Learning with Maskable Stock Representation for Portfolio Management in Customizable Stock Pools