Financial Context-Rich Strategy for Actor-Critic based Agents: A Deep Reinforcement Learning Approach for Stock Trading
Resumo indisponível. Consulte o paper original.
PAPER / ARXIV:2609.20538
Zhu, Chang
RESUMO
An AI evaluation can be perfectly reproducible and still support the wrong claim. This risk is acute in closed-loop systems: policy determines visited states, observable components, and which failures leave a measurable trace. We propose a claim-safe protocol with three actions. Refuse: abstain when a clean reference stream or matched runtime comparison lacks support. Decompose: report protocol execution, operational false admission, and structural hypotheses separately rather than as one PASS/FAIL label. Refresh: treat distribution-shift alarms as requests to invalidate and recompute reference map, as not fault evidence. We instantiate the protocol on an aggregate-only simulator with 24 policy components, three demand regimes, two fault-mask families, independent development and heldout seeds. The preregistered heldout contains 1,440 cases and 21,600 partition rows. Only 55/72 regime-component units were reference-admitted and 54/55 remained runtime-admitted, making abstention part of the result. Stable false admission was 0/20 represented as one-sided exact 95% upper bound of 0.1391 under a frozen 0.20 rule. Within admitted units, clean affected traffic outpredicted nominal fault-cell fraction: across 540 unit-arm rows in nested 20 component clusters, the cell-minus-traffic negative-log-likelihood difference was 0.1264 nats per row, with component-cluster 95% interval [0.0593, 0.1918]. A drift log shows why null must be reference-relative: clean fault-null streams triggered 15/15, 0/15, 14/15 alarms while only the middle regime detector matched reference. Rather than universal threshold, we contribute an executable contract linking observable support, statistical calibration, and justified claims.
NO MESMO MAPA
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.