AI Science Tracker
EN

Real-time reinforcement learning on coupling correction

Shanghai Advanced Research Institute Chinese Academy of Sciences; Shanghai Institute of Applied Physics Chinese Academy of Sciences · 20 Oct 2026 · physics

Discovery

Real-time reinforcement learning on coupling correction

Source check: Source checked. Not independent scientific confirmation.

Finding: APS PDF Methods specify in-situ dueling DQN with six skew-quadrupole actions and live SSRF training (pp. 2–4, 6–9); Results report 0.15 μm RMS versus 0.62 μm baseline over separate 6-hour windows and adaptation after ID/target changes (pp. 7–9). Publication metadata gives 2026-10-20.

Technology: APS PDF Methods specify in-situ dueling DQN with six skew-quadrupole actions and live SSRF training (pp. 2–4, 6–9); Results report 0.15 μm RMS versus 0.62 μm baseline over separate 6-hour windows and adaptation after ID/target changes (pp

Publication: Peer-reviewed paper · Test context: replication or operational use. Catalog categories, not an evidence grade.

AI contribution
APS PDF Methods specify in-situ dueling DQN with six skew-quadrupole actions and live SSRF training (pp. 2–4, 6–9); Results report 0.15 μm RMS versus 0.62 μm baseline over separate 6-hour windows and adaptation after ID/target changes (pp. 7–9). Publication metadata gives 2026-10-20.
Test and result
APS PDF Methods specify in-situ dueling DQN with six skew-quadrupole actions and live SSRF training (pp. 2–4, 6–9); Results report 0.15 μm RMS versus 0.62 μm baseline over separate 6-hour windows and adaptation after ID/target changes (pp. 7–9). Publication metadata gives 2026-10-20. Evaluation: APS PDF Methods specify in-situ dueling DQN with six skew-quadrupole actions and live SSRF training (pp. 2–4, 6–9); Results report 0.15 μm RMS versus 0.62 μm baseline over separate 6-hour windows and adaptation after ID/target changes (pp. 7–9).
Limitation / still unknown
The operational result is limited to the reported SSRF accelerator, control actions and two six-hour comparison windows; performance at other facilities is not established.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://journals.aps.org/prab/pdf/10.1103/49ns-fryv

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 3/5 · Validation 5/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse physics discoveries.