BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260730T152638Z
LOCATION:
DTSTART;TZID=America/Los_Angeles:20260728T125500
DTEND;TZID=America/Los_Angeles:20260728T125600
UID:dac_DAC 2026_sess315_RESEARCH2805@linklings.com
SUMMARY:Time-Aware Active Calibration and Scheduling for N-Qubit Systems v
 ia Bayesian Optimization and Reinforcement Learning
DESCRIPTION:Ziming Zhao and Tingting Li (Zhejiang University); Xiaofei Yue
  (Beijing Institute of Technology); Zhaoxuan Li (Institute of Computing Te
 chnology, CAS); and Jianwei Yin (Zhejiang University)\n\nFrequent recalibr
 ation of multi-qubit systems is essential to maintain hardware fidelity, b
 ut calibration itself is costly and subject to rapid parameter drift. This
  work presents a novel framework for time-aware active calibration that co
 mbines Bayesian Optimization (BO) and Reinforcement Learning (RL) to adapt
 ively schedule and prioritize calibration experiments under uncertainty. W
 e model the calibration landscape as a directed acyclic graph (DAG), where
  nodes represent hardware parameters (e.g., frequency, amplitude ratio, DR
 AG, or phase) and edges capture inter-parameter dependencies extracted aut
 omatically from Qiskit Experiments metadata. Each parameter's temporal dri
 ft is modeled as an Ornstein-Uhlenbeck (OU) or exponential decay process w
 ith uncertainty bounds. The BO agent selects the next calibration experime
 nt to maximize expected information gain per unit time while respecting sy
 stem-level constraints. A parallel RL/ILP scheduler coordinates cross-qubi
 t calibration and readout tasks to minimize idle time and avoid conflicts.
  \nOur system further incorporates self-triggered recalibration based on o
 nline RB/XEB health metrics, enabling autonomous recovery from drift witho
 ut human intervention. Experimental evaluation on multi-qubit testbeds dem
 onstrates improved calibration efficiency, longer mean time between recali
 brations (MTBC), and reduced overall calibration overhead compared to fixe
 d or purely heuristic scheduling baselines. We provide an open-source impl
 ementation, Cal Orchestrator, which integrates DAG construction, BO-based 
 experiment selection, and RL-based calibration scheduling for scalable qua
 ntum system maintenance.\n\nTopics: Quantum\n\nTrack: DES6. Quantum Comput
 ing, Systems\n\nSession Chairs: Hokeun Kim (Arizona State University); Bae
 kGyu Kim (Daegu Gyeongbuk Institute of Science and Technology); Xiang Chen
  (Peking University); Sangyeob Kim (Yonsei University); Robert M. Radway (
 University of Pennsylvania); Prashanth Mohan (University of Pittsburgh); X
 iaoxuan Yang (University of Virginia, Stanford University); Ramtin Zand (U
 niversity of South Carolina); Hadi Kamali (University of Central Florida);
  Jiafeng Xie (Villanova University); Prabuddha Chakraborty (University of 
 Maine); Junichiro Kadomoto (University of Tokyo); and Yosuke Ueno (RIKEN)\
 n\n
END:VEVENT
END:VCALENDAR