AIMLSystems 2025 · IEEE XplorePUBLISHED
Adaptive Quantum-Enhanced Learning (NISQ Era)
AUTHOR // KUSHAGRA GOYAL·NISQ & SYSTEM LABS
INFORMATION
- TypeResearch Paper
- VenueAIMLSystems 2025 · IEEE Xplore
- Index StatusPUBLISHED
CONTRIBUTIONS
- Real-time classical-quantum feedback loop for parameter tuning on noisy hardware.
- Novel noise-resilient cost-function formulation.
- Empirical validation on actual NISQ processors showing improved classification convergence.
ABSTRACT.
This paper presents a hybrid classical-quantum framework designed for robust Quantum Machine Learning (QML) on Noisy Intermediate-Scale Quantum (NISQ) devices. By implementing adaptive error mitigation protocols and regularizing the parameter updates, we significantly reduce the effects of decoherence and gate noise, yielding a +14% relative performance improvement over standard QML baselines in classifying complex high-dimensional datasets.
METHODOLOGY.
We leverage a Parameterized Quantum Circuit (PQC) coupled with real-time classical optimization. An adaptive error-mitigation layer computes expectations and adjusts classical optimizer parameters dynamically using hardware-specific noise models.
CONFERENCE PRESENTATION GALLERY.
Photo 1AIMLSystems 2025
Photo 2AIMLSystems 2025