BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:[HYBRID] Quantum Computing Seminar: Using Landscape Geometry and C
 haos Theory to Forecast Quantum Advantage
DTSTART:20260624T110000Z
DTEND:20260624T120000Z
DTSTAMP:20260624T055400Z
UID:indico-event-401@events.it4i.cz
CONTACT:training@it4i.cz
DESCRIPTION:Annotation\nCan we look at the mathematical shape and chaos of
  a quantum circuit's landscape before training it\, and know immediately i
 f the algorithm is going to learn? In this talk\, we answer this question 
 by introducing a novel framework that uses landscape geometry and chaos th
 eory to forecast quantum advantage in Variational Quantum Algorithms (VQAs
 ). Moving away from traditional accuracy benchmarking\, we analyse the co
 st function manifold through two complementary lenses: Spectral Hessian A
 nalysis to map local geometric ruggedness\, and Lyapunov Exponents to qu
 antify the chaoticity of optimisation trajectories. Together\, these tool
 s diagnose catastrophic training bottlenecks—such as Barren Plateaus and
  Information Scrambling—at "Epoch Zero." Backed by massive state-vector 
 simulations on the Karolina IT4I HPC cluster\, we show how these topologic
 al properties serve as predictive indicators of an algorithm's viability. 
 This research establishes a quantitative pipeline to determine problem-to-
 algorithm suitability\, ensuring we deploy quantum resources only where ad
 vantage is mathematically viable.\nBenefits for the attendees\, what they 
 will learn:\n\nThe Takeaway: Attendees will learn how to evaluate the trai
 nability of a variational quantum ansatz before wasting valuable GPU/QPU h
 ours on optimisation loops.\nThe Mechanism: Understanding how to use initi
 al state geometry to identify catastrophic bottlenecks like Barren Plateau
 s and divergence before the first gradient step is ever taken.\n\nLevel\nB
 eginner - intermediate\nLanguage\nEnglish\nPrerequisites\nBasic knowledge 
 of quantum mechanics and classical machine learning.\n \nTutor\nVan Binh 
 Henri VU is a quantum algorithms researcher in the Quantum Computing labor
 atory at the IT4Innovations National Supercomputer Center (IT4I). He recei
 ved his Ph.D. in condensed matter physics from the University of Paris-Sac
 lay and the French Alternative Energies and Atomic Energy Commission (CEA
 ) in France. He is currently developing quantum computing to tackle real u
 se cases\, and he is actively participating in maintaining the wellness of
  the quantum hardware at IT4I.\n \nAcknowledgements\n\nThis course was su
 pported by the Ministry of Education\, Youth and Sports of the Czech Repub
 lic through the e-INFRA CZ (ID:90254).\n \n \nAll presentations and educ
 ational materials of this course are provided under the Creative Commons A
 ttribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.\n\nhttps:/
 /events.it4i.cz/event/401/
LOCATION:IT4Innovations (ONLINE and onsite)
URL:https://events.it4i.cz/event/401/
END:VEVENT
END:VCALENDAR