Dissertation Defense: Elizabeth Bennewitz
Description
Title:Â Exploring quantum dynamics with analog quantum simulators
Speaker:  Elizabeth Bennewitz (QuICS)
Date & Time:Â Â October 16, 2026, 10:00am
Where to Attend:Â Â ATL 3100A and Virtual Via Zoom: https://umd.zoom.us/j/2305684852?pwd=czBlbjREek1kQVRITEtmZmhlUHg1dz09&omn=95394874358 Meeting ID: 230 568 4852 Passcode: 3fZcEv
Understanding strongly interacting quantum systems is fundamental to physics, yet modeling their most general dynamics remains computationally intractable on classical hardware. Analog quantum simulation offers a powerful alternative: building physical systems—such as trapped ions or superconducting circuits—that directly mimic target dynamics. However, extracting reliable insights from today's noisy hardware requires tailored research questions, custom experimental protocols, and ultra-precise device characterization. This dissertation advances analog quantum processors from proposal to application by: first, formulating an experimental proposal to probe particle scattering on quantum simulators; second, developing methods to learn many-body analog Hamiltonians with high precision; and third, employing state-of-the-art simulators to observe universal random behavior in strongly interacting dynamics.
First, we present an experimental proposal to probe real-time particle scattering dynamics on analog processors, introducing two state-preparation protocols designed for trapped-ion platforms. Using numerical simulations, we benchmark both protocols, demonstrating high-fidelity wave-packet preparation in experimentally accessible regimes and predicting new inelastic scattering regimes with distinct experimental signatures. Together, these results illustrate how developing complex state preparation expands the utility of analog processors for studying particle scattering. Second, we address hardware characterization by developing a multi-parameter Hamiltonian learning protocol for analog processors. We show that the Hamiltonian governing a superconducting analog-digital quantum processor can be learned by optimizing cross-entropy benchmarking (XEB) against experimental snapshots. Implementing this protocol on local sub-regions of a physical device yields an order-of-magnitude improvement in estimated wavefunction fidelity decay rates, and stitching these learned patches together enables scalable learning for larger devices. Finally, we employ an analog simulator to observe emergent universal random behavior in strongly interacting systems. We show that the Scrooge ensemble—an adaptation of the Haar ensemble—provides an accurate, universal description of this emergent randomness. Combining high-fidelity analog time evolution with digital state preparation and readout allows us to measure a broad suite of physical observables, all of which quantitatively agree with Scrooge ensemble predictions. Overall, this dissertation expands the practical utility of analog quantum simulators by proposing expanded state preparation protocols for scattering, introducing robust characterization methods, and uncovering fundamental many-body physics.
*We strongly encourage attendees to use their full name (and if possible, their UMD credentials) to join the zoom session.*
Speaker:  Elizabeth Bennewitz (QuICS)
Date & Time:Â Â October 16, 2026, 10:00am
Where to Attend:Â Â ATL 3100A and Virtual Via Zoom: https://umd.zoom.us/j/2305684852?pwd=czBlbjREek1kQVRITEtmZmhlUHg1dz09&omn=95394874358 Meeting ID: 230 568 4852 Passcode: 3fZcEv
Understanding strongly interacting quantum systems is fundamental to physics, yet modeling their most general dynamics remains computationally intractable on classical hardware. Analog quantum simulation offers a powerful alternative: building physical systems—such as trapped ions or superconducting circuits—that directly mimic target dynamics. However, extracting reliable insights from today's noisy hardware requires tailored research questions, custom experimental protocols, and ultra-precise device characterization. This dissertation advances analog quantum processors from proposal to application by: first, formulating an experimental proposal to probe particle scattering on quantum simulators; second, developing methods to learn many-body analog Hamiltonians with high precision; and third, employing state-of-the-art simulators to observe universal random behavior in strongly interacting dynamics.
First, we present an experimental proposal to probe real-time particle scattering dynamics on analog processors, introducing two state-preparation protocols designed for trapped-ion platforms. Using numerical simulations, we benchmark both protocols, demonstrating high-fidelity wave-packet preparation in experimentally accessible regimes and predicting new inelastic scattering regimes with distinct experimental signatures. Together, these results illustrate how developing complex state preparation expands the utility of analog processors for studying particle scattering. Second, we address hardware characterization by developing a multi-parameter Hamiltonian learning protocol for analog processors. We show that the Hamiltonian governing a superconducting analog-digital quantum processor can be learned by optimizing cross-entropy benchmarking (XEB) against experimental snapshots. Implementing this protocol on local sub-regions of a physical device yields an order-of-magnitude improvement in estimated wavefunction fidelity decay rates, and stitching these learned patches together enables scalable learning for larger devices. Finally, we employ an analog simulator to observe emergent universal random behavior in strongly interacting systems. We show that the Scrooge ensemble—an adaptation of the Haar ensemble—provides an accurate, universal description of this emergent randomness. Combining high-fidelity analog time evolution with digital state preparation and readout allows us to measure a broad suite of physical observables, all of which quantitatively agree with Scrooge ensemble predictions. Overall, this dissertation expands the practical utility of analog quantum simulators by proposing expanded state preparation protocols for scattering, introducing robust characterization methods, and uncovering fundamental many-body physics.
*We strongly encourage attendees to use their full name (and if possible, their UMD credentials) to join the zoom session.*