For online events, details are sent out the week of the event. Join our community to receive them!
From Better Entanglement Distillation to Multi-formalism Quantum Simulation Tools
Krastanov, Stefan - University of Massachusetts Amherst
Presentation on Thursday, Dec. 1, 2022, noon
Location: MIT CUA Room
We will start with a case study of a new algorithm for the simulation of entanglement purification, in order to open a discussion on the, as of yet unmet, need for multi-physics modeling of quantum information systems. We develop an algorithm for simulating this bedrock of quantum networking that is significantly faster than the Clifford stabilizer tableaux formalism (let alone the exponentially-expensive wavefunction formalism). This lets us perform optimizations over otherwise prohibitively-large parameter spaces, constructing the best purification protocols in existence. However, how do we extend these methods naturally to modeling much more diverse systems?
More generally, how do we address the unmet need for "multi-physics" modeling of quantum information processors at the system scale? We know we will need to mix multiple formalisms for the description of the different types of physics involved in the quantum computing stack. Our aforementioned algorithm would be crucial for the modeling of entanglement distribution, but we would also need Clifford circuits for error correction, and wavefunction-based modeling for the low-level analog physics and noise processes. We will discuss the programming principles we have developed in order to run such simulations cohesively, in a manner in which a wide variety of formalisms can work without conflict and without boilerplate, providing full-stack analog/digital models of our hardware.
Bio: Stefan works across the lower layers of the quantum information technology stack. Much of his work focuses on designing robust analog control schemes for the low-level quantum hardware and building noise-resilient digital circuitry on top of them. This enables optimizations across different levels of abstraction, resulting in far better hardware performance than simple concatenation of control protocols. Stefan is a graduate of Liang Jiang's group from the Yale Quantum Institute. The last few years he was a research scientist at the Englund group at MIT, and he is now starting a quantum information science group as an assistant professor at UMass Amherst.