Keynote Speaker

Andrew Sornborger

Andrew Sornborger

Los Alamos National Laboratory

About

Andrew Sornborger is a staff scientist at Los Alamos National Laboratory in the Information Sciences Group, CCS-3. He has studied quantum computation for 25 years with publications in near-term and fault-tolerant quantum simulation, quantum machine learning, variational quantum algorithms, quantum error mitigation, and other fault-tolerant quantum algorithms. He is the Algorithms Thrust lead and LANL PI in the Quantum Science Center, an NQI Center, and the ASC Beyond Moore's Law Quantum Computing Portfolio lead at LANL. He has published a wide range of articles in quantum computation in high impact journals, including Physical Review Letters, PRX, PRX Quantum, Nature Communications, and Quantum.

Learnable and Unlearnable in Quantum Machine Learning

With the example of unitary learning via a quantum compiling algorithm, I will discuss a number of interesting discoveries that we have made that impact how quantum compiling can be used. These discoveries include 1) exponential reductions in training data required by swapping training state pairs for entanglement, 2) exponential difficulty in the form of barren plateaus for certain learning applications, 3) good generalization and how that can impact variational quantum simulation methods. I will conclude by discussing the extension of quantum compiling in the discrete case to the continuous variable setting and show some recent experimental results in CV quantum compiling.

View Slides

Questions? Contact us!