Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about a groundbreaking algorithm for efficiently learning mildly-interacting fermionic unitaries in this 20-minute conference talk from QTML 2025. Discover how researcher Vishnu Iyer addresses a central open question in quantum learning theory by developing the first algorithm capable of learning fermionic unitaries that are "near Gaussian" - those prepared with a small number of non-Gaussian circuit elements. Explore the algorithm's ability to query an n-mode fermionic unitary U prepared by at most O(t) non-Gaussian gates and return a circuit approximating U to diamond distance ε in polynomial time. Understand the introduction of "unitary Gaussian dimension" as a property that generalizes beyond unitaries with limited non-Gaussian gates to include unitaries requiring up to 2^O(t) non-Gaussian gates for construction. Examine the additional polynomial-time algorithm for distinguishing whether an n-mode unitary meets specific Gaussian dimension criteria. Gain insights into structural results about near-Gaussian fermionic unitaries and their applications in quantum chemistry and many-body physics, presented at Singapore's premier quantum machine learning conference.
