Stay up to date with the latest academic and research topics in the quantum community by joining our live discussions every Friday at 12PM EDT. Tune in to gain insights from experts and engage with a community of quantum enthusiasts!
Curated by: Qiskit (180 videos)
Episode 152 The success of modern deep learning hinges on the ability to train neural networks at scale. Through clever reuse of intermediate information, backpropagation facilitates training through gradient computation at a total cost roughly proportional to running the function, rather than incurring an additional factor proportional to the number of parameters - which can now be in the trillions. Naively, one expects that quantum measurement collapse entirely rules out the reuse of quantum information as in backpropagation. But recent developments in shadow tomography, which assumes access to multiple copies of a quantum state, have challenged that notion. In this talk, we will investigate the feasibility of achieving backpropagation scaling for parameterized quantum models, which is essential for their use at scale, by drawing on results from shadow tomography. Amira is a postdoctoral researcher at the University of Amsterdam, as well as QuSoft, a quantum computing research institution in the Netherlands. She was previously an intern at Google Quantum AI and a predoc researcher at IBM. She holds a Ph.D. in quantum computing from the University of KwaZulu-Natal, during which she was a recipient of Google's Ph.D. fellowship and the Oppenheimer Memorial Trust award.