{"id":1184,"date":"2021-08-04T15:11:51","date_gmt":"2021-08-04T19:11:51","guid":{"rendered":"https:\/\/quantum.ncsu.edu\/ibm-quantum\/?post_type=tribe_events&p=1184"},"modified":"2021-08-31T13:48:08","modified_gmt":"2021-08-31T17:48:08","slug":"nc-state-and-duke-quantum-computing-seminar","status":"publish","type":"tribe_events","link":"https:\/\/quantum.ncsu.edu\/ibm-quantum\/event\/nc-state-and-duke-quantum-computing-seminar\/","title":{"rendered":"NC State and Duke Quantum Computing Seminar"},"content":{"rendered":"
NC State & Duke will host a Collaborative Quantum Computing Seminar Sept 10, 2021!<\/strong><\/p>\n Talk Title:\u00a0 <\/strong>Gate-free state preparation for fast VQE molecular simulations<\/p>\n Speaker:\u00a0 <\/strong>Oinam Romesh Meitei, Postdoctoral Associate, Massachusetts Institute of Technology<\/p>\n Abstract:\u00a0 <\/strong>The variational quantum eigensolver (VQE) is currently the flagship algorithm\u00a0for molecular simulations on near-term quantum computers. The\u00a0algorithm involves implementing a sequence of parameterized gates on quantum\u00a0hardware to generate a target quantum state and measuring the molecular energy.\u00a0The number of gates that can be implemented on current quantum devices remains\u00a0limited mainly due to finite coherence times and frequent gate errors,\u00a0preventing accurate applications to systems with significant entanglement,\u00a0such as strongly correlated molecules.\u00a0In this work, we present an alternate approach (which we refer to as ctrl-VQE)\u00a0where the quantum circuit used for state preparation is removed entirely and\u00a0replaced by a quantum optimal control routine which variationally shapes a\u00a0pulse to drive the initial Hartree-Fock state to the full CI target state.\u00a0The objective function optimized is the ground state molecular energy. By removing the\u00a0quantum circuit, the coherence times required for state preparation can be\u00a0drastically reduced by directly optimizing the pulses at the device level. We\u00a0demonstrate the potential of this method numerically by directly optimizing\u00a0pulse shapes which accurately model the dissociation curves of H2 and HeH+,\u00a0and the ground state energy for LiH. We further present an adaptive pulse\u00a0shaping algorithm to avoid over-parameterization of pulse parameters.<\/p>\n This is a Hybrid Collaborative Event with NC State and Duke. The In Person location for NC State is Venture Place, 2nd Floor, Large Classroom.<\/strong><\/p>\n