“Adaptive quantum circuits” (also known as dynamic circuits or feedforward circuits) is where a sequence of quantum operations is not fixed in advance but is determined by the outcomes of measurements made during execution. It involves a combination of (a) mid-circuit measurements in one part of a circuit, (b) real-time classical computation on the measurement outcomes, and (c) logic to decide on the application of controlled gates elsewhere in the circuit based on the results of the classical computation.
To support research into feedforward circuits, TensorFlow Quantum should support the ability for mid-circuit measurements to control later operations. This violates theoretical assumptions underlying the common approach of preparing a circuit and doing only terminal measurements, and may require significant work to implement
“Adaptive quantum circuits” (also known as dynamic circuits or feedforward circuits) is where a sequence of quantum operations is not fixed in advance but is determined by the outcomes of measurements made during execution. It involves a combination of (a) mid-circuit measurements in one part of a circuit, (b) real-time classical computation on the measurement outcomes, and (c) logic to decide on the application of controlled gates elsewhere in the circuit based on the results of the classical computation.
To support research into feedforward circuits, TensorFlow Quantum should support the ability for mid-circuit measurements to control later operations. This violates theoretical assumptions underlying the common approach of preparing a circuit and doing only terminal measurements, and may require significant work to implement