Hardware Acceleration for Randomized Gate Execution

NISQ era devices can make use of randomized gate sampling to improve results in the near-term. For example, known error mitigation techniques such as randomized compiling and algorithms such as projected quantum kernel methods make use of these. Current control system interfaces often lack the ability to implement such techniques at an optimal rate due to latency and other overheads in configuration of the control stacks, which are rather optimized for repetition of identical programs in a simple loop. Extending such a loop to include randomized gate executions may provide performance improvements to the aforementioned techniques, making them practical for real-world applications. In this talk, control systems suitable for running these methods will be discussed.