Increasing the noise resistance of Quantum Computers using QuantumNAS

One of the problems that plague quantum computers today is the increased susceptibility they have to ‘noise’: false, unintended signals originating from imperfect control sequences, environmental interference and unwanted qubit interactions that hamper the performance of quantum circuits.

Researchers at MIT are working on creating a new technique to make ‘parameterized’ quantum circuits(circuits with adjustable quantum gates) directly noise resistant. They created a framework, called Quantum Noise Adaptive Search(QuantumNAS) that can identify and generate a mapping pattern for the most robust quantum circuit for a particular task, a framework that happens to be less computationally intensive than most other search methods. It can identify quantum circuits that improve the accuracy of machine learning and quantum chemistry tasks.

Associate Professor Song Han notes that “the key idea here is that, without this technique, we have to sample each individual quantum circuit architecture and mapping scenario in the design space, train them, evaluate them, and if it is not good we have to throw it away and start over. But using this method, we can obtain many different circuits and mapping strategies at once with no need for many times of training."

In designing QuantumNAS, the researchers focused on variational quantum circuits, utilizing quantum gates with trainable parameters benefitting machine learning or quantum chemistry tasks. These circuits are usually either hand-designed or use rule-based methods.

Neither of the ‘traditional’ search methods for optimizing circuits work very well for variational quantum circuits: in the naïve search method, which evaluates possible circuits individually, the parameters for each candidate quantum circuit must be trained, resulting in a massive computational overhead. The classical method of introducing more parameters does not improve accuracy in this case, since this naturally requires more quantum gates, increasing noise.