Co-designing Quantum Processors

On the previous pages, we explored that every gate has some probability of failure. While reducing the error introduced by quantum gates, reduce the number of gates (especially two qubit gates) needed, would significantly reduce the amount of reduction needed for near-term applications.

Quantum computers are application specific by nature as quantum computers outperform their classical counterpart on specific but by far not all problem types.

But this is software, right? How can you reduce the number of gates needed other than optimizing your algorithm? It turns out, you can also optimize your quantum processor to fit the specific needs of your application. This is what IQM Quantum Computers is doing with its unique co-design approach.

Co-design describes Besides building an universal QC, IQM is developing application specific chips..

NMR Paper

Safe and efficient contrast agents for MRI

As you see, with changing the lattice of the circuit and thus reducing the number of SWAP-Gates required, will decrease the performance of the circuit significantly. Remember that this is just an example and there are more ways to make quantum computers more effective in the near term. Things like the There are other options one could (Lego bricks/scratch blocks explanation)

Star architecture chip has high connectivity, minimizes SWAPs. Not practical with regular transmon technology Central resonator acts effectively as a qubit Gate fidelities are similar to transmons Can we enable direct couplings?