Thread by @fgrosshans

Now at #PQC, Nicolas Didier from @rigetti on full stack quantum computing with superconducting qubits
#PQC #LTQI

@rigetti Nicolas Didier gives papers usin Rigetti QCS with VQE to compute ground states of H₂, LiH, NaH and a few other chemistry computations.
You can als solve linear problems by variational quantum lienar eigensolvers, tuning A to optimize A|x⟩=|b⟩
#LTQI #PQC

@rigetti Nicolas Didier work on optimizing the hybrid algorithm temselves e.g. using machine learning #PQC #LTQI

@rigetti Nicolas Didier reduces the latency of hybrid algorithms through CPU/QPU collocation, parametric compilation (compile the quantum circuit once in a tunable form) and active reset. LAttency reduces from 1s to 36 ms #LTQI #PQC

@rigetti Nicolas Didier reduced the qubit from percents to tenth of percent through silicon vias and caps.
"PQC #LTQI

@rigetti Nicolsa Didier increased the coherence time T₁ form 50 to 200 µs by “Improv[ing] a little bit every step of fabrication” #LTQI #PQC

@rigetti Nicolas Didier: capavaitively couple a tunable qubit with a fix ferquecy qubit to make tunable coherent exchange and mage a iSWAP gate. The anhamonicity also allows you make a CZ gate
iSWAP:=
1 0 0 0
0 0 –i 0
0 –i 0 0
0 0 0 1

#LTQI #PQC

Nicolas Didier : playing with square roots of iSWAP, one can make a larger set of gates #PQC #LTQI

Nicolas Didier: This flexibility should allow to reduce the depth of algorithms, since “more native gates are better”. He has started to play around with QAOA variants using this flexibility #LTQI #PQC

Nicolas Dider now looks at the efect of flux noise, and specially at slow flux noise, “the nasty one”. There is an AC sweet spot in flux modulation amplitude Φ₀, minimizing its effect, when the whitenoise is small enough #PQC #LTQI

Nicolas Didier goes to the next level of his sweet spot hunt by playing with more parameters #PQC #LTQI