An open quantum system inevitably decoheres due to unwanted interactions with its dissipative environment. Such decoherence is a primary hurdle for constructing a large-scale quantum computer and much research is spent in “closing the system” i.e. improving qubit relaxation time, T1 and dephasing time, Tf. However, decoherence can also be combatted with engineered dissipative interactions, high-fidelity measurements and appropriately chosen drives in a quantum feedback loop. What is the best method to combine these ingredients to preserve coherence? I will answer this question by contrasting two quantum feedback approaches to stabilizing an entangled Bell state of two superconducting qubits for an arbitrary time [1, 2]. The first, conventional, measurement-based approach uses a classical controller to measure the parity of the two qubits and actively correct errors with conditional gates. In contrast, the second, driven-dissipative approach employs unconditional drives and an engineered interaction with a cold reservoir to passively stabilize the desired state. We show that the driven-dissipative method provides greater robustness to faults and thus a higher fidelity entangled state. Next, we add a second layer of feedback and show that by continuously monitoring the dissipative channel involved in the stabilization, the controller can herald the presence of entanglement in real-time and thus boost its fidelity. These general principles may be applied in the future to stabilize entanglement between remote qubits in a modular quantum computer , as well as to build a fault-tolerant logical qubit holding quantum information with minimal hardware .
 Shankar et al., Nature 504, 419 (2013)
 Liu et al., Phys. Rev. X 6, 011022 (2016)
 Didier et al., arXiv:1703.03379
 Mirrahimi et al., New J. Phys. 16, 045014 (2014)