Bad Qubits — Learn Quantum by Doing

Multiple choice

According to Cerezo et al. (2021), how does the choice between global and local cost functions affect barren plateaus in shallow QML circuits?

Both global and local cost functions are equally trainable; only circuit depth matters
Global observables (acting on all qubits at once) give plateaus even at shallow depth, whereas local cost functions (readouts on a few qubits) can stay trainable
Global observables stay trainable at all depths, while local cost functions always produce plateaus
Local cost functions require exponentially more qubits than global ones to avoid plateaus