Be more concrete: take a quantum circuit. You can try to make it a causal model just like a classical circuit. I think (but am unsure and am confused) that this just works.
How do I interpret this circuit (which produces an entangled bell state) as a causal diagram? Well, we have nodes for each input, the H gate has an arrow pointing in from the first input, the CNOT takes in both inputs (that is, the output from H and the second input) and spits out both outputs. And oh, hey, look at that, the outputs are d-connected (aka not independent).
Measurements are a thing that circuits encode. But, if you think measurements aren’t fundamental, then all measurements can be represented as some black box usual quantum interactions with a bunch of extra stuff, and it’s just that the black box doesn’t work so nice in the causal diagram (because e.g. in this circuit for teleportation you have to view the two measurement boxes as really being one measurement box that takes in both inputs and spits out both outputs. I think you should be able to replace any measurement boxes in your circuit except for the final one going into an observer with some other local operator—or maybe with nothing?
Be more concrete: take a quantum circuit. You can try to make it a causal model just like a classical circuit. I think (but am unsure and am confused) that this just works.
How do I interpret this circuit (which produces an entangled bell state) as a causal diagram? Well, we have nodes for each input, the H gate has an arrow pointing in from the first input, the CNOT takes in both inputs (that is, the output from H and the second input) and spits out both outputs. And oh, hey, look at that, the outputs are d-connected (aka not independent).
Measurements are a thing that circuits encode. But, if you think measurements aren’t fundamental, then all measurements can be represented as some black box usual quantum interactions with a bunch of extra stuff, and it’s just that the black box doesn’t work so nice in the causal diagram (because e.g. in this circuit for teleportation you have to view the two measurement boxes as really being one measurement box that takes in both inputs and spits out both outputs. I think you should be able to replace any measurement boxes in your circuit except for the final one going into an observer with some other local operator—or maybe with nothing?