Learn¶
Conditioning selects records that already occurred. An intervention changes the physical operation that generates later data. In a quantum laboratory, an intervention can replace an instrument, prepare a new state, or route an incoming system through a selected channel.
A replacement map discards the input and prepares a chosen state. It removes dependence of that output state on the incoming system. This operation can break an incoming correlation. Merely selecting one measurement outcome usually does not perform the same physical change.
Measurement can disturb a quantum system. Its causal role must include both the outcome and the state transformation. A classical do expression does not by itself specify that transformation. Write the laboratory operation, its available setting, and its retained outcome before calculating effects.
Practice¶
Solution
After conditioning on outcome 1, B is |1>. After an unselected local replacement, B is still I/2. The local action cannot change the remote unconditional marginal.
Check your understanding¶
Which operation is an input-independent replacement?
A. Keeping only outcome 0
B. Reading the input label
C. Discarding the input and preparing sigma
Show the answer
C. Discarding the input and preparing sigma
A replacement map has the same normalized output for every input. Selecting measurement records is conditioning and generally gives a different joint procedure.
Sources¶
This lesson draws on Costa & Shrapnel, 2016Leifer & Spekkens, 2013Watrous, 2018.
Next: L6-05 · Discovery and informational access
- Costa, F., & Shrapnel, S. (2016). Quantum causal modelling. New Journal of Physics, 18(6), 063032. 10.1088/1367-2630/18/6/063032
- Leifer, M. S., & Spekkens, R. W. (2013). Towards a Formulation of Quantum Theory as a Causally Neutral Theory of Bayesian Inference. Physical Review A, 88(5). 10.1103/PhysRevA.88.052130
- Watrous, J. (2018). The Theory of Quantum Information. https://cs.uwaterloo.ca/~watrous/TQI/