Can intermediate state resources guide an STP compiler?
Schedule a finite library of STP state conversions, reuse states from unsuccessful attempts, and bound the resulting compilation cost.
Discuss this questionFix a finite library of invariant encoded resource states and STP conversion protocols, each with specified measurement outcomes, conditional state changes, trace-distance error, storage requirements, and probabilities. Given a target classical-output circuit and accuracy , construct a uniform adaptive scheduler that prepares and consumes these resources while retaining useful failure branches. Bound the total expected and high-probability costs, including fresh mixed qubits, measurements, feed-forward, and retries, and compare with a specified gate-by-gate STP compiler.
The chosen library and cost model are part of the proposal. Arbitrary physical target states are unavailable from mixed input under bare invariant real STP; logical targets must specify their encoding.
Why this is interesting
The mixed-state figure shows that entropy, entanglement, symmetry, and branch probability evolve differently. A compiler needs the operational usefulness of the intermediate state and the cost of recovering from each outcome.
Explicit variants
An interaction calculus for prepared pieces
Tabulate singlet–singlet, singlet–triplet, triplet–triplet, and singlet–mixed-qubit conversions, including both conditional states and probabilities. Extend to symmetric blocks or labeled coupling trees. Young diagrams and spin labels must be accompanied by multiplicity correlations: joining blocks is not a classical operation on labels alone.
A place to start
Build a small conversion library for singlets, symmetric blocks, and encoded multiplicity states. Check its branches exactly, then compare independent preparation with recycling on a repeated primitive. Use known coupling-tree preparation procedures as a baseline.
Related work
These works provide context and established results. They need not be the origin of this exact formulation.
Discuss this question
Reading is public. Posting requires a GitHub account. You can also write to Kunal.