The fusion-minimization problem for photonic graph states is formalized as minimum trail cover; most bounded variants are NP-hard, but heuristics plus a TSP reduction give near-optimal fusion counts in benchmarks.
Generating graph states with a single quantum emitter and the minimum number of fusions
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Graph states are the key resources for measurement- and fusion-based quantum computing with photons, yet their creation is experimentally challenging. We optimize a hybrid graph-state generation scheme using a single quantum emitter and linear optics Bell-state measurements called fusions. We first generate a restricted class of states from a single quantum emitter and then apply fusions to create a target graph state, where we use a dynamic programming approach to find the construction that requires the lowest possible number of fusions. Our analysis yields a lookup table for constructing $\sim 2.8\times 10^7$ non-isomorphic graph states with the minimum number of fusions. The lookup table covers all graph states with up to eight qubits and several other ones with up to 14 qubits. We present construction protocols of selected graph states and provide the lookup table. For large graph states that are not in the lookup table, we derive bounds for the required number of fusions using graph-theoretic properties. Finally, we use the lookup table to search for the best graph codes for loss-tolerant encodings, given a fixed number of fusions for their construction.
citation-role summary
citation-polarity summary
fields
quant-ph 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Finding trail covers: near-optimal decompositions of graph states as linear fusion networks
The fusion-minimization problem for photonic graph states is formalized as minimum trail cover; most bounded variants are NP-hard, but heuristics plus a TSP reduction give near-optimal fusion counts in benchmarks.