SlotLayer 2
Frame an operator as a ground-state question
Given a Hamiltonian reachable as hamiltonian-access, plus the caller's own declaration that the quantity wanted is its lowest eigenvalue rather than its time evolution, return the same operator typed as a ground-state-problem -- the entry `ground-state-energy`, and every method realising it, is written against.
A Hermitian operator reachable as a sum of terms, as sparse-access oracles or as a block-encoding -- exactly what hamiltonian-access already promises -- plus the one thing no prior process can hand over: the caller's own declaration that the quantity being asked for is that operator's lowest eigenvalue rather than, say, its state at a later time.
The identical operator, now typed as a ground-state-problem. Nothing about the Hamiltonian is transformed; the whole content of this step is the declaration attached to it.
This one, drawn
From Hamiltonian you can query to Hamiltonian whose ground state is wanted
A circle is an object you are holding. Each line between the two ends is one recorded way through this slot; where a way is built from smaller slots, those are its own lines. Circles are named on hover, and each one is a link.
Nothing drawn here has a recorded way through it that this figure leaves shut. See it on the map
Why this is a layer
This is the thinnest layer the map draws -- it changes no bit of the operator, only what is declared about it -- and it is still drawn rather than folded into `ground-state-energy`'s own contract, for the reason `ground-state-problem` is a narrower state than `hamiltonian-access` at all: the declaration is not decoration (state-vocabulary.ts says so directly), and a route that has not made it should not be able to walk into this region by accident. Jiang, Kalev, Mruczkiewicz and Neven name exactly this framing, in their own first section, while introducing an unrelated ternary-tree fermion-to-qubit encoding: "fermion-to-qubit mapping is a key ingredient in any quantum simulation protocol, e.g., the variational quantum eigensolver (VQE)" -- one sentence about VQE's own inputs, not about their encoding, which is why it sources a framing step rather than a region. W32 read that same paper's encoding for a competing-methods slot at `hamiltonian-access` and refused it (R2, plans/leona-map-scaling-rules.md); this slot is not that proposal -- it takes a bare Hamiltonian and the caller's own declaration and returns the pair, and the declaration is precisely what an automatic encoding cannot supply. It stands on one method rather than two contested ones, by the owner's own ruling on ai-ops 195 (which closes ai-ops 64's "some problems can be solved using VQE by different framing and preparation of the problem itself"): a slot that only names what a caller already knows they want has nothing to contest the way `ansatz-construction` or `ground-state-energy` itself do, and inventing a second method here would manufacture a competition the paper making the claim does not contain.
Records for what it handles
The catalogue's own entries for the objects on either side of this contract.
It takes: Hamiltonian you can query
- Fermi-Hubbard dimer operator (Jordan-Wigner encoded)
- H₂ molecular qubit Hamiltonian (STO-3G, 2-qubit tapered)
- Heisenberg XXZ spin-chain operator
- Classical Ising Hamiltonian operator
- Bose–Hubbard Hamiltonian
- Electronic-structure Hamiltonian
- Fermi–Hubbard Hamiltonian
- Heisenberg Hamiltonian
- Ising cost Hamiltonian
- Kitaev-chain Hamiltonian
- MaxCut cost operator
- Transverse-field Ising Hamiltonian
- Weighted Pauli-sum Hamiltonian
- XY spin Hamiltonian
- XYZ spin Hamiltonian
- Transverse-field Ising model operator
Ways to do this
1 method recorded
- The mapping paper that names VQE as the reason to frame a ground-state question
Realises `ground-state-framing` on one citation: a ternary-tree fermion-to-qubit mapping paper whose own first section states, in passing, that an efficient encoding matters because a caller downstream -- naming VQE by name -- is going to ask for a ground state. The mapping itself is not drawn here; only the sentence that motivates it is.
Routes that skip this layer
No recorded route avoids this step.
This is a step inside
- Variational quantum eigensolver
Prepare a parameterised trial state on the quantum computer, measure the Hamiltonian's expectation value in it, and let a classical optimiser move the parameters. The quantum computer never runs a long coherent evolution; it runs a short one many times, and the loop closes through a classical number.
In the Atlas
No record in the Atlas covers this yet. The catalogue is circuits and primitives; this part of the literature is not in it.