Barriers to dequantization of near-term molecular electronic structure methods
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.
Kai: We're looking at the paper: "Barriers to dequantization of near-term molecular electronic structure methods".
Mira: Here's the gist of the paper. The paper investigates simulation hardness results for quantum circuit families relevant to near-term chemical ground state estimation, specifically focusing on non-orthogonal multi-reference methods like Tensor Network Quantum Eigensolver (TNQE) and No-Reference Quantum Eigensolver (NOQE).
Kai: The core argument establishes that certain quantum ansätze, particularly those involving orbital rotations and fermionic magic state inputs, are not classically simulable in the worst case under the generalized P not equal to NP conjecture.
Mira: A key technical result is Theorem 6, which proves that orbital rotation circuits (G hat) are universal under post-selection when supplied with four-qubit fermionic magic state inputs.
Kai: This universality implies two simulation hardness results: GapP-hardness of closed simulation up to a constant multiplicative factor (Corollary 7) and worst case hardness of classical sampling (Corollary 8), assuming the polynomial hierarchy does not collapse.
Paper summary: Kai: So wrapping up, the paper "Barriers to dequantization of near-term molecular electronic structure methods" really argues that certain quantum circuit families for molecular ground state estimation are classically hard to simulate in the worst case under standard conjectures.
Mira: They focus on non-orthogonal multi-reference methods and show that specific ansätze, like those using orbital rotations and magic state inputs, resist classical simplification through dequantization or sampling approaches.
Lev: The title itself points to the barrier—dequantization—which is trying to turn a quantum problem into a simpler classical one. This paper shows that for the systems we are interested in, that simplification doesn't work well when you use these specific reference states.
Kai: For someone listening who just wants the big picture, it means that if you're building near-term chemistry tools, relying on these specific quantum ansätze gives you a theoretical guarantee against being able to simulate those results efficiently classically in the worst case.
Mira: It’s about showing that for systems with both static and dynamic correlations, we need scalable quantum computers using these types of reference states to get the accurate ground state estimations we're aiming for.
Conclusion: Kai: So we're wrapping up on this paper "Barriers to dequantization of near-term molecular electronic structure methods." It’s basically about showing that these quantum methods used for chemistry can't be easily turned into simple classical calculations in the worst case.
Mira: Exactly. The authors are focusing on non-orthogonal multi-reference methods, like tensor network quantum eigensolvers and no-reference quantum eigensolvers, and they’re proving those certain circuit setups are classically hard to simulate if you use standard assumptions about mathematics not collapsing.
Lev: From an engineering standpoint, that means if you're trying to build a real chemistry tool based on these ideas, you can't just assume some clever classical trick will make it run fast every time. It’s robust against dequantization and sampling methods.
Kai: So what does this mean for the actual hardware we’re building? If the worst case is hard, it suggests that for large systems with both static and dynamic interactions, we really need those scalable quantum wavefunctions you mentioned earlier to get the right answer.
Mira: Right. The paper points out that even with polynomial-depth reference states, there are still specific limits on what you can approximate efficiently. It shows where the current methods fall short when dealing with complex correlation features like electron cusps.
Lev: I see that. It means we can't just assume any simple circuit will work for every system. We need to be more careful about the reference states we prepare and how much noise those preparations introduce before we even start calculating the energy expectation values.
Kai: So the main implication is that achieving accurate ground state estimations for complex molecules will probably still require a quantum computer using these specific, scalable wavefunction ansätze rather than some shortcut classical method.
Mira: That’s the core message. It sets a realistic expectation for what’s achievable with near-term hardware and where we need to focus our efforts in developing better quantum algorithms.
Oskar Leimkuhler, K. Birgitta Whaley
University of California, Berkeley · Berkeley Quantum Information and Computation Center
quant-ph, cond-mat.str-el, physics.chem-ph
Submitted: 2025-03-26
Updated: 2026-11-10
Comments: 36 pages, 8 figures
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 100/100
The gist: The paper investigates simulation hardness results for quantum circuit families relevant to near-term chemical ground state estimation…
Key concepts
- Tensor Network Quantum Eigensolver (TNQE)
- A method used to find the lowest energy state of a molecule by representing the quantum wavefunction as a tensor network. It's relevant for near-term simulations, but its simulation hardness depends on the specific quantum ansätze used.
- Orbital Rotation Circuits ($\hat{G}$)
- These are specific types of quantum circuits that involve rotating orbitals within a system. The paper shows these circuits are universal when combined with four-qubit fermionic magic state inputs, leading to simulation hardness results.
- NOQE Ansatz
- A quantum method used for molecular calculations that does not rely on reference states. When combined with polynomial-depth UCCSD reference states, computing an energy approximation is proven to be BQP-hard, meaning it's hard for classical computers.
- Fermionic Magic State Inputs
- These are specific quantum inputs, involving fermionic magic states, that are crucial for making certain orbital rotation circuits universal. They are key ingredients in establishing the worst-case simulation hardness results.
Terminology
Summary
The paper investigates simulation hardness results for quantum circuit families relevant to near-term chemical ground state estimation, specifically focusing on non-orthogonal multi-reference methods like Tensor Network Quantum Eigensolver (TNQE) and No-Reference Quantum Eigensolver (NOQE). p. 1 A
linear combination of these reference states provides the
ansatz for our recently developed tensor network quantum eigensolver (TNQE) [8], which extends the density
matrix renormalization group (DMRG) [22] optimization
routine beyond the typical constraint of one-dimensional
area law entanglement, and has achieved significantly
more accurate and resource-efficient ground state energy
estimates than a single-reference UCCSD benchmark in
preliminary tests on small molecules. The second class
of reference states is represented by UCCD or related
UCC correlators acting on rotated Slater determinants,
which in linear combination provide the ansatz for the
non-orthogonal quantum eigensolver (NOQE) [7]. Birgitta Whaley†
Department of Chemistry, University of California, Berkeley and
Berkeley Quantum Information and Computation Center,
University of California, Berkeley, CA 94720, USA
(Dated: February 6, 2026)
We prove classical simulation hardness, under the generalized P ̸= NP conjecture, for quantum
circuit families with applications in near-term chemical ground state estimation. The proof exploits
a connection to particle number conserving matchgate circuits with fermionic magic state inputs,
which are shown to be universal for quantum computation under post-selection, and are therefore not
classically simulable in the worst case, in either the strong (multiplicative) or weak (sampling) sense.Where the paper says it
Where the paper says it
A key technical result is Theorem 6, which proves that orbital rotation circuits (G hat) are universal under post-selection when supplied with four-qubit fermionic magic state inputs. p. 1 We present these results as evidence that superpolynomial quantum advantage in quantum chemistry is
theoretically achievable using methods suitable for nearterm hardware, noting that both of these ansatz classes
are designed to compactly represent energy eigenstates
of systems characterized by a combination of static and
dynamic electron correlation. OVERVIEW OF MAIN RESULTS
Our main technical result is to prove that orbital rotation circuits, which we denote by Gˆ, are universal under
post-selection when provided with four-qubit fermionic
magic state inputs (Theorem 6). OVERVIEW OF MAIN RESULTS
Our main technical result is to prove that orbital rotation circuits, which we denote by Gˆ, are universal under
post-selection when provided with four-qubit fermionic
magic state inputs (Theorem 6). From this follow two
simulation hardness results: GapP-hardness of closed
simulation up to a constant multiplicative factor (Corollary 7), as well as worst case hardness of classical sampling (Corollary 8), under the widely held complexity
theoretic assumption that the polynomial hierarchy does
not collapse [2, 23, 24].Where the paper says it
Where the paper says it
These results are applied to quantum non-orthogonal multi-reference methods used for computing off-diagonal matrix elements between reference states. p. 0 The proof exploits
a connection to particle number conserving matchgate circuits with fermionic magic state inputs,
which are shown to be universal for quantum computation under post-selection, and are therefore not
classically simulable in the worst case, in either the strong (multiplicative) or weak (sampling) sense. We apply this result to quantum non-orthogonal multi-reference methods designed for near-term
hardware by ruling out certain dequantization strategies for computing the off-diagonal matrix elements between reference states. In Section V, we show how these results apply to
the off-diagonal matrix element computations in quantum non-orthogonal multi-reference methods, which diagonalize within a subspace of correlated low-depth reference states states expressed in different orbital bases. Corollary 13 now implies that computing an additive
approximation to the expected energy using the multireference NOQE ansatz, with polynomial-depth generalized UCCSD reference states, is BQP-hard (see Section
V B). Corollary 13 now implies that computing an additive
approximation to the expected energy using the multireference NOQE ansatz, with polynomial-depth generalized UCCSD reference states, is BQP-hard (see Section
V B). Furthermore, Theorem 6 and its corollaries now
imply that computing the off-diagonal matrix elements
between non-universal reference states that include orbital rotations, together with sufficient fermionic magic,
are robust against dequantization by tensor network contraction, matchgate simulation, or any sampling-based
approach (see Section V C).Where the paper says it
Where the paper says it
Where the paper says it
The paper concludes that the worst case for any classical simulation algorithm requiring efficient closed simulation of output probabilities up to a multiplicative factor or efficient sampling from probability distributions of these ansätze is ruled out under the generalized P not equal to NP conjecture. p. 3 While
#P and GapP are equivalent under polynomial-time reductions (P
#P = P
GapP), the approximation of a #P sum
up to a multiplicative factor is enabled by Stockmeyer’s
algorithm in BPPNP ⊆ Σ3 [27], whereas approximating a GapP sum up to a multiplicative factor is also a
GapP-hard problem [23]. This is the basis for quantum
speedups in the weak simulation of universal quantum
circuits, as an efficient classical sampler would enable a
multiplicative approximation of the output probabilities
by Stockmeyer’s algorithm in Σ3 [23, 28]. We
also prove BQP-completeness of closed simulation up to a
constant additive error for a more general class of quantum chemistry circuits, namely generalized UCCSD circuits consisting of single and double excitation gates with
polynomial depth in the system size. In the second part, we apply our proof of simulation hardness to quantum non-orthogonal multi-reference
methods for chemical ground state estimation [7, 8, 19,
20], which perform a classical diagonalization of the
Hamiltonian (energy) operator in a subspace of lowdepth reference states that are expressed in different orbital bases, providing a route to accurate computation of
electronic energies for molecular systems possessing both
static and dynamic electron correlations. We apply this result to quantum non-orthogonal multi-reference methods designed for near-term
hardware by ruling out certain dequantization strategies for computing the off-diagonal matrix elements between reference states. We demonstrate these quantum speedups for two choices of ansatz
that incorporate both static and dynamic correlations to model the electronic eigenstates of molecular systems: linear combinations of orbital-rotated matrix product states, which are preparable in
linear depth, and linear combinations of states prepared by generalized UCCSD circuits of polynomial depth, for which computing the expectation values of local fermionic observables up to a
constant additive error is BQP-complete. Furthermore, Theorem 6 and its corollaries now
imply that computing the off-diagonal matrix elements
between non-universal reference states that include orbital rotations, together with sufficient fermionic magic,
are robust against dequantization by tensor network contraction, matchgate simulation, or any sampling-based
approach (see Section V C). Examples are given in Section VI, including orbitalrotated matrix product states (Section VI B) and unitary
cluster Jastrow circuits (Section VI C), both of which can
be prepared by circuits of linear depth in the number of
qubits. In the second part, we apply our proof of simulation hardness to quantum non-orthogonal multi-reference
methods for chemical ground state estimation [7, 8, 19,
20], which perform a classical diagonalization of the
Hamiltonian (energy) operator in a subspace of lowdepth reference states that are expressed in different orbital bases, providing a route to accurate computation of
electronic energies for molecular systems possessing both
static and dynamic electron correlations. These methods specifically target an important class of systems containing many significant molecules involved in catalysis
arXiv:2503.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
In contrast, the paper also details the computational feasibility of approximating expectation values within a well-conditioned subspace prepared by generalized UCCSD circuits of polynomial depth. p. 0 We apply this result to quantum non-orthogonal multi-reference methods designed for near-term
hardware by ruling out certain dequantization strategies for computing the off-diagonal matrix elements between reference states. We demonstrate these quantum speedups for two choices of ansatz
that incorporate both static and dynamic correlations to model the electronic eigenstates of molecular systems: linear combinations of orbital-rotated matrix product states, which are preparable in
linear depth, and linear combinations of states prepared by generalized UCCSD circuits of polynomial depth, for which computing the expectation values of local fermionic observables up to a
constant additive error is BQP-complete. Under this
mapping the fermion operators are expressed in terms of
the Pauli X, Y , and Z gates as
aˆp 7→ (Xp − iYp)Zp−1 · · ·Z1, (15)
where Xp denotes an X gate applied on qubit p, etc. ,
and we have dropped the hat notation for one- and twoqubit gates. Any logical
single-qubit gate can then be mapped directly to particle
number conserving matchgates via its decomposition into
Euler angles [42], as illustrated in Figure 1, corresponding to a complex-valued orbital rotation on the dual-rail
qubit. To implement an arbitrary logical computation one
gate is required that is not in the PN conserving matchgate set [43]. We present these results as evidence that superpolynomial quantum advantage in quantum chemistry is
theoretically achievable using methods suitable for nearterm hardware, noting that both of these ansatz classes
are designed to compactly represent energy eigenstates
of systems characterized by a combination of static and
dynamic electron correlation. OVERVIEW OF MAIN RESULTS
Our main technical result is to prove that orbital rotation circuits, which we denote by Gˆ, are universal under
post-selection when provided with four-qubit fermionic
magic state inputs (Theorem 6).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
1
No classical simulation hardness for orbital rotation circuits with four-qubit fermionic magic state inputs is proven under post-selection, which rules out certain dequantization strategies for computing off-diagonal matrix elements between reference states. This result suggests that near-term quantum hardware can achieve superpolynomial quantum advantage in simulating molecular electronic structure.
2
The AI system can perform ground state estimation for molecular systems characterized by both static and dynamic electron correlation using hybrid TNQE or NOQE methods, as the paper demonstrates that these methods are robust against dequantization by tensor network contraction, matchgate simulation, or any sampling-based approach
in the hardest case.
3
The system can efficiently approximate the expected energy of non-orthogonal multi-reference ansätze up to a constant additive error if they are prepared by polynomial depth generalized UCCSD circuits, as this is shown to be BQP complete.
4
The AI system can resolve off-diagonal matrix elements between non-universal reference states that include orbital rotations robustly against dequantization up to an additive error, provided the underlying circuit family encodes four-qubit fermionic magic states.
Abstract
We prove classical simulation hardness, under the generalized not equal to conjecture, for quantum circuit families with applications in near-term molecular electronic ground and excited state estimation. The proof exploits a connection to particle number conserving matchgate circuits with fermionic magic state inputs, which are shown to be universal for quantum computation under postselection, and are therefore not classically simulable in the worst case, in either the strong (multiplicative) or weak (sampling) sense. We apply these results to non-orthogonal quantum eigensolvers designed for near-term hardware by ruling out sample-based dequantization strategies to additively approximate the off-diagonal matrix elements between orbital-rotated correlated reference states that satisfy an irreducible fermionic magic property, which combines anticoncentration in any orbital basis with an average-case multiplicative approximation hardness conjecture for the associated matrix polynomials. We apply this complexity analysis to two choices of ansatz that incorporate both static and dynamic correlations to model the electronic eigenstates of molecular systems. These are linear combinations of orbital-rotated matrix product states, which are preparable in linear depth, and linear combinations of states prepared by generalized unitary coupled cluster with singles and doubles or unitary cluster Jastrow circuits of polynomial depth, for which computing the expectation values of local fermionic observables up to a constant additive error is-complete. We discuss the implications for achieving practical quantum advantage in resolving the electronic structure of catalytic systems composed from multivalent transition metal atoms using near-term quantum hardware.
Sources
- Classical computational simulation of the FeMo-cofactor model to chemical accuracy and its implications
- Many-hypercube codes: High-rate quantum error-correcting codes for high-performance fault-tolerant quantum computing
Related papers
- Computational regimes in matrix-product-state-based quantum trajectory simulations
- Toward chemically accurate quantum simulations on quantum hardware: A quantum-HPC hybrid approach
- Transversal surface-code game with reconfigurable qubits
- Fulqrum: a generalized framework for quantum subspace diagonalization
- Flux-Activated Resonant Control of a Bosonic Quantum Memory
- Machine learning the effects of many quantum measurements