Group averaging and the geometry of physical quantum states: from normalizable solutions to solvable loop quantum cosmology

arXiv:2610.24027 · gr-qc, hep-th, math-ph, math.MP · Submitted 2026-10-16 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: General Relativity and Quantum Cosmology. The day in general relativity and quantum cosmology.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Group averaging and the geometry of physical quantum states".

Mira: Here's the gist of the paper. Group averaging constructs physical states from normalizable kinematical seeds in canonical quantum gravity where physical states are typically not normalizable in the kinematical Hilbert space When averaging is an orthogonal projection, admissible kinematical rays form a holomorphic vector bundle over the physical ray space Observables that descend to the bundle base are characterized by the vanishing of the Hessian of ⟨Cb 2⟩ along their flow on physical states This work explores how group averaging relates canonical structures, physical geometry, and Dirac observables in solvable loop quantum cosmology.

Kai: The Geometry of Physical State Space The paper details the geometric formulation of quantum mechanics where a ray is a point on a Kähler manifold and expectation values generate Hamiltonian motions that preserve its quantum geometry The physical ray space is characterized as the zero set of the function ⟨Cb 2⟩ for normalizable physical vectors.

Mira: When averaging is an orthogonal projection, the admissible kinematical ray space is a holomorphic vector bundle over the physical ray space, and its fibers are strictly larger than gauge orbits Observables that descend to the bundle base are characterized by the vanishing of the Hessian of ⟨Cb 2⟩ along their flow on physical states.

Kai: The Nature of Averaging and Lifts Group averaging assigns the same physical ray to many seeds, and when averaging is an orthogonal projection, it results in a holomorphic vector bundle where the embedded physical space is its zero section For distributional physical states, group averaging on an explicit Sobolev test space yields an affine bundle of representatives whose sections are K¨ahler-isometric embeddings of the physical ray space whose images contain no exact solutions The test-space inner product singles out one of these embeddings, and narrowing the seed profile produces Weyl sequences of approximate solutions.

Title and authors: Kai: So we've touched on the setup and how they use group averaging to build physical states from seeds, and now we need to talk about what they actually propose as improvements on this paper, "Group averaging and the geometry of physical quantum states: from normalizable solutions to solvable loop quantum cosmology." >

Mira: Well, one major improvement they suggest is that this framework allows you to construct those physical states from normalizable kinematical seeds by viewing them as sections of a holomorphic vector bundle over the physical ray space >

Lev: That’s useful because it gives us a geometric way to understand how averaging maps multiple seeds onto one physical ray, which is more tangible than just saying we have an equivalence class >

Kai: And then there's the idea about observables; they identify observables that descend to the base of this bundle by checking if "the Hessian of the squared constraint vanishes along their flow on physical states" >

Mira: That vanishing Hessian condition, I think that’s a precise mathematical requirement derived from Proposition one and it’s how you rigorously select what descends to the base > <ref:2610.24027#pg2>

Lev: For running this on real hardware, we need to know if that condition is computationally feasible or if it introduces too much overhead in terms of error correction >

Kai: And for distributional states, they use group averaging on an explicit Sobolev test space which yields an affine bundle of representatives whose sections are embeddings of the physical ray space whose images contain no exact solutions >

Mira: That’s interesting because those embeddings not containing exact solutions suggests we're dealing with things that aren't classical solutions in this context >

Lev: So, to summarize the improvements, it’s about giving us a clearer geometric language for states and a specific test for observables, even when the states aren't normalizable >

Kai: Exactly. It takes something abstract—the non-normalizable physical states—and gives us a structured way to describe their relationship with the kinematical seeds >

Mira: So, we’ve seen the setup and the proposed fixes for handling those tricky distributional cases in this paper on "Group averaging and the geometry of physical quantum states: from normalizable solutions to solvable loop quantum cosmology" >

The paper's summary: Kai: To summarize what this paper is really doing, it’s taking canonical structures and physical geometry and Dirac observables in solvable loop quantum cosmology and showing how they all connect through the lens of group averaging >

Mira: It shows that physical histories are actually distributions in these models, and mean-ergodic averaging results in zero when you use this specific construction >

Lev: That’s a strong claim because it connects the canonical description to a specific geometric outcome, implying that the dynamics simplify significantly under this projection >

Kai: They show that when you apply group averaging to solvable loop quantum cosmology, it reproduces the Klein–Gordon physical inner product on an explicit test space >

Mira: And they point out that because of how the scalar-field momentum enters the constraint quadratically, there's a constraint spectrum edge near the shell >

Lev: That spectral edge effect is what makes things biased; it suggests that canonical seeds of small universes are pushed off the shell because of this structure >

Kai: They also mention that this edge separates the Page–Wootters state from the slice of the Klein–Gordon solution, which is a specific piece of physical history >

Mira: So, they’re connecting these canonical structures—the constraints, the geometry, and those observables—and showing what happens when you push past normalizability >

Lev: It’s about showing that even in these solvable models, the geometry isn't always simple; the canonical section mode-dependence is something to watch out for >

The paper's improvements: Kai: Now let’s talk about what this paper suggests as improvements, beyond just the initial setup. They show that by applying group averaging to solvable loop quantum cosmology, they reproduce the Klein–Gordon physical inner product on an explicit test space >

Mira: And they highlight that because of the quadratic nature of the scalar-field momentum in constraints, there's a constraint spectrum edge near the shell, which makes the canonical section mode-dependent and biases canonical seeds of small universes off the shell >

Lev: From my perspective on error correction, that quadratic momentum term is what really matters for running on real hardware because it dictates where we need to place our approximations to keep things stable >

Kai: And they also discuss distributional states, where averaging on an explicit Sobolev test space yields an affine bundle of representatives whose sections are embeddings of the physical ray space whose images contain no exact solutions >

Mira: That’s a way to handle non-normalizable states by getting these embeddings, and then narrowing the seed profile produces Weyl sequences of approximate solutions >

Lev: So, the core improvement here is that they provide concrete mechanisms for handling those non-normalizable states—they give us a process to select approximate physical histories >

Kai: It’s about giving us a way to handle the complexity of these states without just saying they don't exist in the standard Hilbert space >

Conclusion: Mira: So, wrapping up on this paper, it shows that physical histories in models like solvable loop quantum cosmology are distributions, mean-ergodic averaging is zero, and the rigging construction on a test space gives an affine bundle >

Kai: And they confirm that the canonical section depends on the mode and is not K¨ahler isometric in solvable loop quantum cosmology >

Lev: The spectral-edge effects discussed in Section six rely only on the clock momentum entering quadratically, and that should persist in other deparametrized models with relativistic clocks >

Kai: They establish which parts of the relation between embedded solutions, averaging fibers and lifts are canonical, even though the paper notes that only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle >

Mira: So, to end this up: they’ve mapped out how to get from normalizable seeds to physical states using group averaging in solvable loop quantum cosmology >

Lev: I just want to add that the key takeaway for real physics is that you have a concrete way to select approximate physical histories using these test spaces and clock models >

Kai: That's it for this one. We’ve seen how group averaging helps us understand the geometry of physical quantum states in solvable loop quantum cosmology >

Centro de Ciencias Matemáticas, Universidad Nacional Autónoma de México

gr-qc, hep-th, math-ph, math.MP

Submitted: 2026-10-16

Updated: 2026-10-28

Comments: 26 pages. Discussion expanded

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 83/100

The gist: Group averaging constructs physical states from normalizable kinematical seeds in canonical quantum gravity where physical states are typically not normalizable in the kinematical Hilbert space When

Terminology

Summary

Group averaging constructs physical states from normalizable kinematical seeds in canonical quantum gravity where physical states are typically not normalizable in the kinematical Hilbert space When averaging is an orthogonal projection, admissible kinematical rays form a holomorphic vector bundle over the physical ray space Observables that descend to the bundle base are characterized by the vanishing of the Hessian of ⟨Cb 2⟩ along their flow on physical states This work explores how group averaging relates canonical structures, physical geometry, and Dirac observables in solvable loop quantum cosmology.

The Geometry of Physical State Space

The paper details the geometric formulation of quantum mechanics where a ray is a point on a Kähler manifold and expectation values generate Hamiltonian motions that preserve its quantum geometry The physical ray space is characterized as the zero set of the function ⟨Cb 2⟩ for normalizable physical vectors. When averaging is an orthogonal projection, the admissible kinematical ray space is a holomorphic vector bundle over the physical ray space, and its fibers are strictly larger than gauge orbits Observables that descend to the bundle base are characterized by the vanishing of the Hessian of ⟨Cb 2⟩ along their flow on physical states.

The Nature of Averaging and Lifts

Group averaging assigns the same physical ray to many seeds, and when averaging is an orthogonal projection, it results in a holomorphic vector bundle where the embedded physical space is its zero section For distributional physical states, group averaging on an explicit Sobolev test space yields an affine bundle of representatives whose sections are K¨ahler-isometric embeddings of the physical ray space whose images contain no exact solutions The test-space inner product singles out one of these embeddings, and narrowing the seed profile produces Weyl sequences of approximate solutions.

Loop Quantum Cosmology Results

Applying group averaging to solvable loop quantum cosmology reproduces the Klein–Gordon physical inner product on an explicit test space Because the scalar-field momentum enters the constraint quadratically, the constraint spectrum has an edge near the shell, which makes the canonical section mode-dependent and biases canonical seeds of small universes off the shell This edge also separates the Page–Wootters state from the slice of the Klein–Gordon solution.

Distributional States and Clock Models

For parametrized oscillators with distributional physical states, averaging on an explicit Sobolev test space yields an affine bundle of representatives whose sections are embeddings of the physical ray space whose images contain no exact solutions In solvable loop quantum cosmology, group averaging reproduces the Klein–Gordon physical inner product The canonical lift feels the edge, and for small universes, canonical seeds have ⟨Cb⟩ > 0 because they lie on average on the side pϕ > Θ of the shell.

Relevance for Gravitational Constraints

The analysis separates physical rays, physical geometry and Dirac observables from those that depend on a choice of seeds, test space or clock The results show that the canonical section depends on the mode and is not K¨ahler isometric in solvable loop quantum cosmology The spectral-edge effects of Section 6 rely only on the clock momentum entering quadratically, and should persist in other deparametrized models with relativistic clocks. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical.

Summary of Key Findings

The paper demonstrates that physical histories in models like solvable loop quantum cosmology are distributions, mean-ergodic averaging is zero, and the rigging construction on a test space gives an affine bundle The canonical section depends on the mode and is not K¨ahler isometric in solvable loop quantum cosmology The spectral-edge effects of Section 6 rely only on the clock momentum entering quadratically, and should persist in other deparametrized models with relativistic clocks. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. The clock and sLQC results use explicit, surjective trace maps on the Sobolev spaces (26) and (36). Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle. The paper establishes which parts of the relation between embedded solutions, averaging fibers and lifts are canonical. The scope of each statement should be kept explicit. Only the last carries the physical inner product at fixed clock reading, and none of them is a section of the averaging bundle.

Improvements for AI systems

  1. Vector bundle construction for physical states: The system can construct physical states from normalizable kinematical seeds by viewing them as sections of a holomorphic vector bundle over the physical ray space, which allows for a geometric understanding of how averaging maps multiple seeds to one physical ray.

  2. Observables selection criterion: The AI can identify observables that descend to the base of this bundle by checking if the Hessian of the squared constraint vanishes along their flow on physical states, which is a precise condition derived from Proposition 1.

  3. Handling distributional states: For physical states that are not normalizable, the system can use group averaging on an explicit Sobolev test space to yield an affine bundle of representatives, allowing it to select a specific representative by narrowing the seed profile to produce Weyl sequences of approximate solutions.

  4. Constraint spectrum analysis in solvable loop quantum cosmology: The AI can analyze how the constraint spectrum has an edge near the shell due to the quadratic nature of scalar-field momentum in constraints, and use this edge to determine if canonical seeds are biased off the shell.

  5. Distinguishing physical and kinematical structures: The system can separate physical rays, physical geometry and Dirac observables from those that depend on a choice of seeds, test space or clock, allowing it to rigorously compare different descriptions of evolution in gravitational models.

Abstract

In canonical quantum gravity the physical states solve the constraints but are typically not normalizable in the kinematical Hilbert space. Group averaging constructs them from normalizable kinematical seeds. We describe the geometry of this seed-to-state map in the language of geometric quantum mechanics. When averaging is an orthogonal projection, the admissible kinematical rays form a holomorphic vector bundle over the physical ray space. The embedded exact solutions form its zero section, and its fibers are strictly larger than gauge orbits. Observables descend to the base exactly when the Hessian of the squared constraint vanishes along their flow. For distributional physical states, group averaging on an explicit Sobolev test space yields an affine bundle of representatives. Its sections are embeddings of the physical ray space whose images contain no exact solutions. The test-space inner product singles out one of them, and narrowing the seed profile gives Weyl sequences of approximate solutions. We apply this to solvable loop quantum cosmology. Group averaging reproduces the Klein--Gordon physical inner product on an explicit test space. Because the scalar-field momentum enters the constraint quadratically, the constraint spectrum has an edge near the shell. The canonical section is then mode-dependent and not Kähler isometric, and the canonical seeds of small universes are biased off the shell. The edge also separates the Page--Wootters state from the slice of the Klein--Gordon solution. An ideal clock shows none of these effects. The results separate the canonical structures, namely physical rays, physical geometry and Dirac observables, from those that depend on a choice of seeds, test space or clock.

Sources

Related papers