Pith. sign in

REVIEW 2 major objections 36 references

Conditioned free-energy density of proteins using unbalanced solutions to constraint satisfaction problems

T0 review · 2 major / 0 minor · reviewed 2026-06-28 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Free-energy computation for conditioned protein spin models reduces to unbalanced 2-to-1 norm solved by SDP.

desk verdict The reduction of conditioned Curie-Weiss log-partition functions to unbalanced 2-to-1 norm with an SDP algorithm is the actual new piece, but the Ubiquitin modeling step sits on an untested assumption. read the letter →

arxiv 2606.01329 v1 pith:HJIUV4LG submitted 2026-05-31 cs.LG q-bio.BM

Pratik Worah , Subhash Khot , Srinivasa Varadhan This is my paper · ORCID
classification cs.LGq-bio.BM
keywords free-energyCurie-WeissHamiltonianunbalancednormsemidefiniteprogrammingproteinflexibilityUbiquitinconstraintsatisfactionlog-partitionfunction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper shows that the log-partition function of conditioned inhomogeneous Curie-Weiss spin Hamiltonians equals an unbalanced 2-to-1 norm computation. It supplies a polynomial-time semidefinite programming algorithm together with a proof of the minimum unbalance attained. The same reduction is used on the protein Ubiquitin by beginning with its crystal structure, sampling alternate backbone shapes on the free-energy surface, and locating flexible segments that leave secondary structure unchanged. A reader would care because the method supplies a tractable way to map conformational flexibility from a single starting structure without exhaustive simulation.

What carries the argument

Reduction of the conditioned free-energy (log-partition function) to an unbalanced 2-to-1 norm computation, solved via semidefinite programming with a proven lower bound on unbalance.

What would settle it

Direct comparison of the flexible regions predicted for Ubiquitin against experimental measures such as NMR order parameters or crystallographic B-factors; mismatch in the identified segments would falsify the model.

Watch

Extended reading notes

Core claim

Computing the log-partition function (free-energy) of conditioned inhomogeneous Curie-Weiss spin Hamiltonians reduces to an unbalanced 2 to 1 norm computation, and design a polynomial-time SDP algorithm for this problem with a lower bound proof for the amount of unbalance achieved. Applied to the protein Ubiquitin, the framework starts from a known crystal structure, explores alternative backbone conformations across the free-energy landscape, and identifies flexible regions of the protein while preserving its native secondary structure.

Load-bearing premise

That an inhomogeneous Curie-Weiss spin Hamiltonian with the stated conditioning accurately captures the free-energy landscape of a real protein such as Ubiquitin when started from its crystal structure.

Editorial extensions

If this is right

  • The SDP algorithm computes the free-energy value in polynomial time.
  • The solution is guaranteed to achieve at least the proved lower bound on unbalance.
  • The framework can enumerate alternative backbone conformations for Ubiquitin while keeping native secondary structure fixed.
  • Flexible regions of the protein are identified directly from the crystal structure input.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same reduction might be tested on other proteins whose crystal structures are known to check whether flexibility predictions generalize.
  • The link between constraint-satisfaction norms and biophysical Hamiltonians could be examined for other molecular systems that admit spin-like representations.
  • If the unbalance bound is tight, it may limit the range of conformations reachable by the model and suggest where additional constraints would be needed.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 0 minor

Summary. The paper claims that computing the log-partition function of conditioned inhomogeneous Curie-Weiss spin Hamiltonians reduces to an unbalanced 2→1 norm computation, for which a polynomial-time SDP algorithm is designed together with a lower-bound proof on the achievable unbalance. The framework is then applied to Ubiquitin, starting from its crystal structure, to explore backbone conformations across the free-energy landscape and identify flexible regions while preserving native secondary structure.

Significance. If the reduction and SDP algorithm hold, the work supplies an efficient, polynomial-time method for free-energy computation in this class of conditioned spin models, together with a provable guarantee on unbalance; this algorithmic contribution would be of interest in constraint-satisfaction and statistical-physics settings. The protein application, however, hinges on an unverified modeling assumption whose validity is not demonstrated.

major comments (2)
  1. [Abstract / application section] Abstract and application section: the claim that the conditioned Curie-Weiss Hamiltonian plus conditioning accurately reproduces the free-energy landscape of Ubiquitin (allowing identification of flexible regions from the crystal structure) is load-bearing for the biological results, yet no validation against MD trajectories, NMR order parameters, or B-factors is supplied.
  2. [Abstract] Abstract: the asserted mathematical reduction of the log-partition function to unbalanced 2→1 norm is stated without derivation steps, explicit verification of the reduction, or empirical controls on the SDP algorithm, preventing assessment of the central algorithmic claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below, indicating planned revisions where the manuscript can be improved without misrepresenting the work.

read point-by-point responses

  1. Referee: [Abstract / application section] Abstract and application section: the claim that the conditioned Curie-Weiss Hamiltonian plus conditioning accurately reproduces the free-energy landscape of Ubiquitin (allowing identification of flexible regions from the crystal structure) is load-bearing for the biological results, yet no validation against MD trajectories, NMR order parameters, or B-factors is supplied.

    Authors: We agree that the manuscript supplies no direct validation of the Ubiquitin results against MD trajectories, NMR order parameters, or B-factors. The protein example is presented as an illustration of the algorithmic framework rather than a claim of quantitative biological accuracy. In revision we will add an explicit limitations paragraph in the application section that states the mean-field modeling assumptions, notes the absence of such validation, and cites prior literature on Curie-Weiss-type models for backbone flexibility. We will also moderate the abstract wording to describe the output as “candidate flexible regions identified under the model” rather than implying direct reproduction of the experimental landscape. revision: partial

  2. Referee: [Abstract] Abstract: the asserted mathematical reduction of the log-partition function to unbalanced 2→1 norm is stated without derivation steps, explicit verification of the reduction, or empirical controls on the SDP algorithm, preventing assessment of the central algorithmic claim.

    Authors: The abstract is intentionally concise, but the full reduction is derived in Section 2 (Theorem 1 and its proof), small-instance verification appears in Section 3, and SDP performance with empirical controls is reported in Section 4. To address the concern we will expand the abstract by one sentence that sketches the reduction at high level and insert explicit section references for the derivation and experiments. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: mathematical reduction and SDP algorithm are self-contained derivations

full rationale

The paper's core claim is a reduction of the log-partition function for conditioned inhomogeneous Curie-Weiss Hamiltonians to an unbalanced 2→1 norm computation, together with a polynomial-time SDP algorithm and unbalance lower bound. No equations, definitions, or steps in the provided abstract or description reduce this result to a fitted parameter, self-citation chain, or ansatz imported from prior work by the same authors. The protein application (Ubiquitin modeling) is presented as an empirical use of the framework rather than a load-bearing derivation that feeds back into the math. This satisfies the default expectation of a non-circular paper whose central result rests on an external mathematical argument.

Assumptions & free parameters 0 free parameters · 0 assumptions · 0 invented entities

Abstract supplies insufficient detail to enumerate free parameters, axioms, or invented entities; all fields left empty pending full text.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Conditioned free-energy density of proteins using unbalanced solutions to constraint satisfaction problems." pith.science (2026). https://pith.science/paper/HJIUV4LG

@misc{pith2026260601329,
  author       = {Pith},
  title        = {Pith review of: Conditioned free-energy density of proteins using unbalanced solutions to constraint satisfaction problems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HJIUV4LG}},
  note         = {Machine review of arXiv:2606.01329}
}
abstract

We show that computing the log-partition function (free-energy) of conditioned inhomogeneous Curie--Weiss spin Hamiltonians reduces to an unbalanced $2 \to 1$ norm computation, and design a polynomial-time SDP algorithm for this problem with a lower bound proof for the amount of unbalance achieved. Applied to the protein Ubiquitin, the framework starts from a known crystal structure, explores alternative backbone conformations across the free-energy landscape, and identifies flexible regions of the protein while preserving its native secondary structure.

Figures

Figures reproduced from arXiv: 2606.01329 by the authors.

Figure 1
Figure 1. Free-energy landscape F(y) = opt(y) − y for the Ubiquitin Hamiltonian (n = 76). The concave shape confirms a unique free-energy optimum at y ∗ ≈ 592 (≈ 7.8 n), well above the energy￾only value y = n. 3.3 Conditioned Structural Ensemble via ε-Sweep The PDB ground-truth Ramachandran plot for 1UBQ shows Ubiquitin’s mixed α/β fold: 40 β-sheet residues, 16 αR-helix, 5 left-handed α, and 13 coil (Supplementary [PITH_FULL… view at source ↗
Figure 2
Figure 2. Free-energy-conditioned Ramachandran plots [ [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Secondary structure composition vs. ε. Left: Coarse classification (β-sheet at 57%, α-helix at 34%) is flat across conditioning levels. Right: Helix subtypes (αR, 310, α-broad) show only 1–2 residue fluctuations. Per-residue strip diagram [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Per-residue secondary structure strip diagram across the [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Per-residue ψ drift relative to ε = 0.1. Rows correspond to conditioning levels; columns to residues. Approximately 20 residues shift by up to 30◦ at ε ≥ 0.7. The ϕ drift (not shown) is identically zero. 3.5 Validation Pipeline self-consistency. At ε = 0.9 (nearest to …
Figure 6
Figure 6. Figure 6: Ramachandran plots at ε = 0.9: energy-only (y = n, left) vs. free-energy (y = y ∗ , right). Unbalance factor on the protein Hamiltonian [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Relative unbalance factor vs. ε for the Ubiquitin Hamiltonian. Shaded regions show 68% and 95% confidence intervals over Gaussian rounding draws. 4 Discussion The results above demonstrate that an SDP framework can be used to compute conditioned free￾energy states of p…
Figure 8
Figure 8. Figure 8: Synthetic ℓ1-PCA validation of Algorithm 1 with B = I10 and A drawn from the 10 × 10 GOE. (a) Unconstrained baseline is approximately balanced. (b) ε = 0.5 yields more unbalanced solutions. (c) Objective ratio is approximately linear in ε. (d) Relative unbalance is nea…
Figure 9
Figure 9. Figure 9: Hamiltonian matrix for Ubiquitin, constructed from PDB-derived J-coupling constants. [PITH_FULL_IMAGE:figures/full_fig_p025_9.png]
Figure 10
Figure 10. Figure 10: Ramachandran plot of the 1UBQ crystal structure (PDB ground truth). The 74 backbone [PITH_FULL_IMAGE:figures/full_fig_p025_10.png]
Figure 11
Figure 11. Figure 11: Round-trip pipeline validation: forward Karplus [PITH_FULL_IMAGE:figures/full_fig_p026_11.png]
Figure 12
Figure 12. Figure 12: Angular RMSD (degrees) between SDP-recovered dihedral angles and the 1UBQ crystal [PITH_FULL_IMAGE:figures/full_fig_p027_12.png]
Figure 13
Figure 13. Figure 13: Per-residue secondary structure vs. ε at α = 0.7. Compare with [PITH_FULL_IMAGE:figures/full_fig_p027_13.png]
Figure 14
Figure 14. Figure 14: Per-residue secondary structure vs. ε at α = 0.8. Compare with [PITH_FULL_IMAGE:figures/full_fig_p027_14.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 2 canonical work pages

  1. [1]

    Approximating the cut-norm via grothendieck’s inequality.SIAM Journal on Computing, 35(4):787–803, 2006

    Noga Alon and Assaf Naor. Approximating the cut-norm via grothendieck’s inequality.SIAM Journal on Computing, 35(4):787–803, 2006

    2006

  2. [2]

    Kinch, R

    Minkyung Baek, Frank DiMaio, Ivan Anishchenko, Justas Dauparas, Sergey Ovchinnikov, Gyu Rie Lee, Jue Wang, Qian Cong, Lisa N. Kinch, R. Dustin Schaeffer, Claudia Millán, Hahnbeom Park, Carson Adams, Caleb R. Glassman, Andy DeGiovanni, Jose H. Pereira, Andria V. Rodrigues, Alberdina A. van Dijk, Ana C. Ebrecht, Diederik J. Opperman, Theo Sagmeister, Christ...

    2021

  3. [3]

    On the computational complexity of Ising spin glass models.Journal of Physics A: Mathematical and General, 15(10):3241–3253, 1982

    Francisco Barahona. On the computational complexity of Ising spin glass models.Journal of Physics A: Mathematical and General, 15(10):3241–3253, 1982

    1982

  4. [4]

    Inapproximability of matrixp→q norms.SIAM Journal on Computing, 52(1):132– 155, 2023

    Vijay Bhattiprolu, Mrinal Kanti Ghosh, Venkatesan Guruswami, Euiwoong Lee, and Madhur Tulsiani. Inapproximability of matrixp→q norms.SIAM Journal on Computing, 52(1):132– 155, 2023

    2023

  5. [5]

    Approximatingoperatornormsviageneralizedkrivinerounding.CoRR,abs/1804.03644, 2018

    Vijay Bhattiprolu, Mrinalkanti Ghosh, Venkatesan Guruswami, Euiwoong Lee, and Madhur Tulsiani. Approximatingoperatornormsviageneralizedkrivinerounding.CoRR,abs/1804.03644, 2018

    work page arXiv 2018

  6. [6]

    Inapproximability of Matrix $p\rightarrow q$ Norms

    Vijay Bhattiprolu, Mrinalkanti Ghosh, Venkatesan Guruswami, Euiwoong Lee, and Madhur Tulsiani. Inapproximability of matrix p→q norms.CoRR, abs/1802.07425, 2018

    work page Pith review arXiv 2018

  7. [7]

    Stephen A. Cook. The complexity of theorem-proving procedures. InProceedings of the Third Annual ACM Symposium on Theory of Computing, STOC ’71, page 151–158, New York, NY, USA, 1971. Association for Computing Machinery

    1971

  8. [8]

    Ising Critical Behavior of Inhomogeneous Curie-Weiss Models and Annealed Random Graphs.Communications in Mathematical Physics, 348, 2016

    Sander Dommers, Cristian Giardina, Claudio Giberti, Remco van der Hofstad, and Maria Luisa Prioriello. Ising Critical Behavior of Inhomogeneous Curie-Weiss Models and Annealed Random Graphs.Communications in Mathematical Physics, 348, 2016

    2016

Show all 36 references
  1. [9]

    Annealed central limit theorems for the Ising model on random graphs.Latin American Journal of Probability and Mathematical Statistics, XIII, 2016

    Cristian Giardina, Claudio Giberti, Remco van der Hofstad, and Maria Luisa Prioriello. Annealed central limit theorems for the Ising model on random graphs.Latin American Journal of Probability and Mathematical Statistics, XIII, 2016

    2016

  2. [10]

    Goemans and David P

    Michel X. Goemans and David P. Williamson. Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming.J. ACM, 42(6):1115–1145, November 1995

    1995

  3. [11]

    Protein Structure Prediction with Lattice Models.Handbook of Computational Molecular Biology, 2005

    William Hart and Alantha Newman. Protein Structure Prediction with Lattice Models.Handbook of Computational Molecular Biology, 2005

    2005

  4. [12]

    Polynomial-time approximation algorithms for the Ising model.SIAM Journal on Computing, 22(5):1087–1116, 1993

    Mark Jerrum and Alistair Sinclair. Polynomial-time approximation algorithms for the Ising model.SIAM Journal on Computing, 22(5):1087–1116, 1993

    1993

  5. [13]

    John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ron- neberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko, Alex 12 Bridgland, Clemens Meyer, Simon A. A. Kohl, Andrew J. Ballard, Andrew Cowie, Bernardino Romera-Paredes, ...

    2021

  6. [14]

    Contact Electron-Spin Coupling of Nuclear Magnetic Moments.The Journal of Chemical Physics, 30(1):11–15, 1959

    Martin Karplus. Contact Electron-Spin Coupling of Nuclear Magnetic Moments.The Journal of Chemical Physics, 30(1):11–15, 1959

    1959

  7. [15]

    Grothendieck-type inequalities in combinatorial optimization

    Subhash Khot and Assaf Naor. Grothendieck-type inequalities in combinatorial optimization. Communications on Pure and Applied Mathematics, 65(7):992–1035, 2012

    2012

  8. [16]

    Mean-absolute deviation portfolio optimization model and its applications to Tokyo stock market.Management Science, 37(5):519–531, 1991

    Hiroshi Konno and Hiroaki Yamazaki. Mean-absolute deviation portfolio optimization model and its applications to Tokyo stock market.Management Science, 37(5):519–531, 1991

    1991

  9. [17]

    Lewis, Oliver F

    Andrew Leaver-Fay, Michael Tyka, Steven M. Lewis, Oliver F. Lange, James Thompson, Ron Jacak, Kristian W. Kaufman, P. Douglas Renfrew, Colin A. Smith, Will Sheffler, Ian W. Davis, Seth Cooper, Adrien Treuille, Daniel J. Mandell, Florian Richter, Yih-En Andrew Ban, Sarel J. Fle...

    2011

  10. [18]

    Lovell, Ian W

    Simon C. Lovell, Ian W. Davis, W. Bryan Arendall, Paul I. W. de Bakker, J. Michael Word, Michael G. Prisant, Jane S. Richardson, and David C. Richardson. Structure validation by Cα geometry: ϕ, ψ and Cβ deviation.Proteins: Structure, Function, and Bioinformatics, 50(3):437–450, 2003

    2003

  11. [19]

    Ross, Markus K

    Chuanjie Lu, Chao Wu, Dmitry Ghoreishi, Wei Chen, Lingle Wang, Wolfgang Damm, Gail A. Ross, Markus K. Dahlgren, Ellery Russell, Christopher D. Von Bargen, Robert Abel, Richard A. Friesner, and Edward D. Harder. OPLS4: Improving force field accuracy on challenging regimes of ch...

    2021

  12. [20]

    Markopoulos, George N

    Panos P. Markopoulos, George N. Karystinos, and Dimitris A. Pados. Optimal Algorithms for L1-subspace Signal Processing.IEEE Transactions on Signal Processing, 62(19):5046–5058, 2014

    2014

  13. [21]

    Sign-constrained least squares estimation for high-dimensional regression

    Nicolai Meinshausen. Sign-constrained least squares estimation for high-dimensional regression. Electronic Journal of Statistics, pages 1607–1631, 2013

    2013

  14. [22]

    Palmer, Charles D

    Arthur G. Palmer, Charles D. Kroenke, and J. Patrick Loria. Nuclear magnetic resonance methods for quantifying microsecond-to-millisecond motions in biological macromolecules. Methods in Enzymology, 339:204–238, 2001

    2001

  15. [23]

    R. K. Pathria and Paul D. Beale.Statistical Mechanics. Academic Press, 3rd edition, 2011

    2011

  16. [24]

    Corey, and H

    Linus Pauling, Robert B. Corey, and H. R. Branson. The structure of proteins: two hydrogen- bonded helical configurations of the polypeptide chain.Proceedings of the National Academy of Sciences, 37(4):205–211, 1951. 13

    1951

  17. [25]

    G. N. Ramachandran, C. Ramakrishnan, and V. Sasisekharan. Stereochemistry of polypeptide chain configurations.Journal of Molecular Biology, 7(1):95–99, 1963

    1963

  18. [26]

    TALOS+: a hybrid method for predicting protein backbone torsion angles from NMR chemical shifts.Journal of Biomolecular NMR, 44(4):213–223, 2009

    Yang Shen and Ad Bax. TALOS+: a hybrid method for predicting protein backbone torsion angles from NMR chemical shifts.Journal of Biomolecular NMR, 44(4):213–223, 2009

    2009

  19. [27]

    Aramini, Gaohua Liu, Alexander Eletsky, Yibing Wu, Kiran K

    Yang Shen, Oliver Lange, Frank Delaglio, Paolo Rossi, James M. Aramini, Gaohua Liu, Alexander Eletsky, Yibing Wu, Kiran K. Singarapu, Alexander Lemak, Alexandr Ignatchenko, Cheryl H. Arrowsmith, Thomas Szyperski, Gaetano T. Montelione, David Baker, and Ad Bax. Consistent blind...

    2008

  20. [28]

    Non-negative least squares for high-dimensional linear models: Consistency and sparse recovery without regularization.Electronic Journal of Statistics, 2013

    Martin Slawski and Matthias Hein. Non-negative least squares for high-dimensional linear models: Consistency and sparse recovery without regularization.Electronic Journal of Statistics, 2013

    2013

  21. [29]

    Supersparse linear integer models for optimized medical scoring systems.Machine Learning, 102(3):349–391, 2016

    Berk Ustun and Cynthia Rudin. Supersparse linear integer models for optimized medical scoring systems.Machine Learning, 102(3):349–391, 2016

    2016

  22. [30]

    Vijay-Kumar, Charles E

    S. Vijay-Kumar, Charles E. Bugg, and William J. Cook. Structure of ubiquitin refined at 1.8 å resolution.Journal of Molecular Biology, 194(3):531–544, 1987

    1987

  23. [31]

    Vuister and Ad Bax

    Geerten W. Vuister and Ad Bax. Quantitative J correlation: a new approach for measuring homonuclear three-bond J(HNHα)coupling constants in 15N-enriched proteins.Journal of the American Chemical Society, 115(17):7772–7777, 1993

    1993

  24. [32]

    Watson, David Juergens, Nathaniel R

    Joseph L. Watson, David Juergens, Nathaniel R. Bennett, Brian L. Trippe, Jason Yim, Helen E. Eisenach, Woody Ahern, Andrew J. Borst, Robert J. Ragotte, Lukas F. Milles, Basile I. M. Wicky, Nikita Hanikel, Samuel J. Pellock, Alexis Courbet, William Sheffler, Jue Wang, Preetham ...

    2023

  25. [33]

    Angular dependence of1J(Ni, Cαi)and 2J(Ni, Cα(i−1)) coupling constants measured in J-modulated HSQCs.Journal of Biomolecular NMR, 23:47–55, 2002

    Julia Wirmer and Harald Schwalbe. Angular dependence of1J(Ni, Cαi)and 2J(Ni, Cα(i−1)) coupling constants measured in J-modulated HSQCs.Journal of Biomolecular NMR, 23:47–55, 2002. 14 A Supplementary Information A.1 Notation Table 2: Summary of notation used in the paper. Symbo...

    2002

  26. [34]

    A guarantee on the unbalance factor as measured by the log-mutual-coherence.6

  27. [35]

    The second follows from Theorem 7; the first from Theorem 5, which relies on Lemma 6

    A guarantee on the expectedℓ1 objective of the rounded solution. The second follows from Theorem 7; the first from Theorem 5, which relies on Lemma 6. Theorem 5.Let ]Bias be the solution from Algorithm 1, and letρ(X) := AM({∥Bi+Bj ∥4 X }i,j∈[n]) 1 2 GM({∥Bi+Bj ∥4 X }i,j∈[n]) 1...

  28. [36]

    Theorem 7.The rounded solution from Algorithm 2 has a constant approximation factor ofp π 2 for theℓ 1 objective relative to the SDP solution

    Furthermore, assume that∀i, j∈ [n] : ∥ui + uj∥X > 0, and letρ(X) := AM({∥Bi+Bj ∥4 X }i,j∈[n]) 1 2 GM({∥Bi+Bj ∥4 X }i,j∈[n]) 1 2 · AM({∥Bi∥2 X ∥Bj ∥2 X }i,j∈[n]) 1 2 HM({∥Bi∥2 X ∥Bj ∥2 X }i,j∈[n]) 1 2 , then: ρ(X)· Y i,j∈[n] 1 + 2· ⟨u i,u j⟩X ∥ui∥2 X +∥u j∥2 X 1 n2 ≥ 1 n2 · X i...

Pith tools

Reviewed June 28, 2026 · model on record in the stance chip above.