Pith. sign in

REVIEW 5 major objections 6 minor 1 cited by

THOI: An efficient and accessible library for computing higher-order interactions enhanced by batch-processing

T0 review · 5 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read THOI computes all higher-order interactions for systems up to 30 variables in hours on a laptop, far outpacing existing open-source toolboxes.

desk verdict THOI is a genuinely useful engineering contribution that makes exhaustive HOI analysis feasible for ~30 variables, but its accuracy validation is circular and needs an independent reference check before the numbers can be trusted. read the letter →

arxiv 2501.03381 v1 pith:YYBNRUXE submitted 2025-01-06 cs.SC

classification cs.SC
keywords higher-orderinteractionsO-informationsynergyredundancyGaussiancopulabatchprocessingPyTorchsimulatedannealing
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

THOI is a Python library that claims to make exhaustive higher-order interaction (HOI) analysis practical for systems with up to 30 continuous variables, and heuristic searches practical for larger ones. It computes the four standard multivariate information measures — total correlation, dual total correlation, O-information ($\Omega$), and S-information — from Gaussian-copula entropy estimates, batched through PyTorch tensor operations. On a 30-variable benchmark with 1000 samples, the paper reports computing every interaction order from 3 to 30 in 5.8 hours using under 3 GB of memory, where compared open-source libraries would take an extrapolated 221 days, 2 days with a 240 GB memory overload, or roughly 17 years. The authors validate accuracy on synthetic probabilistic graphical models with known synergy and redundancy structure, apply the tool to fMRI data from wakeful and deeply anesthetized subjects, and exhaustively analyze 920 datasets in under 30 minutes on a laptop. The claim matters because pairwise analyses can miss collective dependencies, and the combinatorial cost has previously blocked routine HOI use.

What carries the argument

The load-bearing object is the batched sub-covariance matrix pipeline built on the Gaussian copula entropy estimator. For data with T samples, the copula transform turns each variable's marginal into a standard normal via ranks, so the whole system is summarized by one N by N covariance matrix; every n-plet's entropy then reduces to a determinant of a submatrix. THOI forms a batch of these submatrices by masking the full covariance matrix with binary index tensors, computes all determinants at once in PyTorch, and combines them with additions and subtractions into total correlation, dual total correlation, $\Omega$, and S-information. A second mechanism, independent-variable padding, inserts an identity-matrix block so n-plets of different orders can share one batch, and the known entropy of a standard normal (1.4189 nats per padded variable) is subtracted afterward.

What would settle it

Compute the reference $\Omega$ for the synthetic Gaussian R- and S-systems of Supplementary A.5 with an independent closed-form calculation or with a different entropy estimator, and compare it against THOI's output; a systematic offset would show that the validation cannot detect such a bias.

Watch

Extended reading notes

Core claim

The paper's central claim is that the combinatorial explosion of HOI metrics can be defeated in practice by restructuring the computation as batched linear algebra. Instead of estimating joint probability densities directly, THOI applies the Gaussian copula transform once to obtain a covariance matrix, extracts the sub-covariance matrix for every k-variable combination using binary masks, and computes determinants in parallel batches to obtain entropies and therefore the total correlation, dual total correlation, $\Omega$, and S-information. The authors report that this makes exhaustive HOI computation over all combinations of a 30-variable system finish in 5.8 hours on a laptop, a task they extrapolate would take HOI_toolbox about 221 days and JIDT roughly 17 years. For systems beyond exhaustive reach, the paper provides greedy and simulated-annealing heuristics and shows on a 100-variable concatenated system that both recover the known synergistic and redundant subsystems. It also reports that deep anesthesia compresses the range of $\Omega$ values in fMRI data and that four principal components explain about 95 percent of the variance across 920 datasets.

Load-bearing premise

The accuracy validation assumes that THOI's Gaussian entropy and $\Omega$ formulas are themselves correct, because the reference ground-truth $\Omega$ values in Supplementary A.5 are computed with THOI using the same formulas, so a systematic implementation bias would be invisible to the comparison.

Editorial extensions

If this is right

  • Exhaustive HOI analysis becomes a routine laptop task for systems up to roughly 30 variables, turning a computation previously estimated at 221 days or 17 years into a same-day job.
  • All four measures (total correlation, dual total correlation, $\Omega$, and S-information) are produced in one run, and population-level summaries can be computed on the fly without storing the exponentially many individual values.
  • The greedy and simulated-annealing heuristics identify known synergistic and redundant subsystems in a 100-variable system, making heuristic HOI searches usable beyond exhaustive scale.
  • In the anesthesia application, the paper reports a compression of the $\Omega$ range, with an eight-region interaction shifting from synergy-dominated to redundancy-dominated and a 14-region interaction losing redundancy while remaining redundancy-dominated.
  • The 920-dataset benchmark supports a four-dimensional description of complex systems: overall interdependence, overall independence, proportion of synergistic n-plets, and O-information balance.

Reading between the lines

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

  • Because the Gaussian copula estimator only captures covariance-level dependence after rank normalization, the empirical $\Omega$ values for fMRI and real data are copula-level summaries; coupling that lives in higher moments or non-Gaussian structure could go undetected.
  • The masked-submatrix batching idea is not specific to information metrics and could accelerate other subset-indexed statistics, such as partial correlations, in the same memory regime.
  • The order at which $\Omega$ is maximized or minimized, which the heuristics can approximate cheaply, is treated in the paper as a proxy for the proportion of synergistic n-plets; testing that proxy on non-Gaussian synthetic families would clarify how general the link is.
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, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper introduces THOI, a PyTorch-based Python library for computing higher-order information-theoretic measures (TC, DTC, O-information, S-information) from continuous multivariate data using Gaussian-copula entropy estimation. The main claims are: (i) batched, parallelized computation of all subsets for systems up to about 30 variables, with reported timings (5.8 hours for all orders 3–30 on a 30-variable system) and memory footprint below 3 GB; (ii) heuristic search (greedy and simulated annealing) for larger systems, validated on synthetic probabilistic graphical models with known redundancy/synergy structure; (iii) empirical demonstrations on fMRI data in wakefulness vs. deep anesthesia and on a benchmark of 920 datasets. The paper's central efficiency claim is supported by benchmarks against HOI_toolbox, HOI, and JIDT, while the accuracy claim is validated only against ground-truth values that are themselves computed with THOI's own implementation of the Gaussian entropy and O-information formulas.

Significance. If the efficiency claims hold, THOI is a practically useful tool: exhaustive higher-order interaction analysis for N up to ~30 is currently infeasible in most open-source libraries, and the paper demonstrates a concrete speedup and low memory footprint, with the additional strength that the software is designed to run on standard laptops and supports batched multi-dataset and on-the-fly analyses. The heuristic validation on synthetic PGM systems is also valuable because it tests whether greedy and simulated annealing recover known redundancy/synergy subsystems. However, the paper's accuracy validation is weakened by the circularity identified in Supplementary A.5, and the benchmark section lacks error bars and does not include a code repository link, which limits the reproducibility of the headline numerical claims.

major comments (5)
  1. [Supplement A.5 and §2.3] The validation of metric accuracy is circular: Supplementary A.5 states 'Using the aforementioned ground-truth covariance matrices, the entropy equation (eq. 8) and THOI, we computes all the ground truth metrics of each system.' The PGM covariance matrices are analytically specified, but the ground-truth O-information values are evaluated with THOI's own implementation of Eq. 8 and Eq. 6. Any systematic bias in the determinant routine, the entropy constant, or the composition of Eq. 6 is shared by the reference and the measured values, so the reported agreement cannot detect such a bias. What the experiments do validate is the finite-sample behavior of the covariance/copula estimator and the behavior of the greedy and simulated-annealing heuristics. This point is load-bearing because the paper presents THOI as a tool whose accuracy on TC, DTC, O, and S is established; an independent reference implementation (e.g., analytic Gaussian entropies computed with a different library, or a non-parametric estimator on the same covariance-specified models) should be added to break the circularity.
  2. [§2.2 and Figure 1B,C] The headline speed comparisons are not accompanied by error bars or repeated runs, and the comparison may not be apples-to-apples: THOI uses batched PyTorch operations, while HOI_toolbox, HOI, and JIDT are reported 'without leveraging additional parallelization' and, for JIDT, run in their default mode. The extrapolations (JIDT would take ~17 years, HOI_toolbox ~221 days) are presented without uncertainty ranges. Since the central efficiency claim is quantitative, the paper should report at least three repeated timings per condition, state the number of CPU threads used for each library, and clarify whether the comparison libraries were given any parallelization or memory-usage settings that would be fair relative to THOI.
  3. [§3.4.1, Eq. for padding] The padding derivation states H(Σk + I^{n-k}) = H(Σk) + H(I^{n-k}) because the added component is independent. This is correct only if the block-diagonal structure of the padded covariance matrix is preserved; the text notes that the sampled sub-covariance matrix and the independent components 'are not sorted to be separated, but maintain the original positions of the variables in the full original covariance matrix.' Under that construction the padded matrix is not block-diagonal with an identity block in a fixed corner, so the stated identity H(Σk) = H(Σk + I^{n-k}) − (n−k)·1.4189 requires justification. The degree of freedom in the placement of the independent components changes the determinant of the padded matrix only through row/column permutations, so the identity may still hold, but the paper should state this explicitly and either prove it or reference the permutation-invariance of the determinant.
  4. [§2.4 and §3.6] The fMRI analysis claims large-effect-size differences with Cohen's |d| > 3 and Wilcoxon p < 0.001, but these are explicitly uncorrected for multiple comparisons ('Wilcoxon p >0.001' is the stated threshold in Figure 3C and the text says 'not corrected'). Given that the GA explores many candidate n-plets at many orders, the reported p-values are likely inflated by selection; the paper should report the actual number of hypotheses tested and apply an appropriate multiple-comparison correction or state clearly that the results are exploratory. This does not undermine the library's computational contribution, but it is load-bearing for the paper's empirical claim that anesthesia 'compresses' the range of O-information.
  5. [General reproducibility] The manuscript never provides a code repository URL or version number for THOI, and the benchmark environment (Intel Core i9, 64 GB RAM, Linux Mint 21) does not specify the number of CPU cores or the PyTorch version. Since the paper is about a software library and the main claims are computational, the absence of a repository link prevents the reader from verifying the implementation, re-running the benchmarks, or breaking the circular validation described in Supplement A.5. A public repository with the benchmark scripts should be added before publication.
minor comments (6)
  1. [Abstract and §1] The abstract says 'we analyzed over 900 real-world and synthetic datasets,' while §2.5 reports 920 datasets; the numbers should be consistent.
  2. [§3.6] The sentence 'The anesthesia analysis described in Section 3.6 was conducted using a publicly available dataset of 17 healthy adults' is self-referential; it should refer to Section 2.4 and should not describe the analysis in a section that is about the methods.
  3. [Supplement A.3.2, Eq. (16)] Equation (16) has a typo: 'Σ(rank(rank(X))/(T+1))' appears to use rank twice, and the intended rank transformation is not clearly defined; the paper should define rank(·) precisely and fix the duplicate application.
  4. [§3.3.1] In the enumeration of required terms, 'H(Xj)' is defined but the displayed list subsequent text speaks of 'the entire system,' which is fine; however, the notation H(X^n_{-j}) is introduced only in the list and should be defined before use in Eq. (6) or in the list itself.
  5. [Figure 3 and §2.4] Figure 3C uses 'Wilcoxon p >0.001' as a threshold, but the text says 'no significant difference was found (Wilcoxon p >0.001)'; this is either a typo (p < 0.001) or a mis-statement of the test result, and it should be corrected.
  6. [References] Reference [30] is cited for the Gaussian copula estimator but the supplementary text has placeholder '[cite ince, cite]' and '[cite]' in Supplement A.3; these should be replaced with actual citations, or removed if redundant.

Circularity Check

1 steps flagged · score 6.0 of 10

Accuracy validation is partially circular: 'ground-truth' Ω values are computed by THOI itself, so any bias in THOI's entropy or Ω implementation cancels in the comparison.

  1. self definitional [Supplementary Methods A.5, final paragraph; used in Results Section 2.3 and Discussion]
    "Using the aforementioned ground-truth covariance matrices, the entropy equation (eq. 8) and THOI, we computes all the ground truth metrics of each system."

    The validation of THOI's accuracy compares THOI's outputs to 'ground truth' values that are themselves produced by THOI: the same covariance matrices, the same Gaussian entropy formula (eq. 8), and the same library code are used to define the reference values and to compute the values being tested. Any systematic error in THOI's determinant, entropy, or Ω-composition code enters both sides identically and cancels in the comparison. The reported agreement therefore validates the finite-sample covariance estimation and heuristic search, but cannot validate the correctness of THOI's metric computation.

full rationale

The paper's central performance claims are externally grounded: the 30-variable exhaustive benchmark, the 221-day/17-year extrapolations for HOI_toolbox and JIDT, the <3GB memory footprint, and the 920-dataset run are empirical benchmarks against independent tools and real hardware, and are not circular. The optimization heuristics are also evaluated against the structure of the concatenated PGM (e.g., max/min Ω saturating at orders 20 and 40), which is an independent design property. The circularity is confined to the accuracy-validation loop: Supplementary A.5 defines the 'ground-truth' Ω values by applying eq. (8) and THOI to the PGM covariance matrices. Because the same implementation is used to generate the reference and the tested values, any bias in THOI's entropy/Ω code is invisible to the validation. This is load-bearing for the claim that THOI is an accurate tool for HOI analysis, though it does not undermine the speed/scalability results. No other circularity was found: the mathematical definitions of TC, DTC, Ω, and S are standard, the Gaussian copula estimator is cited to external literature, and the proposed PCA 'framework' is a post-hoc analysis of computed features rather than a derivation from the target result.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The core efficiency claim rests on standard linear algebra and PyTorch batch operations. The accuracy claim rests on the Gaussian-copula entropy model and on synthetic ground truth values that are computed with THOI itself, which is a circular validation. The free parameters are heuristic repeat counts and synthetic coupling constants chosen by hand.

free parameters (3)
  • Number of GA repeats = 10,000
    Chosen by hand to achieve convergence in Figure 2; results depend on this value (Supplementary Figure 6 shows convergence varies with repeats).
  • Number of SA repeats = 100
    Chosen by hand; SA achieves comparable results to GA with 100 repeats, but this is an arbitrary setting rather than a principled stopping rule.
  • Coupling parameter c in PGM synthetic systems = 0.5 (weak), 1 (strong)
    Hand-chosen to define weak and strong synergistic/redundant systems; not fitted to data, but it sets the scale of the ground truth.
assumptions (5)
  • standard math Sklar's theorem: any joint distribution can be separated into marginals and a copula (Supplementary A.3.1).
    Foundation for the Gaussian copula entropy estimation used throughout the paper.
  • standard math Gaussian entropy formula H(X) = 0.5 log((2πe)^n |Σ|) (Eq. 8).
    Closed-form entropy for multivariate Gaussian variables, used to compute all HOI measures.
  • standard math Bias correction formula for Gaussian entropy estimator (Eq. 9).
    Adopted from Schurmann 2004 [62]; needed for finite-sample entropy estimates.
  • domain assumption Additivity of Omega for independent components: A ⊥ B ⇒ Ω(A+B) = Ω(A) + Ω(B).
    Used in Supplementary A.5 to predefine maximal and minimal Omega in the concatenated 100-variable system; cited from [19].
  • domain assumption The Gaussian copula transformation preserves relevant dependency structure for the datasets analyzed.
    The whole method assumes that rank-transforming to a Gaussian copula yields accurate joint entropy estimates for continuous non-Gaussian data; the paper itself acknowledges this limitation in the Discussion.

how reviews work

0 comments
Cite this review

Pith. "Pith review of THOI: An efficient and accessible library for computing higher-order interactions enhanced by batch-processing." pith.science (2026). https://pith.science/paper/YYBNRUXE

@misc{pith2026250103381,
  author       = {Pith},
  title        = {Pith review of: THOI: An efficient and accessible library for computing higher-order interactions enhanced by batch-processing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YYBNRUXE}},
  note         = {Machine review of arXiv:2501.03381}
}
read the original abstract

Complex systems are characterized by nonlinear dynamics, multi-level interactions, and emergent collective behaviors. Traditional analyses that focus solely on pairwise interactions often oversimplify these systems, neglecting the higher-order interactions critical for understanding their full collective dynamics. Recent advances in multivariate information theory provide a principled framework for quantifying these higher-order interactions, capturing key properties such as redundancy, synergy, shared randomness, and collective constraints. However, two major challenges persist: accurately estimating joint entropies and addressing the combinatorial explosion of interacting terms. To overcome these challenges, we introduce THOI (Torch-based High-Order Interactions), a novel, accessible, and efficient Python library for computing high-order interactions in continuous-valued systems. THOI leverages the well-established Gaussian copula method for joint entropy estimation, combined with state-of-the-art batch and parallel processing techniques to optimize performance across CPU, GPU, and TPU environments. Our results demonstrate that THOI significantly outperforms existing tools in terms of speed and scalability. For larger systems, where exhaustive analysis is computationally impractical, THOI integrates optimization strategies that make higher-order interaction analysis feasible. We validate THOI accuracy using synthetic datasets with parametrically controlled interactions and further illustrate its utility by analyzing fMRI data from human subjects in wakeful resting states and under deep anesthesia. Finally, we analyzed over 900 real-world and synthetic datasets, establishing a comprehensive framework for applying higher-order interaction (HOI) analysis in complex systems.

Figures

Figures reproduced from arXiv: 2501.03381 by the authors.

Figure 1
Figure 1. Efficient computation of HOI using batch processing of covariance matrices. A) A set of multivariate time series X is transformed using the Gaussian copula approach, generating covariance matrices Σ for each dataset. The covariance matrices are then sub-sampled using a batch of k-plet indices, defined by a binary mask applied to Σ, yielding the sub-covariance matrices Σ k for each k-plet. These sub-covariance matric… view at source ↗
Figure 2
Figure 2. Within-order optimization with greedy and simulated annealing algorithms A, B) Maximum (red) and minimum (blue) Ω obtained by greedy and SA algorithms for a 100-variable system composed of strong/weak R and S systems and an independent system, each with 20 variables. Dashed horizontal lines indicate ground truth for the weak systems, and dotted lines represent the sum of weak and strong systems (red for R, blue for … view at source ↗
Figure 3
Figure 3. A) Estimated maximum and minimum Ω via the GA for awake (green) and deep anesthesia state (purple). Inset shows the reduction of minimum Ω at lower orders of interaction. B) Average maximum (red) and minimum (blue) effect size obtained from a GA tailored to amplify the difference between the two conditions. Shaded areas denote the range from the minimum to the maximum value for each optimization procedure. C) Distri… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: A) Spearman correlation matrix of features across datasets. Colors code different types of features. ’prop. syn-plets’ is the proportion of synergy-dominated n-plets out of the total number of n-plets. ’order max Ω’ and ’order min Ω’ is the order where Ω was maximized …
Figure 5
Figure 5. Figure 5: Sub-covariance matrices sampled with padding to allow different covariance matrix sizes in a single batch. 1) First, a mask is applied to the full covariance matrices using a masked encoding of the n-plets (each with a different number of masked variables) to obtain ea…
Figure 6
Figure 6. Figure 6: Dependence of the greedy algorithm with the number of repeats. Same as Figure 3A, C and E, but for a system with N=30 variables. Left panel shows the maximum (red) and minimum (blue) O-information for different number of repeats (initial conditions). Note that for redu…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Shannon invariants: A scalable approach to information decomposition

    cs.IT 2025-04 conditional novelty 5.0 of 10

    The average degree of redundancy equals the normalized sum of single-source mutual informations, and the average degree of vulnerability equals the normalized sum of leave-one-out conditional mutual informations.

Reference graph

Works this paper leans on

65 extracted references · 58 canonical work pages · cited by 1 Pith paper

  1. [1]

    Jensen, H. J. Self-organized criticality: emergent complex behavior in physical and biological systems, vol. 10 (Cambridge university press, 1998)

  2. [2]

    Bar-Yam, Y .Dynamics of complex systems (CRC Press, 2019)

  3. [3]

    & Boccara, N

    Boccara, N. & Boccara, N. Modeling complex systems, vol. 1 (Springer, 2010)

  4. [4]

    Complex systems: Network thinking

    Mitchell, M. Complex systems: Network thinking. Artificial intelligence 170, 1194–1212 (2006)

  5. [5]

    & Petri, G

    Battiston, F. & Petri, G. Higher-order systems (Springer, 2022)

  6. [6]

    Battiston, F. et al. Networks beyond pairwise interactions: Structure and dynamics. Physics reports 874, 1–92 (2020)

  7. [7]

    Boccaletti, S. et al. The structure and dynamics of networks with higher order interactions. Physics Reports 1018, 1–64 (2023)

  8. [8]

    E., Mediano, P

    Rosas, F. E., Mediano, P. A., Gastpar, M. & Jensen, H. J. Quantifying high-order interdependencies via multivariate extensions of the mutual information. Physical Review E 100, 032305 (2019)

Show all 65 references
  1. [9]

    F., Pope, M., Faskowitz, J

    Varley, T. F., Pope, M., Faskowitz, J. & Sporns, O. Multivariate information theory uncovers synergistic subsystems of the human cerebral cortex. Communications biology 6, 451 (2023)

  2. [10]

    & Beggs, J

    Timme, N., Alford, W., Flecker, B. & Beggs, J. M. Synergy, redundancy, and multivariate information measures: an experimentalist’s perspective. Journal of computational neuroscience 36, 119–140 (2014)

  3. [11]

    James, R. G. & Crutchfield, J. P. Multivariate dependence beyond shannon information. Entropy 19, 531 (2017)

  4. [12]

    Shannon, C. E. A mathematical theory of communication. The Bell system technical journal 27, 379–423 (1948)

  5. [13]

    Cover, T. M. & Thomas, J. A. Information theory and statistics. Elements of information theory 1, 279–335 (1991)

  6. [14]

    P., Koberle, R., Van Steveninck, R

    Strong, S. P., Koberle, R., Van Steveninck, R. R. D. R. & Bialek, W. Entropy and information in neural spike trains. Physical review letters 80, 197 (1998)

  7. [15]

    & Bialek, W

    Rieke, F., Warland, D. & Bialek, W. Coding efficiency and information rates in sensory neurons. Europhysics Letters 22, 151 (1993)

  8. [16]

    Casali, A. G. et al. A theoretically based index of consciousness independent of sensory processing and behavior. Science translational medicine 5, 198ra105–198ra105 (2013). 16 THOI Torch High-Order Interactions

  9. [17]

    & Edelman, G

    Tononi, G., Sporns, O. & Edelman, G. M. A measure for brain complexity: relating functional segregation and integration in the nervous system. Proceedings of the National Academy of Sciences 91, 5033–5037 (1994)

  10. [18]

    Castro, P. et al. Dynamical structure-function correlations provide robust and generalizable signatures of consciousness in humans. Communications Biology 7, 1224 (2024)

  11. [19]

    E., Gutknecht, A., Mediano, P

    Rosas, F. E., Gutknecht, A., Mediano, P. A. & Gastpar, M. Characterising high-order interdependence via entropic conjugation. arXiv preprint arXiv:2410.10485 (2024)

  12. [20]

    Gatica, M. et al. High-order interdependencies in the aging brain. Brain connectivity 11, 734–744 (2021)

  13. [21]

    Herzog, R. et al. High-order brain interactions in ketamine during rest and task: a double-blinded cross-over design using portable eeg on male participants. Translational Psychiatry 14, 310 (2024)

  14. [22]

    Changes in high-order interaction measures of synergy and redundancy during non-ordinary states of consciousness induced by meditation, hypnosis, and auto-induced cognitive trance

    Kumar, P.et al. Changes in high-order interaction measures of synergy and redundancy during non-ordinary states of consciousness induced by meditation, hypnosis, and auto-induced cognitive trance. NeuroImage 293, 120623 (2024)

  15. [23]

    Stramaglia, S., Scagliarini, T., Daniels, B. C. & Marinazzo, D. Quantifying dynamical high-order interdepen- dencies from the o-information: an application to neural spiking dynamics. Frontiers in Physiology11, 595736 (2021)

  16. [24]

    L., Varley, T

    Newman, E. L., Varley, T. F., Parakkattu, V . K., Sherrill, S. P. & Beggs, J. M. Revealing the dynamics of neural information processing with multivariate information decomposition. Entropy 24, 930 (2022)

  17. [25]

    Scagliarini, T. et al. Gradients of o-information: Low-order descriptors of high-order dependencies. Physical Review Research 5, 013025 (2023)

  18. [26]

    & Rosas, F

    Scagliarini, T., Marinazzo, D., Guo, Y ., Stramaglia, S. & Rosas, F. E. Quantifying high-order interdependencies on individual patterns via the local o-information: Theory and applications to music analysis. Physical Review Research 4, 013184 (2022)

  19. [27]

    M., Rosas, F

    Medina-Mardones, A. M., Rosas, F. E., Rodríguez, S. E. & Cofré, R. Hyperharmonic analysis for the study of high-order information-theoretic signals. Journal of Physics: Complexity 2, 035009 (2021)

  20. [28]

    Lizier, J. T. Jidt: An information-theoretic toolkit for studying the dynamics of complex systems. Frontiers in Robotics and AI 1, 11 (2014)

  21. [29]

    Non-parametric entropy estimation toolbox (npeet)

    Ver Steeg, G. Non-parametric entropy estimation toolbox (npeet). Non-parametric entropy estimation toolbox (NPEET) (2000)

  22. [30]

    Ince, R. A. et al. A statistical framework for neuroimaging data analysis based on mutual information estimated via a gaussian copula. Human brain mapping 38, 1541–1573 (2017)

  23. [31]

    Mahajan Pranav, M. D. Hoi toolbox: Higher-order interactions toolbox. https://github.com/ PranavMahajan25/HOI_toolbox (2021)

  24. [32]

    Neri, M. et al. Hoi: A python toolbox for high-performance estimation of higher-order interactions from multivariate data. Journal of Open Source Software 9, 7360 (2024)

  25. [33]

    & Sun, Z

    Ma, J. & Sun, Z. Mutual information is copula entropy. Tsinghua Science & Technology 16, 51–54 (2011)

  26. [34]

    Paszke, A. et al. PyTorch: An imperative style, high-performance deep learning library. In Wallach, H.et al. (eds.) Advances in Neural Information Processing Systems 32, 8024–8035 (Curran Associates, Inc., 2019). URL https: //proceedings.neurips.cc/paper/2019/file/bdbca288fee7...

  27. [35]

    & Grassberger, P

    Kraskov, A., Stögbauer, H. & Grassberger, P. Estimating mutual information. Physical review E 69, 066138 (2004)

  28. [36]

    Herzog, R. et al. Genuine high-order interactions in brain networks and neurodegeneration. Neurobiology of Disease 175, 105918 (2022)

  29. [37]

    Hourican, C. et al. Efficient search algorithms for identifying synergistic associations in high-dimensional datasets. Entropy 26, 968 (2024)

  30. [38]

    Nemirovsky, I. E.et al. An implementation of integrated information theory in resting-state fmri. Communications Biology 6, 692 (2023)

  31. [39]

    Demertzi, A. et al. Human consciousness is supported by dynamic complex patterns of brain signal coordination. Science advances 5, eaat7603 (2019)

  32. [40]

    Sarasso, S. et al. Consciousness and complexity: a consilience of evidence. Neuroscience of Consciousness 2021, niab023 (2021). 17 THOI Torch High-Order Interactions

  33. [41]

    M., Bryant, A

    Cliff, O. M., Bryant, A. G., Lizier, J. T., Tsuchiya, N. & Fulcher, B. D. Unifying pairwise interactions in complex dynamics. Nature Computational Science 3, 883–893 (2023)

  34. [42]

    Nonnegative entropy measures of multivariate symmetric correlations

    Te Sun, H. Nonnegative entropy measures of multivariate symmetric correlations. Information and Control 36, 133–156 (1978)

  35. [43]

    Information theoretical analysis of multivariate correlation.IBM Journal of research and development 4, 66–82 (1960)

    Watanabe, S. Information theoretical analysis of multivariate correlation.IBM Journal of research and development 4, 66–82 (1960)

  36. [44]

    Sequence to sequence learning with neural networks

    Sutskever, I. Sequence to sequence learning with neural networks. arXiv preprint arXiv:1409.3215 (2014)

  37. [45]

    Neural machine translation by jointly learning to align and translate.arXiv preprint arXiv:1409.0473 (2014)

    Bahdanau, D. Neural machine translation by jointly learning to align and translate.arXiv preprint arXiv:1409.0473 (2014)

  38. [46]

    Attention is all you need

    Vaswani, A. Attention is all you need. Advances in Neural Information Processing Systems (2017)

  39. [47]

    H., Leiserson, C

    Cormen, T. H., Leiserson, C. E., Rivest, R. L. & Stein, C. Greedy algorithms. In Introduction to Algorithms, chap. 16, 414–443 (MIT Press, Cambridge, MA, 2009), 3rd edn

  40. [48]

    Kirkpatrick, S., Gelatt Jr, C. D. & Vecchi, M. P. Optimization by simulated annealing. science 220, 671–680 (1983)

  41. [49]

    Nemirovsky, I. E. et al. An implementation of integrated information theory in resting-state fmri. https: //openneuro.org/datasets/ds003171/versions/2.0.1 (2023). OpenNeuro dataset ds003171

  42. [50]

    A., Marshall, G

    Fernández Slezak, D., Suárez, C., Cecchi, G. A., Marshall, G. & Stolovitzky, G. When the optimal is not the best: parameter estimation in complex biological models. PloS one 5, e13283 (2010)

  43. [51]

    & Koch, C

    Tononi, G., Boly, M., Massimini, M. & Koch, C. Integrated information theory: from consciousness to its physical substrate. Nature reviews neuroscience17, 450–461 (2016)

  44. [52]

    Luppi, A. I. et al. A synergistic core for human brain evolution and cognition. Nature Neuroscience 25, 771–782 (2022)

  45. [53]

    Herzog, R. et al. Neural mass modeling for the masses: Democratizing access to whole-brain biophysical modeling with fastdmf. Network Neuroscience 1–23 (2024)

  46. [54]

    Gatica, M. et al. High-order functional redundancy in ageing explained via alterations in the connectome in a whole-brain model. PLoS computational biology 18, e1010431 (2022)

  47. [55]

    Santos, F. A. et al. Emergence of high-order functional hubs in the human brain. bioRxiv 2023–02 (2023)

  48. [56]

    & Amico, E

    Santoro, A., Battiston, F., Lucas, M., Petri, G. & Amico, E. Higher-order connectomics of human brain function reveals local topological signatures of task decoding, individual identification, and behavior. Nature Communications 15, 10244 (2024)

  49. [57]

    Hindriks, R. et al. Higher-order functional connectivity analysis of resting-state functional magnetic resonance imaging data using multivariate cumulants. Human brain mapping 45, e26663 (2024)

  50. [58]

    J., Westlund, E

    Golubski, A. J., Westlund, E. E., Vandermeer, J. & Pascual, M. Ecological networks over the edge: hypergraph trait-mediated indirect interaction (tmii) structure. Trends in ecology & evolution 31, 344–354 (2016)

  51. [59]

    Grilli, J., Barabás, G., Michalska-Smith, M. J. & Allesina, S. Higher-order interactions stabilize dynamics in competitive network models. Nature 548, 210–213 (2017)

  52. [60]

    Alvarez-Rodriguez, U. et al. Evolutionary dynamics of higher-order interactions in social networks. Nature Human Behaviour 5, 586–595 (2021)

  53. [61]

    & Karsai, M

    Cencetti, G., Battiston, F., Lepri, B. & Karsai, M. Temporal properties of higher-order interactions in social networks. Scientific reports 11, 7028 (2021)

  54. [62]

    Bias analysis in entropy estimation

    Schürmann, T. Bias analysis in entropy estimation. Journal of Physics A: Mathematical and General 37, L295 (2004)

  55. [63]

    Dependence modeling with copulas (CRC press, 2014)

    Joe, H. Dependence modeling with copulas (CRC press, 2014)

  56. [64]

    Fonctions de répartition à n dimensions et leurs marges

    Sklar, M. Fonctions de répartition à n dimensions et leurs marges. In Annales de l’ISUP, vol. 8, 229–231 (1959)

  57. [65]

    & Friedman, N

    Koller, D. & Friedman, N. Probabilistic graphical models: principles and techniques (MIT press, 2009). 18 THOI Torch High-Order Interactions A Supplementary Materials A.1 Key concepts HOI hold the potential to uncover intricate statistical relationships in complex systems by o...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.