Pith. sign in

REVIEW 4 major objections 4 minor 42 references

Theoretical and Practical Analysis of Fr\'echet Regression via Comparison Geometry

T0 review · 4 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read Nonparametric Fréchet regression on bounded-diameter CAT(K) spaces is claimed to attain Euclidean-type convergence rates, with exponential concentration of sample Fréchet means as the supporting mechanism.

desk verdict The paper's central rate theorem is not proven—its own proof derives a slower variance term—and the concentration bound is an unclosed sketch; the existence/uniqueness part is standard. read the letter →

arxiv 2502.01995 v1 pith:KHNDB7MD submitted 2025-02-04 stat.ML cs.AIcs.LG

Masanari Kimura , Howard Bondell This is my paper · ORCID
classification stat.MLcs.AIcs.LG MSC 62G0862R3053C2362G20
keywords FréchetregressionCAT(K)spacescomparisongeometrynonparametricconcentrationinequalitiescurvatureboundsmanifold-valueddatakernelsmoothing
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

This paper aims to show that nonparametric Fréchet regression—regression when the response lies in a metric space—keeps the classical Euclidean rates of kernel smoothing whenever the space satisfies the CAT(K) curvature condition. The central result states that for a complete CAT(K) space of diameter at most $D$, a $\beta$-Hölder regression function, and standard kernel weights, the mean squared error obeys $\sup_{x \in X_0} \mathbb{E}[d^2(\hat{\mu}^*_n(x), \mu^*(x))] = O(1/(n h_n^d) + h_n^{2\beta})$, the usual bias-variance trade-off. The paper also claims an exponential concentration inequality for sample Fréchet means, $L^p$ convergence at rate $n^{-p/2}$, and angle-stability results linking curvature to the local directional geometry of the estimator. If these claims hold, curvature bounds—not manifold smoothness—are what govern the statistical difficulty, with non-positive curvature requiring no diameter restriction.

What carries the argument

The load-bearing object is the strong geodesic convexity constant $\alpha(K,D)$ of the squared distance function: in a CAT(K) space of diameter $D$, the Fréchet functional satisfies $F(z) - F(\mu) \geq \alpha(K,D) d^2(z,\mu)$, with $\alpha = 1/2$ for $K \leq 0$ and a positive curvature-dependent constant for $K > 0$ when $D < \pi/(2\sqrt{K})$. This inequality turns statistical error into a functional gap, which is then controlled by a uniform empirical-process tail bound for $\sup_{z \in M} |F_n(z) - F(z)|$. The secondary machinery is comparison-geometry angle technology: comparison triangles in the constant-curvature model space and Alexandrov angles, which supply the angle-stability lemmas and the local jet expansion of the Fréchet functional.

What would settle it

A Monte Carlo check on a low-dimensional sphere with known $K>0$ and diameter $D < \pi/(2\sqrt{K})$ could settle the concentration claim: compare the observed tail $P(d(\hat{\mu}_n, \mu) > \epsilon)$ with the bound in Theorem 3.7. If the tail decays like $\exp(-c n \epsilon^2)$ rather than $\exp(-c n \epsilon^4)$, or if the prefactor $(\alpha(K,D) D/\delta)^m$ cannot be made finite with any explicit covering radius $\delta$, then the exponential concentration step—and the rate theorem that integrates it—fails as stated.

Watch

Extended reading notes

Core claim

The paper's discovery is a set of comparison-geometry guarantees for Fréchet regression in CAT(K) spaces. It proves existence and uniqueness of conditional Fréchet means under diameter constraints, gives an exponential concentration bound for the sample Fréchet mean with constant $\alpha(K,D)$ coming from strong geodesic convexity, and establishes pointwise almost-sure consistency for kernel-weighted estimators. The rate theorem, stated as Theorem 3.11, asserts that in a complete CAT(K) space of diameter at most $D$, with a $\beta$-Hölder regression function and standard kernel weights, $\sup_{x \in X_0} \mathbb{E}[d^2(\hat{\mu}^*_n(x), \mu^*(x))] = O(1/(n h_n^d) + h_n^{2\beta})$, matching Euclidean nonparametric rates. The proof runs through a bias-variance decomposition with a local population measure and uses strong convexity of the Fréchet functional to convert distance error into functional gap; a separate set of Alexandrov-angle lemmas shows that angles at the conditional Fréchet mean vary Lipschitzly with Wasserstein perturbations of the conditional measures.

Load-bearing premise

The load-bearing premise is that the empirical Fréchet functional converges uniformly over the whole space with a specific exponential tail; the paper asserts this bound, with constants depending on an undefined net radius, rather than proving it.

Editorial extensions

If this is right

  • Bandwidth selection for Fréchet regression can follow the Euclidean recipe: balance $1/(n h_n^d)$ against $h_n^{2\beta}$.
  • Negative-curvature CAT(K) spaces give convexity constant $1/2$ with no diameter restriction, so Hadamard-type spaces are the friendliest setting for the theory.
  • Positive-curvature spaces need support diameter below $\pi/(2\sqrt{K})$; beyond that, uniqueness and the rates can break down.
  • The exponential concentration and $L^p$ rate $n^{-p/2}$ imply sample Fréchet means are as efficient as Euclidean averages in the bounded-diameter regime.
  • Angle stability means directional features around the conditional Fréchet mean inherit Lipschitz continuity in the predictor, which matters for shape and directional statistics.

Reading between the lines

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

  • A testable extension the paper leaves implicit: stereographic projection into hyperbolic space acts as a variance-stabilizing transform for spherical responses, analogous to the log transform, and one could formally characterize when it reduces Fréchet regression risk under heteroscedastic angular noise.
  • Because the rate proof relies only on the weight LLN condition and strong geodesic convexity, similar comparison-geometry arguments should transfer to local-polynomial or random-forest-weighted Fréchet regression.
  • The appendix's $\epsilon$-approximate CAT(K) results suggest the whole theory can be made robust: with comparison error $\epsilon$, convexity degrades by $O(\epsilon D)$ and uniqueness weakens to a $O(\sqrt{\epsilon})$ neighborhood, extending the guarantees to nearly-CAT data.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper studies nonparametric Fréchet regression for metric-space-valued responses, focusing on complete CAT(K) spaces. It claims existence, uniqueness, and stability results for Fréchet means; an exponential concentration inequality for sample Fréchet means; L^p moment bounds; pointwise consistency of kernel-based Fréchet regression; and a sup-norm mean-squared-error rate O(1/(n h_n^d) + h_n^{2β}) for β-Hölder regression functions. The theoretical statements are in Section 3 with proofs in Appendix B. The empirical section compares Fréchet regression on spherical versus hyperbolic representations of spherical data and reports lower MSE for the hyperbolic mapping.

Significance. If the main results were valid, the paper would provide a useful unification: classical Euclidean nonparametric rates would transfer to bounded-curvature metric spaces, with constants depending only on K and the diameter, and the angle-stability and jet-expansion results would add geometric insight. The paper is clearly organized and makes concrete, falsifiable claims, and it includes Python code and dataset details for the experiments. However, the central proof chain has internal inconsistencies: the main concentration bound is asserted with undefined quantities and no derivation, the L^p proposition integrates a tail with the wrong exponent, and the variance term in the central rate theorem is computed at the square-root scale but reported at the linear scale. As a result, the paper's headline convergence-rate result is not established by the supplied arguments.

major comments (4)
  1. [B.2, Theorem 3.7] The core of the proof is an unstated uniform empirical-process bound. After a fixed-z Hoeffding inequality, the proof asserts P[sup_{z in M} |F_n(z)-F(z)| >= t] <= c'_1 exp(-c'_2 n t^2) with c'_1 = 2(alpha(K,D)D/delta)^m and c'_2 = alpha(K,D)/(8D^2), attributed to "standard references in manifold-valued statistics" without a citation. The net radius delta is never defined, the dimension m is not specified for a general CAT(K) space, and no chaining or Lipschitz argument connects a delta-net bound to the supremum over M. Since all downstream consistency and rate results depend on this bound, Theorem 3.7 is not proven as stated.
  2. [B.2, Theorem 3.7 vs. Proposition 3.8] The tail exponent in Theorem 3.7 is exp(-n(alpha(K,D) epsilon^2)^2/(8D^2)) = exp(-O(n epsilon^4)), but the proof of Proposition 3.8 integrates exp(-c_2 n epsilon^2) and concludes E[d^p(muhat_n, mu)] = O(n^{-p/2}). Integrating the epsilon^4 tail gives O(n^{-p/4}), so for p=2 it gives O(n^{-1/2}), not O(n^{-1}). Proposition 3.8's claimed L^p rate is therefore not a consequence of Theorem 3.7; the proof of Proposition 3.8 restates a tail bound that contradicts the theorem it cites.
  3. [B.2, Theorem 3.11] The proof derives E[d^2(muhat*_n(x), mutilde*_n(x))] <= E[Delta_n(x)]/alpha(K,D), and then explicitly states E[Delta_n(x)] = O((n h_n^d)^{-1/2}). The proof concludes a variance component of size O((n h_n^d)^{-1/2}), while the theorem's displayed rate (10) claims O(1/(n h_n^d)). No argument is supplied to convert the square-root bound into the claimed linear-in-1/(n h_n^d) bound. This is a missing factor of (n h_n^d)^{1/2}, not a harmless constant mismatch, so the stated squared-error rate does not follow from the proof.
  4. [B.2, Theorem 3.11] The proof also relies on the assertion that a straightforward Hoeffding/Bennett-type argument gives E[Delta_n(x)] = O((n h_n^d)^{-1/2}) even though muhat*_n(x) appears inside the empirical process term and depends on the whole sample. No Efron-Stein, bounded-differences, or U-statistic calculation is given. Even if the preceding concentration theorem were correct and the tail-exponent mismatch were fixed, this step would still need a rigorous derivation before the bias-variance decomposition in Theorem 3.11 is established.
minor comments (4)
  1. [Section 3.2, Assumption 3.9] Equation (6) writes E[f(x) | X = x], but f is a function on M, so the expression should be E[f(Y) | X = x] or an equivalent conditional expectation with respect to the response variable.
  2. [B.2, Theorem 3.11] In the displayed definition of Delta_n(x), the same term d^2(Y_i, muhat*_n(x)) appears in both summands; the second summand should involve d^2(Y_i, mutilde*_n(x)) or otherwise the expression does not match the two terms in the preceding display.
  3. [B.1, Lemma 3.2] The proof expands {d(p, m_n) - d(y, p)}^2 as d(p, m_n)^2 - 2d(p, m_n) + d^2(y, p); the linear term should be -2d(p, m_n)d(y, p). The displayed algebra and the subsequent bound need correction.
  4. [All sections] The manuscript alternates between symbols such as muhat in Theorem 3.7 and muhat_n elsewhere, and between mu^*, mu^*(x), and mu*_n(x). Standardizing the notation for the sample Fréchet mean and the local population mean would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper's central results are not defined in terms of its own prior work or fitted to the data they predict; missing references and a variance-rate inconsistency are correctness concerns, not circular reasoning.

full rationale

Walking the derivation chain, I find no step in which a prediction or first-principles result is equivalent by construction to its inputs. The Fréchet-mean existence/uniqueness results (Lemmas 3.1–3.3, Propositions 3.4–3.5) are standard CAT(K) arguments and do not presuppose the paper's own conclusions. Theorem 3.7's concentration bound and Proposition 3.8's Lp moment bound are stated with constants α(K,D), δ, and m that are not fitted to the quantities being predicted; even though several constants are left unspecified, this is an incompleteness, not a circular reduction. Theorem 3.11's rate proof does not fit the rate to data, and its bias-variance decomposition is not definitionally identical to the conclusion. The only self-citations (Kimura & Hino 2021, 2022; Kimura 2021; Kimura & Bondell 2024) appear in background remarks or the limitations paragraph and carry no load in the proofs. The paper does, however, contain missing support that should be weighed: in the proof of Theorem 3.7 the uniform empirical-process bound is asserted with constants "from standard references in manifold-valued statistics" without any citation, and δ is never defined; similarly, the proof of Theorem 3.11 appeals to "standard references on manifold-valued kernel regression" for the bound E[Δ_n(x)] = O((n h_n^d)^{-1/2}) but then inserts this into E[d²(μ̂*, μ̃*)] ≤ E[Δ]/α, yielding a variance term that does not match the claimed O(1/(n h_n^d)) in (10). These are proof-completeness and correctness defects, not evidence that the conclusions are circularly defined from their own inputs.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The central claims rest on standard CAT(K) assumptions, plus several unproved or informal assumptions: strict convexity, a uniform strong-convexity constant, smoothness of conditional distributions, and an empirical-process sup bound. No free parameters are fitted to data, but the constants alpha and delta in the main bounds are not actually specified.

free parameters (2)
  • alpha(K,D) strong convexity constant = not computed; asserted alpha = 1/2 for K <= 0 in Theorem 3.7 proof
    All concentration and rate results scale with this constant, but no rigorous value is derived. The formulas in the proof are asserted as 'one can take' and are not consistent with the theorem statement.
  • delta net radius in Theorem 3.7 = undefined
    The constant delta appears in the prefactor 2(alpha D / delta)^m in Eq. (4) and in c'_1, but no net construction, radius choice, or covering-number bound is given anywhere in the paper.
assumptions (6)
  • standard math CAT(K) comparison inequalities hold for all geodesic triangles used in proofs
    Definition 2.1; used in Lemma 3.1, Lemma 3.12, and all downstream arguments.
  • domain assumption The squared distance function is strictly geodesically convex on M
    Lemma 3.3 assumes this rather than proving it from CAT(K). Many CAT(0) spaces, such as trees, are not strictly geodesically convex in this sense.
  • domain assumption There exists a uniform strong convexity constant alpha(K,D) > 0 for all Fréchet functionals with support diameter D
    Invoked in the proofs of Theorems 3.7, 3.10, and 3.11. No derivation from CAT(K) axioms is provided, and the value changes across the proof.
  • domain assumption Conditional distributions P[Y in . | X = x] vary smoothly in x with an O(h^beta) local bias
    Theorem 3.11 includes this informally as a bullet assumption and uses it to bound d(tilde mu_n(x), mu*(x)) = O(h^beta); no quantitative smoothness condition is stated.
  • domain assumption Riemannian exponential maps and Taylor expansions exist at the Fréchet mean
    Section 3.4, Lemma 3.16, and Proposition 3.17 assume manifold smoothness and exp_z; general CAT(K) spaces need not be manifolds.
  • ad hoc to paper Empirical process sup bound over M with exponential tails holds uniformly
    The proof of Theorem 3.7 asserts the sup-norm bound with constants from 'standard references in manifold-valued statistics', but gives no reference or derivation, and the net radius delta is undefined.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Theoretical and Practical Analysis of Fr\'echet Regression via Comparison Geometry." pith.science (2026). https://pith.science/paper/KHNDB7MD

@misc{pith2026250201995,
  author       = {Pith},
  title        = {Pith review of: Theoretical and Practical Analysis of Fr\'echet Regression via Comparison Geometry},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KHNDB7MD}},
  note         = {Machine review of arXiv:2502.01995}
}
read the original abstract

Fr\'echet regression extends classical regression methods to non-Euclidean metric spaces, enabling the analysis of data relationships on complex structures such as manifolds and graphs. This work establishes a rigorous theoretical analysis for Fr\'echet regression through the lens of comparison geometry which leads to important considerations for its use in practice. The analysis provides key results on the existence, uniqueness, and stability of the Fr\'echet mean, along with statistical guarantees for nonparametric regression, including exponential concentration bounds and convergence rates. Additionally, insights into angle stability reveal the interplay between curvature of the manifold and the behavior of the regression estimator in these non-Euclidean contexts. Empirical experiments validate the theoretical findings, demonstrating the effectiveness of proposed hyperbolic mappings, particularly for data with heteroscedasticity, and highlighting the practical usefulness of these results.

Figures

Figures reproduced from arXiv: 2502.01995 by the authors.

Figure 1
Figure 1. Mapping from spherical data into hyperbolic space. [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Visualization of the HYG Stellar database. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Heteroscedasticity in the HYG Stellar dataset. [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Illustrative example of transformed responses. Under the heteroscedastic errors assumption, the appropriate [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: Visualizations for USGS Earthquake catalogue and NOAA Climate dataset. [PITH_FULL_IMAGE:figures/full_fig_p040_5.png]
Figure 6
Figure 6. Figure 6: Heteroscedasticity in the NOAA and USGS datasets. [PITH_FULL_IMAGE:figures/full_fig_p040_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 37 canonical work pages

  1. [1]

    The e-pca and m-pca: Dimension reduction of parameters by information geometry

    Akaho, S. The e-pca and m-pca: Dimension reduction of parameters by information geometry. In 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No. 04CH37541), volume 1, pp.\ 129--134. IEEE, 2004

    work page 2004

  2. [2]

    Natural gradient works efficiently in learning

    Amari, S.-I. Natural gradient works efficiently in learning. Neural computation, 10 0 (2): 0 251--276, 1998

    1998

  3. [3]

    Information geometry and its applications, volume 194

    Amari, S.-i. Information geometry and its applications, volume 194. Springer, 2016

    2016

  4. [4]

    and Nagaoka, H

    Amari, S.-i. and Nagaoka, H. Methods of information geometry, volume 191. American Mathematical Soc., 2000

    work page 2000

  5. [5]

    Information geometry, volume 64

    Ay, N., Jost, J., V \^a n L \^e , H., and Schwachh \"o fer, L. Information geometry, volume 64. Springer, 2017

    work page 2017

  6. [6]

    Lectures on spaces of nonpositive curvature, volume 25

    Ballmann, W. Lectures on spaces of nonpositive curvature, volume 25. Springer Science & Business Media, 1995

    work page 1995

  7. [7]

    and M \"u ller, H.-G

    Bhattacharjee, S. and M \"u ller, H.-G. Single index fr \'e chet regression. The Annals of Statistics, 51 0 (4): 0 1770--1798, 2023

    work page 2023

  8. [8]

    Bierens, H. J. The nadaraya-watson kernel regression function estimator. 1988

    work page 1988

Show all 42 references
  1. [9]

    Bridson, M. R. and Haefliger, A. Metric spaces of non-positive curvature, volume 319. Springer Science & Business Media, 2013

    2013

  2. [10]

    M., Raich, R., Finn, W

    Carter, K. M., Raich, R., Finn, W. G., and Hero III, A. O. Information-geometric dimensionality reduction. IEEE Signal Processing Magazine, 28 0 (2): 0 89--99, 2011

    2011

  3. [11]

    and Grove, K

    Cheeger, J. and Grove, K. Metric and comparison geometry, volume 11. International Press, 2007

    2007

  4. [12]

    G., and Ebin, D

    Cheeger, J., Ebin, D. G., and Ebin, D. G. Comparison theorems in Riemannian geometry, volume 9. North-Holland publishing company Amsterdam, 1975

    1975

  5. [13]

    and M \"u ller, H.-G

    Chen, Y. and M \"u ller, H.-G. Uniform convergence of local fr \'e chet regression with applications to locating extrema and time warping for metric space valued trajectories. The Annals of Statistics, 50 0 (3): 0 1573--1592, 2022

    2022

  6. [14]

    C., Fletcher, P

    Davis, B. C., Fletcher, P. T., Bullitt, E., and Joshi, S. Population shape regression from random design data. International journal of computer vision, 90: 0 255--266, 2010

    2010

  7. [15]

    Spherical regression

    Downs, T. Spherical regression. Biometrika, 90 0 (3): 0 655--668, 2003

    2003

  8. [16]

    Applying inverse stereographic projection to manifold learning and clustering

    Eybpoosh, K., Rezghi, M., and Heydari, A. Applying inverse stereographic projection to manifold learning and clustering. Applied Intelligence, pp.\ 1--15, 2022

    2022

  9. [17]

    and Meyer, F

    Ferguson, D. and Meyer, F. G. Computation of the sample fr \'e chet mean for sets of large graphs with applications to regression. In Proceedings of the 2022 SIAM International Conference on Data Mining (SDM), pp.\ 379--387. SIAM, 2022

    2022

  10. [18]

    Application of the single index methodology to the local Fr \'e chet regression in the context of Object oriented data analysis (OODA)

    Ghosal, A. Application of the single index methodology to the local Fr \'e chet regression in the context of Object oriented data analysis (OODA) . University of California, Santa Barbara, 2023

    2023

  11. [19]

    Fr \'e chet single index models for object response regression

    Ghosal, A., Meiring, W., and Petersen, A. Fr \'e chet single index models for object response regression. Electronic Journal of Statistics, 17 0 (1): 0 1074--1112, 2023

    2023

  12. [20]

    and Petersen, P

    Grove, K. and Petersen, P. Comparison geometry, volume 30. Cambridge University Press, 1997

    1997

  13. [21]

    Robust nonparametric regression with metric-space valued output

    Hein, M. Robust nonparametric regression with metric-space valued output. Advances in neural information processing systems, 22, 2009

    2009

  14. [22]

    Nonpositive curvature: geometric and analytic aspects

    Jost, J. Nonpositive curvature: geometric and analytic aspects. Birkh \"a user, 2012

    2012

  15. [23]

    Generalized t-sne through the lens of information geometry

    Kimura, M. Generalized t-sne through the lens of information geometry. IEEE Access, 9: 0 129619--129625, 2021

    2021

  16. [24]

    and Bondell, H

    Kimura, M. and Bondell, H. Density ratio estimation via sampling along generalized geodesics on statistical manifolds. arXiv preprint arXiv:2406.18806, 2024

    2024 arXiv

  17. [25]

    and Hino, H

    Kimura, M. and Hino, H. -geodesical skew divergence. Entropy, 23 0 (5): 0 528, 2021

    2021

  18. [26]

    and Hino, H

    Kimura, M. and Hino, H. Information geometrically generalized covariate shift adaptation. Neural Computation, 34 0 (9): 0 1944--1977, 2022

    1944

  19. [27]

    and M \"u ller, H.-G

    Lin, Z. and M \"u ller, H.-G. Total variation regularized fr \'e chet regression for metric-space valued data. The Annals of Statistics, 49 0 (6): 0 3510--3533, 2021

    2021

  20. [28]

    Liu, D. C. and Nocedal, J. On the limited memory bfgs method for large scale optimization. Mathematical programming, 45 0 (1): 0 503--528, 1989

    1989

  21. [29]

    Information geometry of u-boost and bregman divergence

    Murata, N., Takenouchi, T., Kanamori, T., and Eguchi, S. Information geometry of u-boost and bregman divergence. Neural Computation, 16 0 (7): 0 1437--1481, 2004

    2004

  22. [30]

    Nadaraya, E. A. On estimating regression. Theory of Probability & Its Applications, 9 0 (1): 0 141--142, 1964

    1964

  23. [31]

    An elementary introduction to information geometry

    Nielsen, F. An elementary introduction to information geometry. Entropy, 22 0 (10): 0 1100, 2020

    2020

  24. [32]

    Peter, A. M. and Rangarajan, A. Information geometry for landmark shape analysis: Unifying shape representation and deformation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31 0 (2): 0 337--350, 2008

    2008

  25. [33]

    and M \"u ller, H.-G

    Petersen, A. and M \"u ller, H.-G. Fr \'e chet regression for random objects with euclidean predictors. The Annals of Statistics, 47 0 (2): 0 691--719, 2019

    2019

  26. [34]

    Random forest weighted local fr \'e chet regression with random objects

    Qiu, R., Yu, Z., and Zhu, R. Random forest weighted local fr \'e chet regression with random objects. Journal of Machine Learning Research, 25 0 (107): 0 1--69, 2024

    2024

  27. [35]

    and Han, K

    Song, D. and Han, K. Errors-in-variables fr 'echet regression with low-rank covariate approximation. Advances in Neural Information Processing Systems, 36: 0 80575--80607, 2023

    2023

  28. [36]

    and Hein, M

    Steinke, F. and Hein, M. Non-parametric regression between manifolds. Advances in neural information processing systems, 21, 2008

    2008

  29. [37]

    C., Wu, Y., and M \"u ller, H.-G

    Tucker, D. C., Wu, Y., and M \"u ller, H.-G. Variable selection for global fr \'e chet regression. Journal of the American Statistical Association, 118 0 (542): 0 1023--1037, 2023

    2023

  30. [38]

    Watson, G. S. Smooth regression analysis. Sankhy \=a : The Indian Journal of Statistics, Series A , pp.\ 359--372, 1964

    1964

  31. [39]

    and Wylie, W

    Wei, G. and Wylie, W. Comparison geometry for the bakry-emery ricci tensor. Journal of differential geometry, 83 0 (2): 0 337--405, 2009

    2009

  32. [40]

    Frequentist model averaging for global fr \'e chet regression

    Yan, X., Zhang, X., and Zhao, P. Frequentist model averaging for global fr \'e chet regression. IEEE Transactions on Information Theory, 2024

    2024

  33. [41]

    Dimension reduction for fr \'e chet regression

    Zhang, Q., Xue, L., and Li, B. Dimension reduction for fr \'e chet regression. Journal of the American Statistical Association, 119 0 (548): 0 2733--2747, 2024

    2024

  34. [42]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.