Pith. sign in

REVIEW 3 major objections 5 minor 63 references

Change Point Detection in the Frequency Domain with Statistical Reliability

T0 review · 3 major / 5 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 Frequency-domain change points can now be tested with p-values that control false positives exactly, even when detection relies on heuristic search.

desk verdict New and useful extension of SI to frequency-domain change points, but the central projection matrix is not a projection; the theorem is repairable with a real fix. read the letter →

arxiv 2502.03062 v2 pith:7QGF3XWV submitted 2025-02-05 stat.ML cs.LG

Akifumi Yamada , Tomohiro Shiraishi , Shuichi Nishino , Teruyuki Katsuoka , Kouichi Taji , Ichiro Takeuchi This is my paper · ORCID
classification stat.MLcs.LG
keywords frequency-domainchangepointdetectionselectiveinferencevalidp-valuesdiscreteFouriertransformtruncatedchidistributionsimulatedannealingparametricprogrammingtypeIerrorcontrol
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's goal is to attach a statistically valid p-value to every change point detected in the frequency domain of a time series. It argues that selective inference—conditioning on the very act of selecting a candidate—removes the double-dipping bias that makes naive p-values near zero even when nothing changed. The proposed method constructs a test statistic from the discrete Fourier transform, conditions on the full heuristic search that produced the candidate, and derives an exactly uniform null p-value. If this is right, engineers who monitor rotating machinery or biosignals can flag a spectral shift as real only when the number is small, with false-positive rate controlled at the chosen level.

What carries the argument

The load-bearing object is the test statistic $T_k(X) = \sigma^{-1}\|P_k X\|$, where $P_k$ is the orthogonal projection onto the subspace spanned by DFT vectors $v^{(d)}$ for the frequencies $d \in \mathcal{D}_k$ at which the candidate $\tau_k$ was detected, weighted by segment lengths and the symmetry constant $c^{(d)}_{\mathrm{sym}}$. Conditioning on the sufficient statistic $Q(X) = (V(X), U(X))$ and on the selection event $A(X)=A(x)$ reduces the null distribution of $T_k$ to a truncated $\chi$-distribution with $\mathrm{tr}(P_k)$ degrees of freedom and a truncation region $Z$. The region $Z$ is the exact set of $z$ for which the heuristic returns the observed candidate set; Algorithm 3 finds it by parametric programming, walking along the parametrized line and gluing together the over-conditioned intervals where every Metropolis decision is fixed.

What would settle it

Run the method on many independent null sequences (known $\sigma$, no true spectral change), record the selective p-value for the first detected candidate, and test uniformity; if $P(p \le 0.05)$ deviates from 0.05 beyond Monte Carlo error, the exactness claim fails. A more direct check: for a fixed observed sequence, evaluate $A(a + bz)$ on a fine grid of $z$ and compare the union of returned over-conditioned intervals against the set of $z$ with $A(a+bz)=A(x)$; any mismatch disproves the truncation-region claim.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2: the selective p-value defined in Eq. (17) satisfies $P_{H_{0,k}}(p^{\mathrm{selective}}_k \le \alpha \mid A(X)=A(x)) = \alpha$ for every $\alpha \in (0,1)$, and hence also unconditionally, so the type I error rate is exactly the nominal level. In plain terms, whenever the null hypothesis holds—no mean shift in any of the frequencies where a candidate was detected—the reported p-value is uniform, and the false-positive probability among reported change points is controlled. The discovery is that this uniformity survives even though the candidate set is produced by a simulated-annealing heuristic, because the line search of Algorithm 3 reconstructs the exact selection event by unioning over-conditioned regions along the one-dimensional slice $X = a + bz$.

Load-bearing premise

The p-values are exact only if the parametric line search of Algorithm 3 enumerates every interval of the slice where the simulated-annealing search returns the observed change-point set; if an interval is missed, the conditioning set is too large and the p-values stop being uniform.

Editorial extensions

If this is right

  • Reported frequency-domain change points carry p-values that are exactly uniform under the null, so a user who rejects at level $\alpha$ will see a false positive with probability at most $\alpha$.
  • The method identifies which frequencies actually shifted at a change point, enabling root-cause diagnosis rather than a single 'something changed' flag.
  • Because the selection-conditional p-value conditions on the heuristic process, the guarantee holds even though the candidate search is only approximately optimal.
  • Over-conditioned competitors control the error rate but waste power; the exact truncation region restores power while keeping the guarantee.
  • The conditioning machinery transfers to any optimizer whose decision boundaries along the slice are quadratic, which includes many simulated-annealing and piecewise-linear heuristics.

Reading between the lines

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

  • The line search's exactness is the linchpin; a cheap validation would be to compare Algorithm 3's $Z$ against a dense grid of $z$ values on real null data to confirm no interval is missed.
  • The same projection-and-condition recipe could test changes in phase, power, or cross-frequency coherence, since it only requires the statistic to be the norm of a Gaussian projection.
  • Correlated noise breaks the guarantee, as the paper's own experiments show; users should whiten or estimate the covariance before trusting $\alpha$.
  • Multivariate extension to multiple sensors is the natural next step—the DFT basis already handles complex spectra, so stacking sensors only changes the projection matrix.
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

3 major / 5 minor

Summary. The paper extends selective inference (SI) to change point detection in the frequency domain. Frequencies are obtained via short-time DFT; candidate change points are selected by a simulated-annealing heuristic that encourages alignment across frequencies, and each selected candidate is tested using a truncated-χ selective p-value. The authors claim exact type I error control (Theorem 2) and support this with synthetic experiments and a bearing dataset. The central construction represents the test statistic as the norm of a projection of the data, computes the truncation region by parametric programming along the line a+bz, and reports p-values that are shown empirically to control type I error while retaining power.

Significance. If the technical gaps are resolved, this is a timely and useful contribution: it is the first SI-based method for frequency-domain change point detection, it targets the practically important setting of changes that appear simultaneously across multiple frequencies, and it ships reproducible code and extensive experiments, including robustness checks for non-Gaussian and correlated noise. The general idea of conditioning on the SA selection event and removing over-conditioning by parametric programming is valuable and may transfer to other heuristic solvers. However, the current manuscript contains a load-bearing error in the definition of the projection matrix, so the main theorems are not established as written.

major comments (3)
  1. [§4.1, Eq. (12)–(14); Appendix B.1] The projection matrix P_k in Eq. (12) is not an orthogonal projection for middle frequencies. For d not in {0, M/2}, c_sym=2 and ||v^(d)||^2 = M/a_len^(d), so P_k v^(d) = 2 v^(d), hence P_k^2 v^(d) = 4 v^(d) ≠ P_k v^(d). Consequently Eq. (14)'s assertion 'P_k = P_k^2, P_k = P_k^T' is false, and the displayed equality T_k(X)=σ^{-1}||P_k X|| does not hold. This invalidates the proof of Theorem 1 in Appendix B.1, which relies on (I−P_k)P_k=0 and on X=a+bz being a real line. The same complex-vs-real problem appears in Appendix C.1: the cost matrix C^{(d)}_{s+1:e} is complex Hermitian, but Eq. (23) treats a^T C b as a real coefficient e1, which is not generally true. The construction is repairable by defining the real symmetric projection \tilde P_k = Σ_{d∈D_k} (a_len^(d)c_sym^(d)/M)(Re(v^(d))Re(v^(d))^T + Im(v^(d))Im(v^(d))^T), which satisfies \tilde P_k^2 = \tilde P_k and X^T \tilde P_k X = Σ (a_len^(d)c_sym^(d)/M)|v^(d)T X|^2, and by redoing the conditioning algebra and quadratic inequalities with this matrix. As written, Theorems 1 and 2 are unproved.
  2. [Appendix C.2, Eq. (24), Algorithm 3] The paper claims that Algorithm 3 computes the exact truncation region Z = {z : A(a+bz)=A(x)} by parametric programming, and this exactness is essential for Theorem 2's equality p_selective ~ Unif(0,1). However, no theorem or proof is given that the iterative interval-enumeration procedure terminates and recovers the whole region. The update 'S ← S ∪ Z_oc(a+bz) for z ∈ S_c' is not an implementable rule for a continuum of intervals; it is not specified how z is selected or how the breakpoints of A(a+bz) are characterized beyond the over-conditioned quadratic inequalities. If the computed region is only a subset of Z, the p-values are computed under over-conditioning and the claimed uniform property fails, although type I error control may remain conservative. This needs to be either rigorously proved or clearly stated as an approximate/exact-in-practice method.
  3. [§1, §4, §5.2] The paper repeatedly claims that 'the probability of the final detected CPs being false positives is theoretically guaranteed to be below the specified significance level' and that the final CP set is reliable. Theorem 2, however, provides a per-selection property for an individual selected candidate k, conditional on that candidate being selected; it does not control the probability that at least one of the multiple detected CPs is false. The experiments in §5.2 test only one randomly selected detected CP candidate per trial, so they do not measure the set-level false positive rate. The authors should clarify the exact nature of the guarantee (per-candidate conditional on selection) or incorporate a multiple-testing correction over the detected CPs and evaluate the corresponding error rate.
minor comments (5)
  1. [§2.1] There is a typo: 'desrcribe' should be 'describe' in the first sentence of Section 2.1.
  2. [§3.4] 'altough' should be 'although' in the paragraph on decrement of temperature.
  3. [§5 and figures] There are repeated misspellings: 'truely' should be 'truly' in the Figure 1 caption and figure descriptions in §5.3; 'T ype I Error Rate' in Figure 5 should be 'Type I Error Rate'.
  4. [§6] The sentence 'The results theoretically confirmed that our method provided an unbiased evaluation based on SI framework' overstates what theorems and simulations establish; 'theoretically' should be removed or the scope clarified, since the paper's theoretical guarantee is exactly what needs repair.
  5. [Appendix D.3] The correlated-noise experiment is reported as failing to control type I error for larger ρ; this limitation is honest and should be retained, but it would be useful to state in the main text that the theoretical guarantee is restricted to i.i.d. Gaussian noise and that the real-data analysis relies on the estimated-variance variant whose validity is only empirical.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the selective p-value is a standard post-selection conditional distribution, not a fit or self-citation reduction.

full rationale

The paper's central claim is a selective-inference validity statement: the p-value in Eq. (17) is defined as the tail probability of the conditional distribution of the test statistic given A(X)=A(x) and Q(X)=Q(x), Theorem 1 identifies that conditional law as a truncated chi distribution, and Theorem 2 applies the probability integral transform. Nothing is fitted to make the p-values uniform; the uniformity is derived from the Gaussian model and the conditioning construction. The test statistic and projection matrix are defined directly from the DFT contrasts and selected change points, not from the p-values, so there is no self-definitional or fitted-input-called-prediction step. The only potentially self-referential element is the citation of Duy & Takeuchi (2022) for the parametric-programming computation of the truncation region in Appendix C.2. That method is independently published, is used as an implementation tool for computing Z, and is not the justification for the conditional-distribution derivation in Theorems 1 and 2; it therefore does not make the central argument circular. A possible technical issue with P_k failing to be an orthogonal projection for middle frequencies would be a correctness defect in the proof, not a circularity, because it would not show that the claimed result is equivalent to its inputs by construction. The empirical comparisons are also self-contained rather than being manufactured by the selection procedure. Overall, the derivation is self-contained with at most a minor non-load-bearing self-citation.

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

The method introduces no new physical or statistical entities. The free parameters are hyperparameters of the heuristic candidate selection, which does not affect the validity of the p-values but affects power. The axioms are the standard SI assumptions plus the unproven exactness of the parametric programming enumeration and the real-symmetric-projection property.

free parameters (3)
  • kappa in gamma penalty = 0.5 (synthetic), 3 (real)
    gamma = kappa * M * sigma^2 * log T controls the penalty for merging CPs across frequencies. The paper states there is no theoretical method to define gamma and sets it heuristically (Appendix A, Eq. 20).
  • SA initial temperature c0 = set via acceptance ratio 0.5
    Initial temperature of simulated annealing is chosen by a preliminary experiment with acceptance ratio eta(c0)=0.5. This affects CP candidate selection but not p-value validity.
  • SA cooling factor lambda = 0.8
    Geometric cooling schedule with lambda=0.8 is set as a practical choice (Section 3.4).
assumptions (5)
  • domain assumption Noise epsilon follows i.i.d. N(0, sigma^2 I_N) with known or separately estimated variance
    Equation (1), Section 2.1. This is required for the truncated chi-distribution in Theorem 1. The paper shows robustness to non-Gaussian but not to strong correlation.
  • domain assumption STFT uses rectangular non-overlapping windows and N is a multiple of M
    Section 2.1. The DFT coefficient model and independence across time blocks rely on this.
  • ad hoc to paper The CP candidate selection algorithm A is deterministic after fixing the random seed
    Section 4, footnote 5. This is needed so the selection event A(X)=A(x) is well-defined.
  • ad hoc to paper The parametric programming line search exactly recovers the truncation region Z
    Appendix C.2 and Algorithm 3. The paper asserts exactness without a formal proof, and the p-value validity depends on it.
  • ad hoc to paper The complex test statistic can be represented as sigma^{-1} ||P_k X|| for a real symmetric projection P_k
    Equation (12)-(14). P_k is built from complex vectors v^(d); the paper asserts P_k = P_k^2 = P_k^T, but this is not shown and may not hold in the complex case.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Change Point Detection in the Frequency Domain with Statistical Reliability." pith.science (2026). https://pith.science/paper/7QGF3XWV

@misc{pith2026250203062,
  author       = {Pith},
  title        = {Pith review of: Change Point Detection in the Frequency Domain with Statistical Reliability},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7QGF3XWV}},
  note         = {Machine review of arXiv:2502.03062}
}
abstract

Effective condition monitoring in complex systems requires identifying change points (CPs) in the frequency domain, as the structural changes often arise across multiple frequencies. This paper extends recent advancements in statistically significant CP detection, based on Selective Inference (SI), to the frequency domain. The proposed SI method quantifies the statistical significance of detected CPs in the frequency domain using $p$-values, ensuring that the detected changes reflect genuine structural shifts in the target system. We address two major technical challenges to achieve this. First, we extend the existing SI framework to the frequency domain by appropriately utilizing the properties of discrete Fourier transform (DFT). Second, we develop an SI method that provides valid $p$-values for CPs where changes occur across multiple frequencies. Experimental results demonstrate that the proposed method reliably identifies genuine CPs with strong statistical guarantees, enabling more accurate root-cause analysis in the frequency domain of complex systems.

Figures

Figures reproduced from arXiv: 2502.03062 by the authors.

Figure 1
Figure 1. Demonstration of the proposed method. Panel (a) shows the original time series signal. Panel (b) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Illustrations of the three local search operations for frequency [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the local search operation which merges two adjacent CP locations. In this figure, [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (16 more)
Figure 4
Figure 4. Figure 4: Schematic illustration of the SI framework. A point in the data space [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Type I error rate. 0.04 0.08 0.12 0.16 0.0 0.2 0.4 0.6 0.8 1.0 Power Proposed OC Bonferroni (a) M = 512. 0.04 0.08 0.12 0.16 0.0 0.2 0.4 0.6 0.8 1.0 Power Proposed OC Bonferroni (b) M = 1024 [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Power. 14 [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 8
Figure 8. Figure 8: In panel (a), the time variation of a frequency spectrum (1920 Hz) where a CP candidate location [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]
Figure 7
Figure 7. Figure 7: Vibration signals of the four bearings in the set No.2 of the IMS bearing dataset. Each signal was [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Results of the CP candidate selection for the signal of bearing 1 in the frequency domain and the [PITH_FULL_IMAGE:figures/full_fig_p016_8.png]
Figure 9
Figure 9. Figure 9: Schematic illustration of the proposed line search method for the identification of the trunca [PITH_FULL_IMAGE:figures/full_fig_p025_9.png]
Figure 10
Figure 10. Figure 10: Computational time in the type I error rate and the power experiments. [PITH_FULL_IMAGE:figures/full_fig_p026_10.png]
Figure 11
Figure 11. Figure 11: Robustness of type I error control for estimated variance. [PITH_FULL_IMAGE:figures/full_fig_p028_11.png]
Figure 12
Figure 12. Figure 12: Robustness of type I error control for non-Gaussian noise. [PITH_FULL_IMAGE:figures/full_fig_p028_12.png]
Figure 13
Figure 13. Figure 13: Robustness of type I error control for correlation of noise. [PITH_FULL_IMAGE:figures/full_fig_p028_13.png]
Figure 14
Figure 14. Figure 14: Sensitivity study of type I error rate control for the hyper-parameter [PITH_FULL_IMAGE:figures/full_fig_p029_14.png]
Figure 15
Figure 15. Figure 15: Sensitivity study of power for the hyper-parameter [PITH_FULL_IMAGE:figures/full_fig_p029_15.png]
Figure 16
Figure 16. Figure 16: Results of bearing 2. In panel (b), p-values were actually computed by considering CP candidates of the 6th harmonic, 3240 Hz, and 3460 Hz (around the 14th and 15th harmonics). 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 Time (Day) 0 2 4 6 8 10 12 14 16 18 Amplitude …
Figure 17
Figure 17. Figure 17: Results of bearing 3. 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 Time (Day) 0 2 4 6 8 10 12 14 16 Amplitude Spectrum p-values Proposed: 0.823 OC: 0.584 Naive: 0.000 (a) Inference on a falsely detected CP can￾didate location for 3540 Hz (the 15th har￾monic) on 0.25–2…
Figure 18
Figure 18. Figure 18: Results of bearing 4. In panel (b), p-values were actually computed by considering not only a CP candidate of the 6th harmonic but also a CP candidate of 3440 Hz (around the 15th harmonic). 30 [PITH_FULL_IMAGE:figures/full_fig_p030_18.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

63 extracted references · 50 canonical work pages

  1. [1]

    Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing

    Emile Aarts and Jan Korst. Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing. John Wiley & Sons, Inc., 1989

    work page 1989

  2. [2]

    Time-dependent spectral analysis of nonstationary time series

    Sudeshna Adak. Time-dependent spectral analysis of nonstationary time series. Journal of the American Statistical Association, 93 0 (444): 0 1488--1501, 1998

    work page 1998

  3. [3]

    Algorithms for the optimal identification of segment neighborhoods

    Ivan E Auger and Charles E Lawrence. Algorithms for the optimal identification of segment neighborhoods. Bulletin of mathematical biology, 51 0 (1): 0 39--54, 1989

    work page 1989

  4. [4]

    Change-point analysis as a tool to detect abrupt climate variations

    Claudie Beaulieu, Jie Chen, and Jorge L Sarmiento. Change-point analysis as a tool to detect abrupt climate variations. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 370 0 (1962): 0 1228--1249, 2012

    work page 1962

  5. [5]

    Post-selection inference for quantifying uncertainty in changes in variance

    Rachel Carrington and Paul Fearnhead. Post-selection inference for quantifying uncertainty in changes in variance. arXiv preprint arXiv:2405.15670, 2024

    work page Pith review arXiv 2024

  6. [6]

    Thermodynamical approach to the traveling salesman problem: An efficient simulation algorithm

    Vladim \' r C ern \`y . Thermodynamical approach to the traveling salesman problem: An efficient simulation algorithm. Journal of optimization theory and applications, 45: 0 41--51, 1985

    work page 1985

  7. [7]

    A statistical change point model approach for the detection of dna copy number variations in array cgh data

    Jie Chen and Yu-Ping Wang. A statistical change point model approach for the detection of dna copy number variations in array cgh data. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 6 0 (4): 0 529--541, 2008

    work page 2008

  8. [8]

    Valid inference corrected for outlier removal

    Shuxiao Chen and Jacob Bien. Valid inference corrected for outlier removal. Journal of Computational and Graphical Statistics, 29 0 (2): 0 323--334, 2020

    2020

Show all 63 references
  1. [9]

    Selective inference for k-means clustering

    Yiqun T Chen and Daniela M Witten. Selective inference for k-means clustering. Journal of Machine Learning Research, 24 0 (152): 0 1--41, 2023

    2023

  2. [10]

    More powerful conditional selective inference for generalized lasso by parametric programming

    Vo Nguyen Le Duy and Ichiro Takeuchi. More powerful conditional selective inference for generalized lasso by parametric programming. Journal of Machine Learning Research, 23 0 (300): 0 1--37, 2022

    2022

  3. [11]

    Computing valid p-value for optimal changepoint by selective inference using dynamic programming

    Vo Nguyen Le Duy, Hiroki Toda, Ryota Sugiyama, and Ichiro Takeuchi. Computing valid p-value for optimal changepoint by selective inference using dynamic programming. In Advances in Neural Information Processing Systems, volume 33, pp.\ 11356--11367, 2020

    2020

  4. [12]

    Quantifying statistical significance of neural network-based image segmentation by selective inference

    Vo Nguyen Le Duy, Shogo Iwazaki, and Ichiro Takeuchi. Quantifying statistical significance of neural network-based image segmentation by selective inference. Advances in Neural Information Processing Systems, 35: 0 31627--31639, 2022

    2022

  5. [13]

    Cad-da: Controllable anomaly detection after domain adaptation by statistical inference

    Vo Nguyen Le Duy, Hsuan-Tien Lin, and Ichiro Takeuchi. Cad-da: Controllable anomaly detection after domain adaptation by statistical inference. In International Conference on Artificial Intelligence and Statistics, pp.\ 1828--1836. PMLR, 2024

    2024

  6. [14]

    A MOSUM procedure for the estimation of multiple random change points

    Birte Eichinger and Claudia Kirch. A MOSUM procedure for the estimation of multiple random change points . Bernoulli, 24: 0 526--564, 2018

    2018

  7. [15]

    Selective sequential model selection

    William Fithian, Jonathan Taylor, Robert Tibshirani, and Ryan Tibshirani. Selective sequential model selection. arXiv preprint arXiv:1512.02565, 2015

    2015 arXiv

  8. [16]

    Multiscale change point inference

    Klaus Frick, Axel Munk, and Hannes Sieling. Multiscale change point inference. Journal of the Royal Statistical Society Series B: Statistical Methodology, 76 0 (3): 0 495--580, 2014

    2014

  9. [17]

    Wild binary segmentation for multiple change-point detection

    Piotr Fryzlewicz. Wild binary segmentation for multiple change-point detection . The Annals of Statistics, 42 0 (6): 0 2243--2281, 2014

    2014

  10. [18]

    Multiple-change-point detection for auto-regressive conditional heteroscedastic processes

    Piotr Fryzlewicz and Suhasini Subba Rao. Multiple-change-point detection for auto-regressive conditional heteroscedastic processes. Journal of the Royal Statistical Society Series B: Statistical Methodology, 76 0 (5): 0 903--924, 2014

    2014

  11. [19]

    Selective inference for hierarchical clustering

    Lucy L Gao, Jacob Bien, and Daniela Witten. Selective inference for hierarchical clustering. Journal of the American Statistical Association, 119 0 (545): 0 332--342, 2024

    2024

  12. [20]

    Analysis of the rolling element bearing data set of the center for intelligent maintenance systems of the university of cincinnati

    William Gousseau, J \'e r \^o me Antoni, Fran c ois Girardin, and Julien Griffaton. Analysis of the rolling element bearing data set of the center for intelligent maintenance systems of the university of cincinnati. In CM2016, Charenton, France, 2016

    2016

  13. [21]

    Kernel change-point analysis

    Za \" d Harchaoui, Eric Moulines, and Francis R Bach. Kernel change-point analysis. In Advances in neural information processing systems, pp.\ 609--616, 2009

    2009

  14. [22]

    Exact post-selection inference for the generalized lasso path

    Sangwon Hyun, Max G'sell, and Ryan J Tibshirani. Exact post-selection inference for the generalized lasso path. Electronic Journal of Statistics, 12 0 (1): 0 1053--1097, 2018

    2018

  15. [23]

    Post-selection inference for changepoint detection algorithms with application to copy number variation data

    Sangwon Hyun, Kevin Z Lin, Max G'Sell, and Ryan J Tibshirani. Post-selection inference for changepoint detection algorithms with application to copy number variation data. Biometrics, 77 0 (3): 0 1037--1049, 2021

    2021

  16. [24]

    An algorithm for optimal partitioning of data on an interval

    Brad Jackson, Jeffrey D Scargle, David Barnes, Sundararajan Arabhi, Alina Alt, Peter Gioumousis, Elyus Gwin, Paungkaew Sangtrakulcharoen, Linda Tan, and Tun Tao Tsai. An algorithm for optimal partitioning of data on an interval. IEEE Signal Processing Letters, 12 0 (2): 0 105-...

    2005

  17. [25]

    Testing for a change in mean after changepoint detection

    Sean Jewell, Paul Fearnhead, and Daniela Witten. Testing for a change in mean after changepoint detection. Journal of the Royal Statistical Society Series B: Statistical Methodology, 84 0 (4): 0 1082--1104, 2022

    2022

  18. [26]

    Optimal detection of changepoints with a linear computational cost

    Rebecca Killick, Paul Fearnhead, and Idris A Eckley. Optimal detection of changepoints with a linear computational cost. Journal of the American Statistical Association, 107 0 (500): 0 1590--1598, 2012

    2012

  19. [27]

    Optimization by simulated annealing

    Scott Kirkpatrick, C Daniel Gelatt Jr, and Mario P Vecchi. Optimization by simulated annealing. science, 220 0 (4598): 0 671--680, 1983

    1983

  20. [28]

    Detecting abrupt changes in a piecewise locally stationary time series

    Michael Last and Robert Shumway. Detecting abrupt changes in a piecewise locally stationary time series. Journal of multivariate analysis, 99 0 (2): 0 191--214, 2008

    2008

  21. [29]

    Optimal segmentation of random processes

    Marc Lavielle. Optimal segmentation of random processes. IEEE Transactions on signal processing, 46 0 (5): 0 1365--1373, 1998

    1998

  22. [30]

    Exact post model selection inference for marginal screening

    Jason D Lee and Jonathan E Taylor. Exact post model selection inference for marginal screening. Advances in neural information processing systems, 27, 2014

    2014

  23. [31]

    Evaluating the statistical significance of biclusters

    Jason D Lee, Yuekai Sun, and Jonathan E Taylor. Evaluating the statistical significance of biclusters. Advances in neural information processing systems, 28, 2015

    2015

  24. [32]

    Exact post-selection inference, with application to the lasso

    Jason D Lee, Dennis L Sun, Yuekai Sun, and Jonathan E Taylor. Exact post-selection inference, with application to the lasso. The Annals of Statistics, 44 0 (3): 0 907--927, 2016

    2016

  25. [33]

    Bearing Data Set, NASA Ames Prognostics Data Repository (http://ti.arc.nasa.gov/project/prognostic-data-repository), NASA Ames Research Center, Moffett Field, CA , 2007

    Jay Lee, Hai Qiu, Gang Yu, Jing Lin, and Rexnord Technical Services. Bearing Data Set, NASA Ames Prognostics Data Repository (http://ti.arc.nasa.gov/project/prognostic-data-repository), NASA Ames Research Center, Moffett Field, CA , 2007

    2007

  26. [34]

    M-statistic for kernel change-point detection

    Shuang Li, Yao Xie, Hanjun Dai, and Le Song. M-statistic for kernel change-point detection. In Advances in Neural Information Processing Systems, pp.\ 3366--3374, 2015

    2015

  27. [35]

    More powerful post-selection inference, with application to the lasso

    Keli Liu, Jelena Markovic, and Robert Tibshirani. More powerful post-selection inference, with application to the lasso. arXiv preprint arXiv:1801.09037, 2018

    2018 arXiv

  28. [36]

    Selective inference in regression models with groups of variables

    Joshua R Loftus and Jonathan E Taylor. Selective inference in regression models with groups of variables. arXiv preprint arXiv:1511.01478, 2015

    2015 arXiv

  29. [37]

    A novel framework of change-point detection for machine monitoring

    Guoliang Lu, Yiqi Zhou, Changhou Lu, and Xueyong Li. A novel framework of change-point detection for machine monitoring. Mechanical Systems and Signal Processing, 83: 0 533--548, 2017

    2017

  30. [38]

    Graph-based structural change detection for rotating machinery monitoring

    Guoliang Lu, Jie Liu, and Peng Yan. Graph-based structural change detection for rotating machinery monitoring. Mechanical Systems and Signal Processing, 99: 0 73--82, 2018

    2018

  31. [39]

    On optimal multiple changepoint algorithms for large data

    Robert Maidstone, Toby Hocking, Guillem Rigaill, and Paul Fearnhead. On optimal multiple changepoint algorithms for large data. Statistics and computing, 27: 0 519--533, 2017

    2017

  32. [40]

    Fisher discriminant analysis with kernels

    Sebastian Mika, Gunnar Ratsch, Jason Weston, Bernhard Scholkopf, and Klaus-Robert Mullers. Fisher discriminant analysis with kernels. In Neural networks for signal processing IX: Proceedings of the 1999 IEEE signal processing society workshop (cat. no. 98th8468), pp.\ 41--48. ...

    1999

  33. [41]

    Valid p-value for deep learning-driven salient region

    Daiki Miwa, Vo Nguyen Le Duy, and Ichiro Takeuchi. Valid p-value for deep learning-driven salient region. In Proceedings of the 11th International Conference on Learning Representation, 2023

    2023

  34. [42]

    Vito M. R. Muggeo and Giada Adelfio. Efficient change point detection for genomic sequences of continuous measurements. Bioinformatics, 27 0 (2): 0 161--166, 2011

    2011

  35. [43]

    Tree-values: selective inference for regression trees

    Anna C Neufeld, Lucy L Gao, and Daniela M Witten. Tree-values: selective inference for regression trees. Journal of Machine Learning Research, 23 0 (305): 0 1--43, 2022

    2022

  36. [44]

    Circular binary segmentation for the analysis of array-based dna copy number data

    Adam B Olshen, E Seshan Venkatraman, Robert Lucito, and Michael Wigler. Circular binary segmentation for the analysis of array-based dna copy number data. Biostatistics, 5 0 (4): 0 557--572, 2004

    2004

  37. [45]

    E. S. Page. Continuous inspection schemes. Biometrika, 41 0 (1/2): 0 100--115, 1954

    1954

  38. [46]

    Real-time financial surveillance via quickest change-point detection methods

    Andrey Pepelyshev and Aleksey S Polunchenko. Real-time financial surveillance via quickest change-point detection methods. Statistics and Its Interface, 10 0 (1): 0 93--106, 2017

    2017

  39. [47]

    Performance evaluation of dna copy number segmentation methods

    Morgane Pierre-Jean, Guillem Rigaill, and Pierre Neuvial. Performance evaluation of dna copy number segmentation methods. Briefings in bioinformatics, 16 0 (4): 0 600--615, 2015

    2015

  40. [48]

    Detection of multiple structural breaks in multivariate time series

    Philip Preuss, Ruprecht Puchstein, and Holger Dette. Detection of multiple structural breaks in multivariate time series. Journal of the American Statistical Association, 110 0 (510): 0 654--668, 2015

    2015

  41. [49]

    Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics

    Hai Qiu, Jay Lee, Jing Lin, and Gang Yu. Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics. Journal of sound and vibration, 289 0 (4-5): 0 1066--1090, 2006

    2006

  42. [50]

    A review and comparison of changepoint detection techniques for climate data

    Jaxk Reeves, Jien Chen, Xiaolan L Wang, Robert Lund, and Qi Qi Lu. A review and comparison of changepoint detection techniques for climate data. Journal of applied meteorology and climatology, 46 0 (6): 0 900--915, 2007

    2007

  43. [51]

    Inference for l 2-boosting

    David R \"u gamer and Sonja Greven. Inference for l 2-boosting. Statistics and computing, 30 0 (2): 0 279--289, 2020

    2020

  44. [52]

    A cluster analysis method for grouping means in the analysis of variance

    Andrew Jhon Scott and Martin Knott. A cluster analysis method for grouping means in the analysis of variance. Biometrics, pp.\ 507--512, 1974

    1974

  45. [53]

    Selective inference for change point detection by recurrent neural network

    Tomohiro Shiraishi, Daiki Miwa, Vo Nguyen Le Duy, and Ichiro Takeuchi. Selective inference for change point detection by recurrent neural network. Neural Computation, pp.\ 1--33, 2024 a

    2024

  46. [54]

    Statistical test for attention maps in vision transformers

    Tomohiro Shiraishi, Daiki Miwa, Teruyuki Katsuoka, Vo Nguyen Le Duy, Kouichi Taji, and Ichiro Takeuchi. Statistical test for attention maps in vision transformers. In Proceedings of the 41st International Conference on Machine Learning, 2024 b

    2024

  47. [55]

    Valid and exact statistical inference for multi-dimensional multiple change-points by selective inference

    Ryota Sugiyama, Hiroki Toda, Vo Nguyen Le Duy, Yu Inatsu, and Ichiro Takeuchi. Valid and exact statistical inference for multi-dimensional multiple change-points by selective inference. arXiv preprint arXiv:2110.08989, 2021

    2021 arXiv

  48. [56]

    Computing valid p-values for image segmentation by selective inference

    Kosuke Tanizaki, Noriaki Hashimoto, Yu Inatsu, Hidekata Hontani, and Ichiro Takeuchi. Computing valid p-values for image segmentation by selective inference. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 9553--9562, 2020

    2020

  49. [57]

    Statistical learning and selective inference

    Jonathan Taylor and Robert J Tibshirani. Statistical learning and selective inference. Proceedings of the National Academy of Sciences, 112 0 (25): 0 7629--7634, 2015

    2015

  50. [58]

    Exact post-selection inference for sequential regression procedures

    Ryan J Tibshirani, Jonathan Taylor, Richard Lockhart, and Robert Tibshirani. Exact post-selection inference for sequential regression procedures. Journal of the American Statistical Association, 111 0 (514): 0 600--620, 2016

    2016

  51. [59]

    Conditional selective inference for robust regression and outlier detection using piecewise-linear homotopy continuation

    Toshiaki Tsukurimichi, Yu Inatsu, Vo Nguyen Le Duy, and Ichiro Takeuchi. Conditional selective inference for robust regression and outlier detection using piecewise-linear homotopy continuation. Annals of the Institute of Statistical Mathematics, 74 0 (6): 0 1197--1228, 2022

    2022

  52. [60]

    Selective inference for change point detection in multi-dimensional sequences

    Yuta Umezu and Ichiro Takeuchi. Selective inference for change point detection in multi-dimensional sequences. arXiv preprint arXiv:1706.00514, 2017

    2017 arXiv

  53. [61]

    Selective inference for latent block models

    Chihiro Watanabe and Taiji Suzuki. Selective inference for latent block models. Electronic Journal of Statistics, 15 0 (1): 0 3137--3183, 2021

    2021

  54. [62]

    Post selection inference with kernels

    Makoto Yamada, Yuta Umezu, Kenji Fukumizu, and Ichiro Takeuchi. Post selection inference with kernels. In International conference on artificial intelligence and statistics, pp.\ 152--160. PMLR, 2018

    2018

  55. [63]

    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.