betaFeedback

WorksMathematics

Polynomial-Time Solvability of the Multi-Dimensional Slope Change Problem

Abstract

Slope change detection seeks to identify changes in the trends of time-ordered signals. We study the problem of fitting a multidimensional time series by continuous piecewise-linear functions with a shared continuous breakpoint and unrestricted slopes. We consider both the dense setting, in which every dimension may change slope, and the sparse setting, in which at most kk dimensions may change slope. Although the resulting optimization problems are nonconvex mixed integer nonlinear programs and may have irrational optimal solutions even for rational data, we prove that all six variants, corresponding to the ℓ1\ell_1, ℓ2\ell_2, and ℓ∞\ell_\infty losses in both settings, are exactly solvable in polynomial time. For least-squares loss, we obtain efficient algorithms requiring O(JT)O(JT) operations in the dense setting and O(J2Tlog⁡J)O(J^2T\log J) operations in the sparse setting, where TT is the number of time points and JJ is the number of dimensions. We implement these least-squares algorithms and evaluate them on simulated data and three real-world time series, namely, surface temperatures (ERA5), hand-tracking data from a mixed-reality headset (HoloAssist), and turbofan engine sensor data (NASA C-MAPSS). The dense algorithm solves instances with up to J=105J=10^5 and T=100T=100, that is, 10710^7 observations, in less than 0.10.1 seconds on a laptop, while the sparse algorithm solves instances with about 10310^3 dimensions and up to T=240T=240 in under a second. Compared with a mixed-integer nonlinear programming (MINLP) formulation solved by Gurobi, our algorithms are several orders of magnitude faster (approximately 8.9×1038.9 \times 10^3 to 3.4×1063.4 \times 10^6 times) on instances where Gurobi terminates within its default optimality tolerance of 0.01%0.01\% optimality gap. On larger instances, Gurobi frequently reaches a 10,000-second time limit with substantial optimality gaps.

Introduction

Slope change detection seeks the location in a time-ordered sequence, called a breakpoint, at which the rate of change of an underlying signal changes. Unlike mean-shift detection, which seeks abrupt changes in the signal level, slope change detection considers changes in trend while preserving continuity. Slope-change models have been used in a variety of applications, including surface-temperature trends in climatology [2, 48, 7, 57], rainfall trends [20], changes in health outcomes measured repeatedly over time [17], and sensor-based monitoring [19]. In many applications, multiple signals are observed at each observation time, and a common event, such as a regime shift or a system fault, may change their trends simultaneously. This motivates fitting a slope change that is shared across the dimensions of a multidimensional time series.

Problem formulation

Let T≥4T \ge4 be an integer, let x1<x2<⋯<xTx_1 < x_2 < \cdots< x_T be rational time coordinates, and let yi=(yi1,…,yiJ)∈QJy_i = (y_i^1,\ldots,y_i^J) \in\mathbb{Q}^J be the observation at time xix_i, so that yijy_i^j is the observation of dimension j∈[J]j \in[J] at time xix_i. For a breakpoint z∈Rz \in\mathbb{R}, every continuous function that is affine on (−∞,z](-\infty,z] and on [z,∞)[z,\infty) can be written uniquely as

x↦α+βx+δ(x−z)+,x \mapsto\alpha+ \beta x + \delta(x-z)_+,

where (t)+:=max⁡{t,0}(t)_+ := \max\{t,0\}. We call zz the breakpoint of this function. This function has slope β\beta and intercept α\alpha before the breakpoint, and slope β+δ\beta+\delta and intercept α−δz\alpha-\delta z after it. Thus δ\delta is the change in slope, and continuity at zz is built into the representation. We fit one such function to each dimension, with a breakpoint common to all dimensions. Given z∈Rz \in\mathbb{R} and α,β,δ∈RJ\alpha,\beta,\delta\in\mathbb{R}^J, the residual of dimension jj at time xix_i is

rij=yij−αj−βjxi−δj(xi−z)+,i∈[T], j∈[J],(1)r_i^j = y_i^j - \alpha_j - \beta_j x_i - \delta_j(x_i-z)_+,\qquad i \in[T],\ j \in[J], \tag*{(1)}

and r∈RTJr \in\mathbb{R}^{TJ} denotes the vector of these TJTJ residuals. For p∈{1,2,∞}p \in\{1,2,\infty\}, define the loss

Lp(r):={∑i,j∣rij∣,p=1,∑i,j(rij)2,p=2,max⁡i,j∣rij∣,p=∞.L_p(r) := \begin{cases} \sum_{i,j} |r_i^j|, & p=1,\\ \sum_{i,j} (r_i^j)^2, & p=2,\\ \max_{i,j} |r_i^j|, & p=\infty. \end{cases}

Given data (xi,yi)i∈[T](x_i,y_i)_{i\in[T]} with xi∈Qx_i \in\mathbb{Q} and yi∈QJy_i \in\mathbb{Q}^J, and k∈{0,1,…,J}k \in\{0,1,\ldots,J\}, the slope change problem is

OPT⁡p,k:=min⁡z,α,β,δ,r{Lp(r):r satisfies (1), ∥δ∥0≤k, z∈[x2,xT−1]}.(2)\operatorname{OPT}_{p,k} := \min_{z,\alpha,\beta,\delta,r} \left\{L_p(r): r\text{ satisfies }{(1)},\ \|\delta\|_0 \le k,\ z \in[x_2,x_{T-1}]\right\}. \tag*{(2)}

The constraint ∥δ∥0≤k\|\delta\|_0 \le k allows at most kk dimensions to change slope, and each remaining dimension is fitted by a single line. We refer to the case k=Jk=J as the dense problem and to the case k<Jk<J as the sparse problem. The constraint z∈[x2,xT−1]z \in[x_2,x_{T-1}] ensures that at least two observations lie on each side of the breakpoint, since a change in slope cannot be inferred from a segment containing a single observation.

Three features make this problem nontrivial. First, the breakpoint zz is a continuous decision variable, and the terms δj(xi−z)+\delta_j(x_i-z)_+ introduce nonconvexity. Second, the breakpoint is shared across all dimensions, coupling their fitting problems. Third, the sparse problem introduces the additional combinatorial problem of selecting which dimensions may change slope. Although the problem can be formulated as a mixed-integer nonlinear program [47], we show that its special structure permits polynomial-time algorithms.

Related literature

Single-dimension time series. Most methods for a single time series allow several breakpoints but restrict the breakpoints to the observation times or a prespecified candidate set. Without continuity, dynamic programming finds exact solutions [38, 5, 39]. With continuity, [25] use functional dynamic programming with an ℓ0\ell_0 penalty on the number of slope changes, and [24] allow unevenly spaced observations and more flexible candidate sets. The convex alternative, ℓ1\ell_1 trend filtering [42, 55], penalizes the second differences of the fitted values, so its breakpoints arise implicitly at observation times. Fewer methods treat breakpoint locations as continuous. For one breakpoint under least squares, the setting closest to ours, [37] obtains the exact fit by examining each interval between consecutive observations, with statistical properties studied in [36, 26] and a grid-search alternative proposed by [44]. For several breakpoints, iterative linearization [46] guarantees only local optimality, while mixed-integer programming provides exact formulations [56, 49, 30, 47].

Multiple-dimension time series with shared breakpoints. Here, the breakpoints are common to all dimensions. [40] estimates a break in the intercept, slope, or both of a linear trend, common to all series in a large panel, by minimizing the total sum of squared residuals over all break dates, and [41] estimates the break date jointly with latent common factors in the errors. For shared changes in the mean, the group fused lasso [11] gives a convex formulation. [19] detects, online, a common change point after which a subset of the dimensions moves from a constant mean to a linear drift, with sparsity modeled through the probability that each dimension is affected. Sparse changes in the mean of high-dimensional time series are studied by [58] and [23]. These works differ from our setting in their restrictions on breakpoint locations, assumptions on the form of the slope change, or statistical detection objectives, and none establishes polynomial-time global optimization for the formulation (2). [47] extends their mixed-integer formulations to shared breakpoints. These formulations are exact, but they are not known to be solvable in polynomial time.

Parametric linear programming. [15] optimizes the value of min⁡{cTx:(A0+λA1)x≤b, x≥0}\min\{c^{\mathsf{T}}x : (A_0+\lambda A_1)x \le b,\ x \ge0\} over a scalar λ\lambda in an interval, extending sensitivity analysis [28]. They prove that both minimizing and maximizing this value are NP-hard, and identify easier cases where A1A_1 has entries of one sign or a single nonzero row or column. For p∈{1,∞}p \in\{1,\infty\} and fixed i∗i^*, our dense problem is the minimization version with λ=z\lambda= z. However, because of the absolute values, A1A_1 always has entries of both signs, and for J≥2J \ge2 it has JJ nonzero columns, so none of these cases applies. We nevertheless solve the dense and sparse problems exactly in polynomial time, even though the optimal zz may be irrational and the sparse problem is not an LP for fixed i∗i^*.

Complexity and connection to ReLU networks. Closely related function recovery problems are NP-hard [60], and [25] notes that it is unclear whether detecting multiple slope changes is NP-hard. We study the setting with a single breakpoint that is shared across dimensions and located continuously. This setting is also the global training of a ReLU network whose input is time, with a single hidden unit (x−z)+(x-z)_+ shared by JJ outputs with affine skip connections, where the breakpoint zz is restricted to [x2,xT−1][x_2,x_{T-1}]. Training ReLU networks is NP-hard for other architectures [22, 14, 29], and optimal weights can be irrational [1, 9]. These results do not imply hardness for ours. Exact polynomial-time training is known for one hidden layer with a single output and fixed input dimension [4], but this does not cover multiple outputs.

Main complexity results

Throughout, ⟨D⟩\langle\mathcal{D}\rangle denotes the binary encoding length of the data D:=(xi,yi)i∈[T]\mathcal{D}:=(x_i,y_i)_{i\in[T]}. The existence and representation of rational solutions are important issues in mixed-integer nonlinear programming. For example, mixed-integer quadratic programming admits polynomial-size certificates of feasibility [21], whereas systems of cubic polynomial inequalities with rational data may have only irrational feasible solutions, and deciding the existence of rational solutions is NP-hard [10]. Problem (2) exhibits both phenomena, that is, optimal breakpoints can be irrational, yet we show that optimal solutions admit polynomial-size algebraic representations and can be computed exactly in polynomial time. To see that optimal breakpoints can be irrational, let T=6T=6, J=2J=2, xi=i−1x_i=i-1 for i∈[6]i\in[6], y1=(−2,2,−2,−2,5,0)y^1=(-2,2,-2,-2,5,0), and y2=(−9,4,8,−6,−4,0)y^2=(-9,4,8,-6,-4,0). Then

OPT⁡2,2=74531−477755737420≈125.4128,\operatorname{OPT}_{2,2}=\frac{74531-\sqrt{477755737}}{420}\approx125.4128,

attained at z⋆=(−58563+5477755737)/49162≈1.0318z^\star=(-58563+5\sqrt{477755737})/49162\approx1.0318. Since 477755737477755737 is not a perfect square, both z⋆z^\star and OPT⁡2,2\operatorname{OPT}_{2,2} are irrational. This irrationality is caused by sharing the breakpoint across dimensions. For J=1J=1 and p=2p=2 a rational optimal breakpoint always exists (see § 3). We therefore specify what an algorithm returns.

Definition 1. An algorithm solves (2) if it returns an optimal breakpoint z⋆z^\star, represented as a real algebraic number by a univariate polynomial with rational coefficients together with an interval with rational endpoints that contains z⋆z^\star and no other root of the polynomial, and returns α⋆,β⋆,δ⋆\alpha^\star,\beta^\star,\delta^\star as explicit rational functions of z⋆z^\star with rational coefficients.

As we shall see, for p∈{2,∞}p\in\{2,\infty\}, our algorithms return an optimal breakpoint z⋆z^\star that is algebraic of degree at most two over Q\mathbb{Q}. For p=1p=1, its degree is at most max⁡{2,2J−2}\max\{2,2J-2\} in the dense problem and at most max⁡{2,2k−2}\max\{2,2k-2\} in the sparse problem. In all cases, the optimal fitting coefficients can be chosen in Q(z⋆)\mathbb{Q}(z^\star), the smallest subfield of R\mathbb{R} containing Q\mathbb{Q} and z⋆z^\star, which consists of the values at z⋆z^\star of rational functions with rational coefficients.

The encoding length of an algebraic number represented as in Definition 1 is the total binary encoding length of the coefficients of the polynomial and the endpoints of the interval. All rational and algebraic numbers arising in our algorithms admit representations of encoding length polynomial in TT, JJ, and ⟨D⟩\langle D\rangle. Standard algorithms for real algebraic numbers perform exact arithmetic, root isolation, sign evaluation, and comparison on these representations in polynomial time [6, 51]. Consequently, the optimal breakpoint, objective value, and fitting coefficients can also be approximated to within 2−t2^{-t} in time polynomial in TT, JJ, ⟨D⟩\langle D\rangle, and tt [16, 51].

Poly-time complexity

Theorem 1. For every p∈{1,2,∞}p\in\{1,2,\infty\} and k∈{0,1,…,J}k\in\{0,1,\ldots,J\}, problem (2) can be solved, in the sense of Definition 1, in time polynomial in TT, JJ, and ⟨D⟩\langle D\rangle. More precisely, the algorithms perform the arithmetic operations counted in Table 1. All numbers involved are rational or algebraic, with encoding length polynomial in TT, JJ, and ⟨D⟩\langle D\rangle. For p∈{2,∞}p\in\{2,\infty\}, these operations include, and the table counts among them, computations of roots of quadratic polynomials and exact comparisons of quadratic algebraic numbers. For p=1p=1, the algorithm additionally performs root isolations and algebraic-number comparisons, counted in Table 1. All these algebraic operations have polynomial bit complexity.

References

  1. [1]Mikkel Abrahamsen, Linda Kleist, and Tillmann Miltzow. Training neural networks is ℜ-complete. In Advances in Neural Information Processing Systems, volume 34, 2021.arxiv.org/abs/2102.09798
  2. [2]Fatemeh Aghaei A., Ewan T. Phillips, and Holger Kantz. Change point detection in ERA5 ground temperature time series. Climatic Change, 178(11):193, 2025.arxiv.org/abs/2507.15045
  3. [3]Andreas Anastasiou and Angelos Papanastasiou. Generalized multiple change-point detection in the structure of multivariate, possibly high-dimensional, data sequences. Statistics and Computing, 33(5):94, 2023.arxiv.org/abs/2211.06856
  4. [4]Raman Arora, Amitabh Basu, Poorya Mianjy, and Anirbit Mukherjee. Understanding deep neural networks with rectified linear units. In International Conference on Learning Representations, 2018.
  5. [5]Jushan Bai and Pierre Perron. Computation and analysis of multiple structural change models. Journal of Applied Econometrics, 18(1):1–22, 2003.DOI
  6. [6]Saugata Basu, Richard Pollack, and Marie-Françoise Roy. Algorithms in Real Algebraic Geometry, volume 10 of Algorithms and Computation in Mathematics. Springer, Berlin, second edition, 2006.
  7. [7]Claudie Beaulieu, Colin Gallagher, Rebecca Killick, Robert Lund, and Xueheng Shi. A recent surge in global warming is not detectable yet. Communications Earth & Environment, 5(1):576, 2024.DOI
  8. [8]Claude Berge. Topological Spaces. Oliver and Boyd, Edinburgh and London, 1963.
  9. [9]Daniel Bertschinger, Christoph Hertrich, Paul Jungeblut, Tillmann Miltzow, and Simon Weber. Training fully connected neural networks is ℜ-complete. In Advances in Neural Information Processing Systems, volume 36, pages 36222–36237, 2023.
  10. [10]Daniel Bienstock, Alberto Del Pia, and Robert Hildebrand. Complexity, exactness, and rationality in polynomial optimization. Mathematical Programming, 197(2):661–692, 2023.
  11. [11]Kevin Bleakley and Jean-Philippe Vert. The group fused lasso for multiple change-point detection, 2011. https://arxiv.org/abs/1106.4199.
  12. [12]Manuel Blum, Robert W. Floyd, Vaughan Pratt, Ronald L. Rivest, and Robert E. Tarjan. Time bounds for selection. Journal of Computer and System Sciences, 7(4):448–461, 1973.DOI
  13. [13]Tim Bollerslev. Generalized autoregressive conditional heteroskedasticity. Journal of econometrics, 31(3):307–327, 1986.
  14. [14]Digvijay Boob, Santanu S. Dey, and Guanghui Lan. Complexity of training ReLU neural network. Discrete Optimization, 44:100620, 2022.
  15. [15]Laurie Boveroux, Margarida Carvalho, Andrea Lodi, and Quentin Louveaux. On the complexity of linear programs with parametric constraint matrices, 2026. https://arxiv.org/abs/2610.20422.
  16. [16]Richard P. Brent and Paul Zimmermann. Modern Computer Arithmetic, volume 18 of Cambridge Monographs on Applied and Computational Mathematics. Cambridge University Press, Cambridge, 2010.DOI
  17. [17]Samuel L Brilleman, Laura D Howe, Rory Wolfe, and Kate Tilling. Bayesian piecewise linear mixed models with a random change point: an application to BMI rebound in childhood. Epidemiology (Cambridge, Mass.), 28(6):827, 2017.DOI
  18. [18]Niamh Cahill, Stefan Rahmstorf, and Andrew C Parnell. Change points of global temperature. Environmental Research Letters, 10(8):084002, 2015.DOI
  19. [19]Yang Cao, Yao Xie, and Nagi Gebraeel. Multi-sensor slope change detection. Annals of Operations Research, 263(1):163–189, 2018.
  20. [20]Antonio de la Casa and Olga Nasello. Breakpoints in annual rainfall trends in Córdoba, Argentina. Atmospheric Research, 95(4):419–427, 2010.DOI
  21. [21]Alberto Del Pia, Santanu S. Dey, and Marco Molinaro. Mixed-integer quadratic programming is in NP. Mathematical Programming, 162(1–2):225–240, 2017.arxiv.org/abs/1407.4798
  22. [22]Santanu S. Dey, Guanyi Wang, and Yao Xie. Approximation algorithms for training one-node ReLU neural networks. IEEE Transactions on Signal Processing, 68, 2020.DOI
  23. [23]Farida Enikeeva and Zaid Harchaoui. High-dimensional change-point detection under sparse alternatives. The Annals of Statistics, 47(4):2051–2079, 2019.DOI
  24. [24]Paul Fearnhead and Daniel Grose. CPOP: Detecting changes in piecewise-linear signals. Journal of Statistical Software, 109:1–30, 2024.
  25. [25]Paul Fearnhead, Robert Maidstone, and Adam Letchford. Detecting changes in slope with an ℓ₀ penalty. Journal of Computational and Graphical Statistics, 28(2):265–275, 2019.DOI
  26. [26]Paul I. Feder. On asymptotic distribution theory in segmented regression problems–identified case. The Annals of Statistics, 3(1):49–83, 1975.DOI
  27. [27]Grant Foster and Stefan Rahmstorf. Global warming has accelerated significantly. Geophysical Research Letters, 53(5):e2025GL118804, 2026.
  28. [28]Robert M. Freund. Postoptimal analysis of a linear program under simultaneous changes in matrix coefficients. Mathematical Programming Study, 24:1–13, 1985.DOI
  29. [29]Vincent Froese and Christoph Hertrich. Training neural networks is NP-hard in fixed dimension. In Advances in Neural Information Processing Systems, volume 36, 2023.arxiv.org/abs/2303.17045
  30. [30]Noam Goldberg, Steffen Rebennack, Youngdae Kim, Vitaliy Krasko, and Sven Leueffer. MINLP formulations for continuous piecewise linear function fitting. Computational Optimization and Applications, 79(1):223–233, 2021.DOI
  31. [31]Leo J Guibas and Robert Sedgewick. A dichromatic framework for balanced trees. In 19th Annual Symposium on Foundations of Computer Science (sfcs 1978), pages 8–21. IEEE, 1978.DOI
  32. [32]Gurobi Optimization, LLC. Gurobi optimizer reference manual, 2026. Version 13.0.
  33. [33]Georg Hahn, Paul Fearnhead, and Idris A Eckley. Bayesproject: Fast computation of a projection direction for multivariate changepoint detection. Statistics and Computing, 30(6):1691–1705, 2020.DOI
  34. [34]James D Hamilton. A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica: Journal of the econometric society, pages 357–384, 1989.DOI
  35. [35]Hans Hersbach, Bill Bell, Paul Berrisford, Shoji Hirahara, András Horányi, Joaquín Muñoz-Sabater, Julien Nicolas, Carole Peubey, Raluca Radu, Dinand Schepers, et al. The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730):1999–2049, 2020.
  36. [36]David V. Hinkley. Inference about the intersection in two-phase regression. Biometrika, 56(3):495–504, 1969.DOI
  37. [37]Derek J. Hudson. Fitting segmented curves whose join points have to be estimated. Journal of the American Statistical Association, 61(316):1097–1129, 1966.DOI
  38. [38]Brad Jackson, Jeffrey D. Scargle, David Barnes, Sundararajan Arabhi, Alina Alt, Peter Giou-mousis, 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(2):105–108, 2005.DOI
  39. [39]Rebecca Killick, Paul Fearnhead, and Idris A. Eckley. Optimal detection of changepoints with a linear computational cost. Journal of the American Statistical Association, 107(500):1590–1598, 2012.DOI
  40. [40]Dukpa Kim. Estimating a common deterministic time trend break in large panels with cross sectional dependence. Journal of Econometrics, 164(2):310–330, 2011.DOI
  41. [41]Dukpa Kim. Common breaks in time trends for large panel data with a factor structure. The Econometrics Journal, 17(3):301–337, 2014.DOI
  42. [42]Seung-Jean Kim, Kwangmoo Koh, Stephen Boyd, and Dimitry Gorinevsky. ℓ₁ trend filtering. SIAM Review, 51(2):339–360, 2009.
  43. [43]Siu Kwan Lam, Antoine Pitrou, and Stanley Seibert. Numba: A LLVM-based Python JIT compiler. In Proceedings of the Second Workshop on the LLVM Compiler Infrastructure in HPC, pages 1–6, 2015.DOI
  44. [44]P. M. Lerman. Fitting segmented regression models by grid search. Journal of the Royal Statistical Society Series C (Applied Statistics), 29(1):77–84, 1980.DOI
  45. [45]Nimrod Megiddo. Linear programming in linear time when the dimension is fixed. Journal of the ACM, 31(1):114–127, 1984.DOI
  46. [46]Vito M. R. Muggeo. Estimating regression models with unknown break-points. Statistics in Medicine, 22(19):3055–3071, 2003.pubmed.ncbi.nlm.nih.gov/12973787
  47. [47]Apoorva Narula, Santanu S. Dey, and Yao Xie. Change-point detection via piecewise linear fitting using MIP, 2026. https://arxiv.org/abs/2602.11947.
  48. [48]Stefan Rahmstorf, Grant Foster, and Niamh Cahill. Global temperature evolution: recent trends and some pitfalls. Environmental Research Letters, 12(5):054001, 2017.DOI
  49. [49]Steffen Rebennack and Vitaliy Krasko. Piecewise linear function fitting via mixed-integer linear programming. INFORMS Journal on Computing, 32(2):507–530, 2020.DOI
  50. [50]Stephen A Ross. The arbitrage theory of capital asset pricing. Journal of Economic Theory, 13(3):341–360, 1976.
  51. [51]Michael Sagralloff and Kurt Mehlhorn. Computing real roots of real polynomials. Journal of Symbolic Computation, 73:46–86, 2016.
  52. [52]Abhinav Saxena, Kai Goebel, Don Simon, and Neil Eklund. Damage propagation modeling for aircraft engine run-to-failure simulation. In 2008 International Conference on Prognostics and Health Management, pages 1–9. IEEE, 2008.DOI
  53. [53]Alexander Schrijver. Theory of linear and integer programming. John Wiley & Sons, 1998.
  54. [54]David Siegmund and E. S. Venkatraman. Using the generalized likelihood ratio statistic for sequential detection of a change-point. The Annals of Statistics, 23(1):255–271, 1995.
  55. [55]Ryan J. Tibshirani. Adaptive piecewise polynomial estimation via trend filtering. The Annals of Statistics, 42(1):285–323, 2014.
  56. [56]Alejandro Toriello and Juan Pablo Vielma. Fitting piecewise linear continuous functions. European Journal of Operational Research, 219(1):86–95, 2012.DOI
  57. [57]Umberto Triacca and Antonello Pasini. Recent structural breaks in global temperature series: Evidence from a changepoint analysis. Climate, 14(6):129, 2026.DOI
  58. [58]Tengyao Wang and Richard J Samworth. High dimensional change point estimation via sparse projection. Journal of the Royal Statistical Society Series B: Statistical Methodology, 80(1):57–83, 2018.
  59. [59]Xin Wang, Taein Kwon, Mahdi Rad, Bowen Pan, Ishani Chakraborty, Sean Andrist, Dan Bohus, Ashley Feniello, Bugra Tekin, Felipe Vieira Frujeri, et al. HoloAssist: An egocentric human interaction dataset for interactive AI assistants in the real world. In 2023 IEEE/CVF International Conference on Computer Vision (ICCV), pages 20213–20224. IEEE, 2023.
  60. [60]Andreas Weinmann and Martin Storath. Iterative Potts and Blake–Zisserman minimization for the recovery of functions with discontinuities from indirect measurements. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 471(2176):20140638, 2015.

Paper details

vlnc:W_j8v3bkmn
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 12

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

large language models

View statement
Looking for related papers…
Contents