Second Maximum of a Gaussian Random Field and Exact (t-)Spacing test

arXiv:2406.18397 · math.ST, cs.LG, math.DG, math.PR, stat.ML, stat.TH · Submitted 2024-06-26 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: AI Radio. Generated commentary on the latest Artificial Intelligence papers.

Tom: I'm Tom.

Jane: And I'm Jane, here with Lu, Meng and our in-house large language model, Lalam.

Tom: Today's paper: "Second Maximum of a Gaussian Random Field and Exact (t-)Spacing test".

Jane: The gist In this article,

Tom: First, who's behind it and why it matters.

Title and authors: Tom: So we're looking at this paper by Azaïs, Dalmao, and De Castro called "Second Maximum of a Gaussian Random Field and Exact (t-)Spacing test". It’s about finding a new way to characterize the distribution of the global maximum when you're dealing with a Gaussian random field on a Riemannian manifold.

Jane: That sounds really technical. Could you break down what that means for someone who isn't working with manifolds every day? What's the big idea here?

Tom: Basically, they introduce this new concept of the second maximum, which they call lambda two <ref:2406.18397#pg1,concept of the second maximum>. This statistic is supposed to be a really powerful tool for characterizing how the global maximum behaves.

Lu: I think what's interesting is how they use it. They condition the distribution of the global maximum on this second maximum and some part of the Riemannian Hessian, which they call <ref:2406.18397#pg2>.

Meng: Conditioning on things sounds complicated when you're trying to ship models. What is this actually going to help us do in practice?

Lalam: The core result they derive is an exact test based on the spacing between these maxima, which they call the spacing test <ref:2406.18397#pg1>. This lets them get a precise distribution for the maximum when conditioned on lambda two and <ref:2406.18397#pg1>.

Jane: An exact test sounds promising. So, what is this spacing test actually doing? Is it something we can use right now?

Tom: It’s not just some theoretical thing. They show that when you know the variance, they get this statistic S sigma(lambda one lambda two):= G/sigma(lambda one/sigma) G/sigma(lambda two/sigma) and it is uniformly distributed on the interval (zero one) under the null hypothesis <ref:2406.18397#pg2>.

Jane: Uniformly distributed on (zero one) <ref:2406.18397#pg2>. So if we run this test, a value falling in a certain range tells us something concrete about whether we have a signal or not. What does that mean for sparse alternatives?

Lu: They investigate this specifically for detecting sparse alternatives in Gaussian symmetric tensors <ref:2406.18397#pg1>. They look at cases like m(times) = lambda zero c(times, t zero) <ref:2406.18397#pg3>.

Meng: I've seen papers on spiked tensor PCA before, and those often rely on asymptotic limits. Does this paper offer something different than what we already have?

Tom: Yeah, they explicitly distinguish their exact results from the asymptotic phase transitions seen in Spiked Tensor PCA <ref:2406.18397#pg3>. This means it's an exact inference procedure valid for any fixed dimension d and sample size <ref:2406.18397#pg5>.

Title and authors: Jane: That’s a big deal if it holds up across different dimensions without needing huge data sets or relying on the system being very large. What about when we don't know the noise level?

Tom: They address that with a studentized version, which they call the t-spacing test <ref:2406.18397#pg3>. This test uses an estimation sigma b built from a Karhunen-Loève expansion <ref:2406.18397#pg5>.

Lu: The t-spacing test is derived using the density of the Student t-distribution <ref:2406.18397#pg5>. Under the null hypothesis, this test statistic T(T one T two /sigma b):= H/sigma b(T one) H/sigma b(T two) is also uniformly distributed on (zero one) <ref:2406.18397#pg5>.

Meng: If it doesn't require knowing the variance in advance, that makes deployment much more robust. What are the practical limitations we need to know?

Tom: Assumption (A5) for this unknown variance case requires at least d + two degrees of freedom <ref:2406.18397#pg1>. They also note that the estimator sigma b two t follows a chi distribution with kappa degrees of freedom under the null hypothesis <ref:2406.18397#pg5>.

Jane: So, to summarize for someone listening who isn't reading this paper, they’ve created a method that doesn't need you to know the exact noise level upfront, and it works exactly even when you are testing sparse models in continuous settings.

Lu: The framework is applied across several areas: Gaussian symmetric tensors <ref:2406.18397#pg1>, continuous sparse deconvolution, and two-layered neural networks with smooth rectifiers <ref:2406.18397#pg1>.

Meng: I'm interested in the geometry part. They talk about the helix property where Xt one(times) exhibits a singularity at t one <ref:2406.18397#pg5>. Does that geometric structure mean anything for how we build our neural network layers?

Tom: It helps clarify the local geometry around the global maximum by linking it to the eigen-decomposition of and directional limiting values near t one <ref:2406.18397#pg5>. This geometric insight informs how local curvature relates to directional sparse solutions.

Jane: It sounds like this paper moves us from just looking at global maximums to understanding the local structure around them, which is helpful for continuous kernel regression.

Lu: Exactly. The maxima lambda one and lambda two correspond directly to the knots of the Continuous LARS algorithm <ref:2406.18397#pg3>. This means the test statistic is intrinsically tied to how that algorithm works in continuous sparse regression problems.

Meng: For deployment, if this test is calibrated under the null hypothesis, it means we can set a threshold and have confidence in that setting for detecting sparse alternatives <ref:2406.18397#pg3>. What about the applications like Super-Resolution?

Title and authors: Tom: They show explicit descriptions of the test statistics for applications like Super-Resolution, where they describe how AI models can perform exact reconstruction using beurling minimal extrapolation <ref:2406.18397#pg5>.

Jane: So, if you're working on image restoration or any kind of sparse reconstruction, this gives you a precise mathematical tool to check if the results are just noise or if there's a real signal.

Lu: The paper also details how they compute these test statistics using explicit expressions in specific cases <ref:2406.18397#pg5>. This is important for implementers because it shows the computation isn't just some black box integral.

Meng: I need to make sure that when I deploy this, the computational cost of calculating these G/sigma terms doesn't blow up on my production hardware. What about scalability?

Tom: The explicit expressions allow for direct computation in specific cases <ref:2406.18397#pg5>. And they mention the framework's applicability to two-spiked tensor models and their generalization <ref:2406.18397#pg5>.

Jane: So we have this exact, non-asymptotic framework for Gaussian random fields on manifolds, a test that works even when the noise variance is unknown, and it applies to things like neural networks and tensor models. It seems like it gives us much tighter control than what we usually have.

Lu: The main contribution is indeed the geometric characterization of lambda two and the derivation of that singular Kac-Rice formula <ref:2406.18397#pg1>. It provides a concrete, non-asymptotic inference procedure for testing sparse alternatives in Gaussian symmetric tensors <ref:2406.18397#pg5>.

Meng: I just hope the real-world performance holds up when we move from the theoretical setting to actual noisy traffic where assumptions might not be perfect.

Tom: That’s the challenge for any method, right? But this paper gives us a very strong foundation to test those assumptions rigorously before we deploy anything <ref:2406.18397#pg5>.

Jane: So, the next step is seeing how these precise spacing tests link up with the actual thresholding methods we use in AI systems. That’s where things get really interesting for practical application.

Lu: I think that's where the real impact will be felt, connecting this geometric understanding to concrete maximum likelihood detection thresholds <ref:2406.18397#pg5>.

Tom: It’s a solid piece of theory that gives us exact tools for inference in these complex spaces, regardless of the underlying noise structure.

Jane: Indeed, it's a lot of work on characterizing these random fields and deriving these precise statistics. We’ll keep an eye on how this connects to the next set of papers we review.

The paper's summary: Tom: So, we're talking about this paper now, "Second Maximum of a Gaussian Random Field and Exact (t-)Spacing test". The main point is they found this way to define the second maximum, lambda two for a Gaussian random field on a curved space.

Jane: So how does that help us? It seems like they're trying to get more precise information about the biggest peaks in our data than just looking at the absolute highest one.

Lu: They condition everything on this second maximum and some part of the curvature, which they call, and that leads to a new way of looking at the distribution of the field's maximum.

Meng: What does "conditioning" mean for me? Does it just make things harder to calculate? Is it adding complexity to my deployment pipeline?

Tom: Not exactly. They use this conditioning to derive a tailored Kac-Rice formula, which gives us an explicit distribution for the maximum when we know these specific statistics, lambda one and lambda two.

Jane: An explicit distribution sounds like a big deal. So they can actually calculate the probability of seeing certain peak spacings.

Lu: They use this to build something called the "spacing test," which is a way to check if what we're seeing is just noise or if we have a genuine signal, based on how far apart these maxima are.

Meng: A test based on spacing? That sounds like it could be useful for checking sparse alternatives in AI models, right?

Tom: Exactly. They show this test works really well when you're testing sparse models in Gaussian symmetric tensors, and they do it exactly—no approximations needed for dimension d.

Jane: So if we can get an exact test that doesn't rely on asymptotic limits, that’s much more reliable than the old methods we use for things like Spiked Tensor PCA.

Lu: They specifically point out how their results differ from those asymptotic phase transitions, which means this inference procedure is valid no matter the dimension or how large your sample size is.

Meng: I need something that works on real traffic, not just in a textbook scenario. So what's the catch for using this method? What's the trade-off?

Tom: The trade-off is that for the unknown variance case, they need at least d + two degrees of freedom to build their test statistic sigma b, and they have to use an estimator built from a Karhunen-Loève expansion.

Jane: So, if you're testing something without knowing your noise level beforehand, you still need some initial data structure to estimate that noise parameter.

Lu: That’s the caveat. But once you have that sigma b, the t-spacing test statistic T(T one T two /sigma b) is exactly uniformly distributed on zero to one under the null hypothesis.

Meng: A uniform distribution on zero to one means we can set a threshold and be certain about our detection probability, regardless of what the true noise level is. That’s what I need for production deployment.

Tom: Right. And they also showed that this framework extends to continuous sparse deconvolution and two-layered neural networks with smooth rectifiers, which are architectures we see all the time in deep learning.

Jane: It sounds like this isn't just a niche math problem; it’s a general tool for analyzing how signals behave in these complex AI structures.

Lu: The geometric part is interesting too. they link the eigen-decomposition of to directional limiting values, which gives us some insight into the local shape around that global maximum peak.

Meng: So you're telling me that by understanding the local curvature encoded in, you can better understand how your sparse AI model is actually finding its solutions?

Tom: Precisely. This geometric interpretation helps clarify what kind of structure we are looking at when we analyze those maxima.

Jane: So, the big picture here is moving from just seeing global maximums to understanding the local geometry around them, which informs how we build our sparse AI models and how they solve regression problems.

Lu: And by linking these maxima to knots in algorithms like Continuous LARS, it connects the statistical test directly to how those continuous sparse solutions are found.

Meng: I'm still thinking about the computational cost on my hardware. Are there explicit expressions that make this calculation feasible for large models?

Tom: Yes, they provide explicit expressions in specific cases, which means you can compute these statistics without relying solely on numerical integration, depending on the specific manifold you are working with.

Jane: So it's a robust framework for inference in continuous sparse regression problems across various AI architectures. Next time we talk about this, we need to figure out how this exact test actually translates into setting detection thresholds for real-world applications.

The paper's improvements: Tom: So, moving on to what they suggest improving about this test, they're not just giving us a static formula; they’re proposing how to make it even more precise for real-world use.

Jane: What kind of improvements are we talking about? Are they fixing some weird edge cases or making the math cleaner?

Lu: They are refining the way the test statistic is constructed, specifically by incorporating information from the singular Kac-Rice formula in a more controlled setting.

Meng: Controlled setting sounds technical. Does that mean it handles things that are usually messy in our production environment better?

Tom: It means they've managed to derive an explicit conditional density of the maximum, which is super useful for problems where the maximum peak itself has some kind of singularity, like in continuous sparse regression.

Jane: So if a peak has a singularity, that’s when standard tests usually break down, and this new formula helps us handle that?

Lu: Exactly. This geometric insight links the eigen-decomposition of the Hessian to directional limiting values near that first maximum peak.

Meng: Linking local curvature to directional sparse solutions sounds like it could help us design better layers in our AI models if we want them to find those sparse solutions more efficiently.

Tom: Yeah, it gives us a way to understand how the local geometry around the global maximum relates directly to the direction of our sparse solution.

Jane: That's a big step because it moves beyond just counting peaks and starts understanding the shape of what we are looking for.

Lu: And they also discuss generalizing this framework, showing it can be applied to even more complex structures, like two-spiked tensor models with their own specific generalization techniques.

Meng: Generalization is good, but what about scalability? If I have a massive model, will the computational cost of calculating these refined statistics still be manageable for me?

Tom: The authors show that they can compute the test statistics using explicit expressions in specific cases, which means you can calculate them without relying solely on heavy numerical integration.

Jane: So you get this refined statistical power without having to run a massive, slow simulation every time you want to check your data.

Lu: And they confirm that this works for various applications like super-resolution and continuous sparse deconvolution, which means the framework is broad.

Meng: I'm still focused on deployment costs. If I use this refined test in production, what’s the actual overhead compared to a simpler method?

Tom: The overhead depends on whether you are using the known variance or estimating it with sigma b; but for two-layered neural networks, they show it remains computationally efficient enough for practical use.

Jane: So they're showing that this high-level mathematical insight translates into something that can actually be deployed in a way that makes sense economically.

Lu: The implication is that we now have a more rigorous, non-asymptotic tool to evaluate the quality of sparse solutions in complex continuous spaces.

Meng: That’s what I want to hear: a tool that gives me better confidence in my model's sparsity structure without requiring me to guess the noise level first.

Conclusion: Tom: So, to wrap up this talk on "Second Maximum of a Gaussian Random Field and Exact (t-)Spacing test," the main thing is that they’ve given us an exact, non-asymptotic way to test for signal in continuous sparse settings.

Jane: It seems like they've built a really robust statistical tool that doesn't depend on relying on big asymptotic approximations when we analyze Gaussian random fields on manifolds.

Lu: They establish the geometric characterization of lambda two, which is key because it links the local curvature of the field directly to how sparse solutions behave.

Meng: For deployment, this means we have a test that works for any fixed dimension d and sample size, which is something I need when I’m trying to deploy models on unpredictable real-world data.

Tom: Exactly. It gives us a concrete inference procedure for testing those sparse alternatives in Gaussian symmetric tensors, which is much better than the rough estimates we usually get.

Jane: And the t-spacing test, especially the version that handles unknown noise variance via sigma b, sounds like it’s going to be very practical for my audience who don't read math.

Lu: It provides exact inference without needing prior knowledge of the noise level, provided you have at least d plus two degrees of freedom to build that estimator.

Meng: I still worry about the estimation part; if that sigma b is off by a lot, does my whole test get garbage results? What’s the actual cost of calculating it?

Tom: The explicit expressions they provide for the test statistics in specific cases are what make it feasible for computation on production hardware.

Jane: So, this paper provides a solid foundation for how AI systems can reliably distinguish between signal and noise in continuous sparse deconvolution and two-layered neural networks.

Lu: They showed that even applications like super-resolution benefit from this framework because of the precise statistics they derive for those scenarios.

Meng: It’s good to see it applied to things like tensor models, which is where I usually look when I’m trying to understand how complex representations are learned in deep architectures.

Tom: The work on Gaussian symmetric tensors and continuous sparse deconvolution shows the versatility of this approach beyond just one type of model.

Jane: So, if you’re interested in applying this to image restoration or neural network analysis, there’s a solid mathematical basis now for doing that inference exactly.

Lu: The paper concludes by detailing deployment procedures for specific manifolds like the sphere S2 and the Stiefel manifold Mn, two of which are very relevant for geometry-heavy applications.

Meng: I’m glad they showed those specific cases; it makes the whole thing feel less abstract and more like something I could actually integrate into my pipeline.

Tom: It does provide that connection to reality. The title, "Second Maximum of a Gaussian Random Field and Exact (t-)Spacing test," really sums up the core contribution here.

Jane: We're leaving this paper with a better tool for rigorous inference in complex AI systems, focusing on the geometry around peak structures and handling unknown noise levels.

Lu: The next logical step is figuring out how these exact spacing tests link up with the actual thresholding methods we use in model evaluation, which is where I think the real impact will be felt.

Meng: That’s what we need to focus on—bridging this theoretical exactness with the operational reality of setting those final detection thresholds.

Institut de Mathématiques de Toulouse · DMEL, CENUR Litoral Norte · Institut Camille Jordan, CNRS UMR 5208 · Institut Universitaire de France (IUF)

math.ST, cs.LG, math.DG, math.PR, stat.ML, stat.TH

Submitted: 2024-06-26

Updated: 2026-10-15

Comments: v4: connection with the Kac-Rice test of Taylor, Loftus and Tibshirani (2016) made explicit; new section on close maxima. Code: https://github.com/ydecastro/tensor-spacing

Code: https://github.com/ydecastro/tensor-spacing

License: http://creativecommons.org/licenses/by-sa/4.0/

Importance score: 86/100

The gist: The gist In this article, we introduce a novel concept of the second maximum of a Gaussian random field defined on a Riemannian manifold and derive an exact test based on its spacing between maxima.

Key concepts

Second Maximum: This is the largest value of a Gaussian random field found away from its global maximum point ($t_1$). It is calculated after removing the value and gradient information at $t_1$. This statistic helps characterize the distribution of where the overall maximum occurs.
Spacing Test: This test uses the distance between the first maximum ($\lambda_1$) and the second maximum ($\lambda_2$). By analyzing this spacing, researchers can create an exact statistical test to determine if a signal is sparse or if a specific alternative exists in models like neural networks.
Riemannian Hessian (Independent Part): This refers to the part of the Riemannian Hessian matrix at the maximum point ($t_1$) that is independent of the field's value and gradient. This component is crucial for defining the statistical properties of the second maximum and is used in constructing a tailored Kac-Rice formula.
t-spacing Test: This is a specific version of the spacing test designed when the variance of the random field is unknown. It uses an estimated variance ($\sigma_b$) derived from a Karhunen-Loève expansion to create a test statistic that remains perfectly calibrated under the null hypothesis, regardless of whether the true noise level is known.

Terminology

Calibration — 1× in this paper · explained in 7 episodes

Summary

The gist In this article, we introduce a novel concept of the second maximum of a Gaussian random field defined on a Riemannian manifold and derive an exact test based on its spacing between maxima.

Theoretical Framework

The core idea involves characterizing the distribution of the global maximum by conditioning it on the second maximum and the independent part of the Riemannian Hessian pg2>. This leads to a tailored Kac-Rice formula that yields an explicit distribution for the maximum conditioned on these statistics pg3>. The second maximum is defined as λ2 ∈ arg max s≠t1∈M of σXt1(s) pg3>. This approach allows for the derivation of an exact test based on the spacing between these maxima, referred to as the spacing test pg3>.

Key Results and Tests

The paper establishes several key results regarding the exactness and power of this test pg4>.

Exact Spacing Test (Known Variance): When the variance σ is known, Theorem 2 states that the test statistic Sσ(λ1, λ2, omega):= Gomega/σ(λ1/σ) Gomega/σ(λ2/σ) ∼ U(0, 1), where Gomega/σ(l) is defined by an integral pg5>. This demonstrates that one minus the ratio (1.3) represents the cumulative distribution function (CDF) of the law of λ1 conditional on λ2 and omega, under the null hypothesis pg5>.
Studentized Spacing Test (Unknown Variance): When variance is unknown, a Studentized version called the t-spacing test is derived using an estimation σb built from the Karhunen-Loève expansion pg5>. Theorem 3 shows that the test statistic T(T1, T2, omega/σb):= Homega/σb(T1) Homega/σb(T2) ∼ U(0, 1), where Homega/σb is defined using the density of the Student t-distribution pg5>.

Applications and Extensions

The framework is shown to be applicable across various fields, including Gaussian symmetric tensors, continuous sparse deconvolution, and two-layered neural networks with smooth rectifiers pg1>.

- Detecting Sparse Alternatives:

The test is expected to have high power on s-sparse alternatives of the form m(·) = λ0c(·, t0) pg5>. In the moderate regime, the spacing λ1 − λ2 grows linearly with γ, and (t1, λ1) weakly converges to the distribution of (t0, σX(t0)), allowing detection pg3>.

- Comparison with Asymptotic Limits:

The paper distinguishes its exact results from asymptotic phase transitions in Spiked Tensor PCA, providing an exact, non-asymptotic inference procedure valid for any fixed dimension d and sample size pg5>.

General Methodology

The methodology involves defining the profile likelihood random field Z(t) = m(t) + σX(t) pg1>. The statistics are computed using Riemannian gradient descent algorithms on M pg5>. The independent part of the Hessian omega is defined as omega:= σR(t1) = σΛ− 1/22(t1) Re(t1) Λ− 1/22(t1) pg5>. The test statistics are then computed using the integral forms Gomega/σ (resp. Homega/σb), which are explicitly given in terms of the determinant of matrices involving omega pg5>.

Manifold and Tensor Examples

The framework is illustrated with several examples, such as Spiked Tensor PCA on the sphere S2 pg5>, Two-Spiked Tensor Model on the Stiefel manifold Mn,2 pg5>, and Super-Resolution Torus T pg5>. The analysis of the profile likelihood random field Z(t) reveals that Xt1(·) exhibits a singularity at t1, forming a structure referred to as the helix pg5>.

Variance Estimation Details

For the unknown variance case, Assumption (A5) requires at least d + 2 degrees of freedom pg1>. The estimator σb 2 t is constructed using a Karhunen-Loève expansion pg5>. Under the null hypothesis, σb 2 t follows a χ-distribution with κ degrees of freedom pg5>. The final t-spacing test statistic T(T1, T2, omega/σb) is shown to be uniformly distributed on (0, 1) under H0 pg5>. This provides an exact inference procedure without prior knowledge of the noise level pg5>.

Conclusion and Future Work

The work provides a framework for applying the spacing test to continuous sparse regression for any Gaussian random field on a Riemannian manifold pg1>. The main contributions include the geometric characterisation of λ2, the singular Kac–Rice formula, and the exact spacing test pg5>. Future work may explore precise links between this method and maximum likelihood detection thresholds pg5>. The paper concludes by detailing deployment procedures for specific manifolds like S2 and the Stiefel manifold Mn,2 pg5>. The appendix provides proofs for key lemmas, such as the helix random field property pg5>. The final section details how to compute the test statistics using numerical integration or explicit expressions in specific cases pg5>. The paper also discusses two-spiked tensor models and their generalization pg5>.

References

Adler, R. J. and Taylor, J. E. (2009). Random fields and geometry pg27

Armentano, D., Azaïs, J.-M., and León, J. R. (2023). On a general Kac-Rice formula for the measure of a level set pg1

Auffinger, A. and Ben Arous, G. (2013). Complexity of random smooth functions on the high-dimensional sphere pg1

Azaïs, J.-M., De Castro, Y., and Gamboa, F. (2015). Spike detection from inaccurate samplings pg1

Azaïs, J.-M., De Castro, Y., and Mourareau, S. (2017). A rice method proof of the null-space property over the grassmannian pg5

Azaïs, J.-M. and Wschebor, M. (2005). On the distribution of the maximum of a gaussian field with d parameters pg27

Candès, E. J. and Fernandez-Granda, C. (2014). Towards a mathematical theory of super-resolution pg5

Chizat, L. (2022). Sparse optimization on measures with over-parameterized gradient descent pg5

De Castro, Y., Gadat, S., Marteau, C., and Maugis-Rabusseau, C. (2021). Supermix: sparse regularization for mixtures pg5

De Castro, Y.

Improvements for AI systems

Azaïs et al.'s framework enables an exact test based on the spacing between these maxima, allowing for exact, non-asymptotic inference procedure valid for any fixed dimension d and sample size when testing sparse alternatives in Gaussian symmetric tensors. This allows AI systems to perform hypothesis testing on high-dimensional data without relying on asymptotic approximations like those used in classical Spiked Tensor PCA literature.
The t-spacing test provides exact, non-asymptotic inference for the spiked tensor model without requiring prior knowledge of the noise level, which is crucial for AI applications where noise variance is often unknown. This allows models to be evaluated against sparse alternatives robustly even when the noise structure is not perfectly characterized.
The t-spacing test can be studentized, allowing exact inference using the studentised spacing λ1/σb − λ2/σb without prior knowledge of σ 2. This capability lets AI systems make precise inferences about signal strength in complex regression problems even when the noise variance is only known up to a scaling factor.
The test statistics, such as Sσ(λ1, λ2, omega):= Gomega/σ(λ1/σ) Gomega/σ(λ2/σ) ∼ U(0, 1), provide a calibrated measure for detecting sparse alternatives in continuous sparse deconvolution and two-layered neural networks with smooth rectifiers. This allows AI to reliably distinguish between signal and noise in these specific architectures.
The framework extends the Kac-Rice formula to a controlled singular setting, enabling the derivation of an explicit conditional density of the maximum, which is used for testing sparse regression problems where maxima are intrinsically linked to knots of the Continuous LARS algorithm. This can be applied in AI systems performing continuous sparse kernel regression.
The analysis links the eigen-decomposition of omega and directional limiting values of the rescaled field near t1, providing a helix interpretation that clarifies local geometry around the global maximum. This geometric insight can inform AI system design by understanding how local curvature encoded in the Hessian relates to directional approaches to sparse solutions.
The framework supports applications like Super-Resolution, where the test statistics are explicitly described, enabling AI models to perform exact reconstruction using beurling minimal extrapolation for recovering fine-scale details from noisy measurements.

Abstract

In this article, we introduce the novel concept of the second maximum of a Gaussian random field defined on a Riemannian manifold: Once the field reaches its global maximum at a point t 1, the second maximum is the largest value of the field away from t 1, after the value and the gradient of the field at t 1 have been removed. This statistic serves as a powerful tool for characterizing the distribution of the global maximum. By employing an ad-hoc Kac-Rice formula, we derive the explicit distribution of the maximum, conditioned on the second maximum and the independent part of the Riemannian Hessian (the part of the Hessian at t 1 that is independent of the value and the gradient of the field at t 1). This approach yields an exact test based on the spacing between these maxima, which we refer to as the spacing test. We investigate the applicability of this test for detecting sparse alternatives in Gaussian symmetric tensors, continuous sparse deconvolution, and two-layered neural networks with smooth rectifiers. Theoretical results are supported by numerical experiments that illustrate the calibration and power of the proposed procedures. More generally, we provide a framework for applying the spacing test to continuous sparse regression for any Gaussian random field on a Riemannian manifold. Furthermore, when the variance-covariance function is known only up to a scaling factor, we derive an exact Studentized version of the procedure, coined the t-spacing test. We show that this test is perfectly calibrated under the null hypothesis and exhibits high power for detecting sparse alternatives. As a corollary, we derive an exact, non-asymptotic test for the spiked tensor model without requiring prior knowledge of the noise level.