A new method solves large-scale sparse group square-root Lasso problems efficiently.
problem Large-scale linearly constrained sparse group square-root Lasso problems.
method Dual semismooth Newton based augmented Lagrangian method (ALM).
result The proposed method efficiently solves the problem with numerical experiments demonstrating its effectiveness.
Novel approach integrates Multivariate Square-root Lasso into Synthetic Control for high-dimensional data.
problem Challenges in practical implementation and computational efficiency of Synthetic Control method for high-dimensional disaggregated data.
method Integrates Multivariate Square-root Lasso into Synthetic Control framework.
result Demonstrates superior computational efficiency without compromising estimation accuracy.
Improved survival analysis using square root Cox's models and neural networks.
problem Feature selection in survival analysis.
method Square root Cox's survival analysis by the fittest linear and neural networks model, directly tuning penalty parameter λ.
result Substantially improved over traditional methods, achieving phase transition in feature selection.
The paper establishes a general oracle inequality for high-dimensional prediction.
problem High-dimensional data in modern sciences.
method Penalized estimators like lasso, scaled lasso, square-root lasso, elastic net.
result Generic estimators can provide consistent prediction with any design matrix.
New method handles correlated and repeated measurements using smoothed multivariate square-root Lasso.
problem Handling correlated and repeated measurements with complex noise structure.
method Proposes a concomitant estimator that uses non-averaged measurements and leverages smoothing theory for optimization.
result Demonstrates practical benefits on various datasets (toy, simulated, real neuroimaging).
Lasso is a seminal contribution to high-dimensional statistics, but it hinges on a tuning parameter that is difficult to calibrate in practice. A partial remedy for this problem is Square-Root Lasso, because it inherently calibrates to the noise variance. However, Square-Root Lasso still requires the calibration of a t…
We propose a simple imputation method for high-dimensional linear regression with missing data.
problem Handling missing covariates in high-dimensional linear regression.
method Impute missing entries with conditional mean of observed covariates and use standard LASSO or square-root LASSO.
result The imputation scheme retains minimax estimation rate and is pivotal for the square-root LASSO.
New method optimizes noise estimation alongside regression coefficients for multimodal neuroimaging data.
problem Heteroscedastic regression models with different noise levels across data sources.
method Generalized Concomitant Multi-Task Lasso for jointly estimating regression coefficients and noise covariance.
result Improved prediction and support identification with correct noise covariance estimation.
Many learning tasks, such as cross-validation, parameter search, or leave-one-out analysis, involve multiple instances of similar problems, each instance sharing a large part of learning data with the others. We introduce a robust framework for solving multiple square-root LASSO problems, based on a sketch of the learn…
The study analyzes convergence rates for sparse pivotal estimators in high-dimensional regression.
problem Sparse pivotal estimation in high-dimensional regression problems.
method Theoretical analysis and comparison of non-smooth + non-smooth optimization problems, including smoothing techniques.
result Minimax sup-norm convergence rates for square-root Lasso-type estimators are derived.
Proximal algorithms work well for SQRT-Lasso despite its nonsmooth loss.
problem Tackles the optimization of SQRT-Lasso regression.
method Applies proximal algorithms without concern for nonsmooth loss.
result Proximal algorithms converge fast with high probability.
New method for NMF without tuning parameter.
problem Finding latent structures in noisy data matrices.
method Inspired by square-root lasso, proposes a tuning-free minimum-volume NMF.
result Optimal tuning parameter value is noise level-independent.
Improved Lasso method for high-dimensional regression with noise estimation.
problem Sparse structure in high-dimensional settings with unknown noise level.
method Smoothed Concomitant Lasso, incorporating noise estimation and efficient solver.
result Increased numerical stability and computational efficiency compared to original formulation.
The lasso and related sparsity inducing algorithms have been the target of substantial theoretical and applied research. Correspondingly, many results are known about their behavior for a fixed or optimally chosen tuning parameter specified up to unknown constants. In practice, however, this oracle tuning parameter is …
The paper improves support recovery guarantees for the group Lasso.
problem Estimating group-sparse signals from noisy measurements.
method Establishes new conditions for accurate group-level support estimation using the group Lasso.
result Allows for nearly as many recoverable nonzero groups as the total number of groups.
New method learns optimal cost for machine learning models.
problem Optimizing machine learning models under distributional uncertainty.
method Data-driven approach to define distributional uncertainty neighborhoods.
result Improves upon various machine learning estimators.
New algorithm finds global minimum for TREX, controlling FDR.
problem Sparse high-dimensional regression with TREX's non-convex optimization.
method Polynomial-time algorithm for non-convex TREX optimization.
result Global minimum found for TREX, controlling FDR.
Study differential properties of matrix square roots in specific cases.
problem Understanding matrix square roots in semi-simple, symmetric, and orthogonal cases.
method Analysis of differential and metric structures of real square roots of matrices under specific conditions.
result Differential properties of matrix square roots in semi-simple, symmetric, and orthogonal cases.
A new method simulates square-root processes efficiently.
problem Simulating square-root processes accurately and efficiently.
method Simulate the integrated square-root process instead of the square-root process itself.
result High precision with low number of time steps, and exact limiting Inverse Gaussian distributions.
Square-root impact law confirmed for option trades.
problem Is market impact similar for stocks and options?
method Analyzed proprietary data of option trades.
result Square-root law holds for option markets.
Guarantees uniform convergence for square-root Lipschitz losses.
problem Uniform convergence guarantees for square-root Lipschitz losses.
method Using Rademacher complexity and square root of scalar loss function Lipschitz constant.
result Generalizes previous results and handles non-smooth loss functions.
The paper shows square roots of certain transformations can't exist on contact manifolds.
problem Existence of square roots of diffeomorphisms on contact manifolds.
method Proving the non-existence of square roots for arbitrary small contactomorphisms.
result There exist arbitrary small contactomorphisms without square roots.
Study finds price impact follows a 'double' square-root law, suggesting mechanical origin.
problem Understanding the origin of price impact in markets.
method Detailed dataset of Tokyo Stock Exchange orders, analyzing single and metaorders.
result Price impact follows a 'double' square-root law, indicating mechanical origin rather than information.
Corrects the misconception that market impact is just volatility.
problem Misinterpretation of market impact as volatility.
method Introducing a simple scaling argument and comparing it to empirical data.
result Market impact is not related to price diffusion.
We analyze how uncertainty in models affects optimization outcomes using Wasserstein distances.
problem Sensitivity of optimization problems to model uncertainty.
method Non-parametric approach using Wasserstein balls to capture uncertainty, providing explicit corrections for value function and optimizer.
result Explicit formulae for first-order corrections to value function and optimizer.
Adversarial training improves linear regression solutions, offering robustness against small perturbations.
problem Vulnerability of linear models to adversarial perturbations.
method Formulated as a min-max problem, adversarial training minimizes the best solution under worst-case attacks.
result Adversarial training yields the minimum-norm interpolating solution in overparameterized models, equivalent to parameter shrinking methods in underparameterized models.
The paper derives oracle inequalities for estimators with fast and slow rates.
problem Developing fast and slow oracle inequalities for estimators.
method Direct study of analysis estimator and adaptation of Dalalyan, Hebiri and Lederer's arguments.
result Constant-friendly rates for (square root) total variation regularized estimators over graphs.
Square-root natural-gradient improves variational inference convergence.
problem Challenges in establishing theoretical convergence guarantees for natural-gradient descent.
method Square-root parameterization for Gaussian covariance.
result Establishes novel convergence guarantees for natural-gradient Gaussian inference.
The Volterra square-root process shows non-uniqueness of limiting distributions and regularity of its law.
problem Non-uniqueness of limiting distributions in the Volterra square-root process.
method Establishing existence of limiting distributions using integrability of the Volterra convolution kernel and exponential-affine transformation.
result The limiting distributions of the Volterra square-root process depend on the initial state and belong to weighted Besov spaces.
The study confirms that market volatility can be explained by correlated metaorders impacting prices in a square-root fashion.
problem Explaining market volatility using metaorders and their impact.
method Generated synthetic market data and analyzed the correlation between order flow and returns.
result The square-root law of market impact is confirmed and can be measured from anonymized trade data.
Extends square root velocity transform to handle curves in shape spaces of manifold-valued data.
problem Handling curves that leave the shape space in shape analysis.
method Generalizes absolutely continuous curves to strong Riemannian manifolds and extends SRVT to handle these curves.
result Computations in the SRVT framework can now be applied to curves in shape spaces of manifold-valued data.
We apply an asymmetric version of Kirman's herding model to volatile financial markets. In the relation between returns and agent concentration we use the square root law proposed by Zhang. This can be derived by extending the idea of a critical mean field theory suggested by Plerou et al. We show that this model is eq…
Solves non-Abelian Rainich problem for SU(2) gauge fields.
problem Existence of local SU(2) Yang-Mills fields with prescribed stress-energy tensor.
method Canonically identifying tensors with Hermitian forms and defining internal square roots of stress-energy tensors.
result Existence of local SU(2) Yang-Mills field is equivalent to a single differential condition on internal square roots of stress-energy tensor.
Revisiting Trade-sign Long-memory and Square-root Law price impact
problem Revisiting the Lillo-Mike-Farmer (LMF) theory and the square-root law (SQRL) of meta-order impact
method Using a coupled discrete reaction-diffusion formulation
result Long-memory of trade signs and square-root law of meta-order impact
Study finds a crossover from linear to square-root market impact based on order volume.
problem Understanding the dynamics of market impact as a function of order volume.
method Used a large dataset of 8 million trades to establish the crossover between linear and square-root market impact regimes. Applied a dynamical theory of liquidity to explain the results.
result Quantitative agreement with data achieved by considering two liquidity time scales: fast and slow.
We confirm the square-root law of market impact on Apple Inc. using a large dataset.
problem Testing the square-root law of market impact on a single U.S. large-cap equity.
method Using a full market-by-order feed, we reconstruct metaorders and calibrate impact using the square-root formula.
result The square-root law is confirmed with a prefactor of 0.34, consistent with worldwide data.
Study two-generated subgroups of homeomorphisms with specific chain supports and find uncountably many isomorphism types.
problem Characterize two-generated subgroups of homeomorphisms with specific chain supports and their square roots.
method Analyze subgroups of Homeo+(I) with supports forming a chain of two intervals, and examine their square roots. result Uncover uncountably many isomorphism types of subgroups with specific chain supports and their square roots.
Agent-based market shows herding cycles with square-root price impact.
problem Understanding herding cycles in agent-based markets.
method Agent-based model with 20,000 retail traders interacting with a single institutional agent.
result Agent discovers multi-cycle predatory strategy with 8-11 complete cycles over 2000 trading days.
New auto-encoder handles varying noise levels without retraining.
problem Auto-encoders degrade in noisy conditions.
method Formalized auto-encoders as transform learning, derived new architecture.
result Models generalize well to different noise levels.
This paper studies confidence regions for robust estimators using Wasserstein distance.
problem Developing robust estimators against model misspecification.
method Wasserstein distributionally robust optimization.
result Asymptotic normality and optimal confidence regions for distributionally robust estimators.
New bounds for Lasso and Group Lasso in high dimensions derived.
problem Estimation error bounds for Lasso and Group Lasso in high-dimensional settings.
method Recent advances in high-dimensional statistics to derive new L2 estimation upper bounds.
result Bounds match optimal minimax rate for Lasso and improve over existing results for Group Lasso.
Develops a model to explain price dynamics under predictable market flow.
problem Diffusive price dynamics paradox under predictable market-order flow.
method Integrates square-root price-impact law into Lillo--Mike--Farmer model.
result Price dynamics are diffusive at long times under predictable market-order flow.
PLS-Lasso integrates dimension reduction into regression for financial index tracking.
problem Dimension reduction and regression are traditionally treated separately in multivariate data analysis.
method PLS-Lasso integrates dimension reduction directly into the regression process, presenting two formulations: PLS-Lasso-v1 and PLS-Lasso-v2.
result PLS-Lasso-v1 and PLS-Lasso-v2 outperform Lasso in financial index tracking.
The paper examines Adaptive Lasso and Transfer Lasso, highlighting their differences and proposing a new method.
problem Comparing and contrasting Adaptive Lasso and Transfer Lasso.
method Theoretical analysis of asymptotic properties and introduction of a new method.
result The Transfer Lasso method reduces non-asymptotic estimation errors compared to Adaptive Lasso.
The square root velocity framework is a method in shape analysis to define a distance between curves and functional data. Identifying two curves if they differ by a reparametrisation leads to the quotient space of unparametrised curves. In this paper we study analytical and topological aspects of this construction for …
Efficiently computes matrix square roots and their inverses for large matrices.
problem Computing matrix square roots and inverses for large matrices efficiently.
method Combines Krylov subspace methods with rational approximation for quadratic-time computation.
result Achieves 4 decimal places of accuracy with fewer than 100 matrix-vector multiplications.
Paper connects surface shape analysis and unbalanced optimal transport.
problem Computing the SRNF shape distance on piecewise linear surfaces.
method Characterizes SRNF shape distance as WFR distance pullback, proposes new algorithm for WFR distance computation.
result Direct computation of SRNF shape distance on piecewise linear surfaces.
Compact bilinear pooling approximates covariance features for faster training.
problem Efficiently approximating covariance features for faster training.
method Compact bilinear pooling extended to polynomial approximations of covariance features.
result The proposed method achieves comparable accuracy with fewer dimensions.