Online learning of linear operators between infinite-dimensional spaces is possible but with limitations.
problem Learning linear operators between infinite-dimensional Hilbert spaces in an online setting.
method Online learning approach for linear operators with bounded p-Schatten norm, proving impossibility for operator norm. result Separation between online learnability and uniform convergence for bounded linear operators.
Uniform linear bounds on volume changes in 3D hyperbolic spaces.
problem Volume variation in hyperbolic 3-manifolds.
method Uniform linear bounds proof for drilling and filling operations.
result Uniform linear bounds on volume variation proved.
Solves low-rank approximation problems in Hilbert spaces.
problem Low-rank approximation in Hilbert spaces.
method Closed-form solutions and error bounds for bounded linear operators.
result Generalization to bounded linear operators from finite dimensions.
New method bounds high-dimensional regression without estimating design covariance.
problem High-dimensional linear regression with random design.
method Error-in-operator approach that incorporates design covariance into empirical risk minimization.
result Dimension-free bounds on excess prediction risk derived.
Unified bounds for neural networks incorporating physical laws.
problem Limitations in existing generalization analyses for PINNs and VPINNs.
method Unified framework using Taylor expansion and Koopman-based analysis.
result High-rank networks can generalize well even with differential operators.
Abstract: A new approach to technical indicators without lag.
problem Defining classical technical indicators as bounded operators for lag-free trading.
method Using linear algebra to redefine technical indicators as bounded operators in l∞(N) space. result Demonstrated the no-lag versions of technical indicators are simpler and more effective.
In an L∞-framework, we present a few extension theorems for linear operators. We focus the attention on majorant preserving and sandwich preserving types of extensions. These results are then applied to the study of price systems derived by a reasonable restriction of the class of equivalent martingale measures…
New analysis proves sketching operators' RIP guarantees for mixture models without importance sampling.
problem Proving sketching operators' Restricted Isometry Property (RIP) for mixture models without assuming importance sampling.
method Proposed alternative analysis based on new deterministic bounds and concentration inequalities.
result Theoretical guarantees for sketching operators without importance sampling.
Optimal multiscale learning of linear operators
problem Statistical and computational limits of learning bounded linear operators between Sobolev spaces
method Reformulate as an infinite-dimensional matrix regression problem with heterogeneous multiscale structure
result Establish minimax rates and construct a finite-resolution blockwise least-squares estimator attaining these rates
New method preserves unitarity for Schrödinger equation learning, reducing errors and improving time generalization.
problem Learning the evolution operator for time-dependent Schrödinger equation with varying Hamiltonians.
method Linear estimator preserving weak unitarity, with theoretical error bounds and time generalization.
result Achieves up to two orders of magnitude smaller relative errors than existing methods.
Study identifies and analyzes three types of errors in learning Fourier operators.
problem Statistical, discretization, and truncation errors in learning Fourier operators.
method Analysis of a Discrete Fourier Transform (DFT) based least squares estimator.
result Established upper and lower bounds on statistical, discretization, and truncation errors.
New method estimates GGLM parameters, overcoming non-convexity.
problem Estimating parameters in GGLM with dependencies.
method Monotone operator-based variational inequality method.
result Guarantees for parameter recovery in GLM and GGLM.
Extended Gauss-Markov theorem for linear estimation with bounded bias.
problem Linear estimation with bounded bias operator.
method Derive optimal estimator formulas for Nuclear and Spectral norms, analyze generalization error.
result Cross-validated Nuclear and Spectral regressors outperform Ridge regression in simulations.
Bounds on spectral gaps of hyperbolic 3-manifolds and orbifolds.
problem Constraining the spectra of Laplace operators on hyperbolic manifolds and orbifolds.
method Linear programming and spectral identities derived from the conformal bootstrap and Selberg trace formula.
result Upper bounds on the first and second Laplacian eigenvalues, and spectral gaps of hyperbolic 3-manifolds and orbifolds.
This paper analyzes divide-and-conquer estimators for functional linear regression without assuming target function in the RKHS.
problem Functional linear regression without target function in RKHS.
method Integral operator approach to establish upper bounds and prove asymptotic optimality.
result Sharp finite sample upper bounds and asymptotic optimality of divide-and-conquer estimators.
Deep learning framework for kernel methods using RKHM and Perron-Frobenius operators.
problem Kernel methods in deep learning with potential overfitting issues.
method Combining RKHM and Perron-Frobenius operator to derive a new Rademacher bound and analyze deep kernel methods.
result Theoretical interpretation of benign overfitting and milder dependency on output dimension.
Study uses SGD to learn operators in Hilbert spaces with convergence analysis.
problem Learning operators in general Hilbert spaces with SGD.
method Proposes weak and strong regularity conditions for convergence analysis.
result SGD converges to best linear approximation of nonlinear operators.
On contact manifolds we describe a notion of (contact) finite-type for linear partial differential operators satisfying a natural condition on their leading terms. A large class of linear differential operators are of finite-type in this sense, and for any such operator we construct a partial connection on a (finite ra…
Study on neural scaling laws for solving linear systems in-context.
problem Theoretical guarantees for solving linear systems using a linear transformer architecture.
method Neural scaling laws and task diversity for in-domain and out-of-domain generalization.
result Novel notion of task diversity for necessary and sufficient condition of generalization under task shifts.
Active data collection improves convergence rates in operator learning.
problem Improving convergence rates in operator learning with linear target and stochastic input.
method Active data collection strategies with mean-zero stochastic process and continuous covariance kernels.
result Achieves arbitrarily fast error convergence rates with eigenvalue decay of covariance kernels.
In this work, we develop a simple algorithm for semi-supervised regression. The key idea is to use the top eigenfunctions of integral operator derived from both labeled and unlabeled examples as the basis functions and learn the prediction function by a simple linear regression. We show that under appropriate assumptio…
On a Riemannian metric-measure space, we establish an Alexandrov-Bakelman-Pucci type measure estimate connecting Bakry-Émery Ricci curvature lower bound, modified Laplacian and the measure of certain special sets. We apply this estimate to prove Harnack inequalities for the modified Laplacian operator and fully non-lin…
I prove the bistability of linear evolution equations x′=A(t)x in a Banach space E, where the operator-valued function A is of the form A(t)=f′(t)G(t,f(t)) for a binary operator-valued function G and a scalar function f. The constant that bounds the solutions of the equation is computed explicitly; it i…
This work provides lower bounds for differentiable games and defines a new condition number.
problem Understanding the fundamental limits of convergence in differentiable games.
method The authors cast saddle-point and min-max problems as 2-player games and use tools from single-objective convex optimization to derive linear lower bounds for convex-concave games. They also introduce a new condition number for games.
result The authors provide linear lower bounds for differentiable games, including n-player games, and introduce a new condition number that captures the possibility of linear rates in games without strong convexity or concavity. Given a complex Hilbert space H, we study the differential geometry of the manifold A of normal algebraic elements in Z=L(H), the algebra of bounded linear operators on H. We represent A as a disjoint union of subsets M of Z and, using the algebraic structure of Z, a torsionfree affine connection ∇ (that is inva…
The paper finds inequalities for eigenvalues of fourth order elliptic operators on Riemannian manifolds.
problem Eigenvalue inequalities for fourth order elliptic operators on Riemannian manifolds.
method Analyzes eigenvalues of fourth order elliptic operators in divergence form with Dirichlet boundary conditions on bounded domains in compact Riemannian manifolds.
result General inequalities for eigenvalues are derived.
Study on variance estimation for dynamic regression with finite sample guarantees.
problem Variance estimation for dynamic linear regression with non-constant observation operator.
method Analysis of the system operator's spectrum to derive variance estimators with finite sample complexity guarantees.
result First known variance estimators with finite sample complexity guarantees for dynamic regression.
Adapting the method of Andrews-Clutterbuck we prove an eigenvalue gap theorem for a class of non symmetric second order linear elliptic operators on a convex domain in euclidean space. The class of operators includes the Bakry-Emery laplacian with potential and any operator with second order term the laplacian whose fi…
The paper studies asymptotic expansions of operators related to Bochner-Schrödinger on Riemannian manifolds.
problem Asymptotic expansions of operators related to Bochner-Schrödinger on Riemannian manifolds.
method Analyzes the Bochner-Schrödinger operator Hp and its function φ(Hp) in L2(X,Lp⊗E), providing an asymptotic expansion of its smooth Schwartz kernel. result The trace of the operator φ(Hp) admits a complete asymptotic expansion in powers of p−1/2 as po∞. This paper presents a general coding method where data in a Hilbert space are represented by finite dimensional coding vectors. The method is based on empirical risk minimization within a certain class of linear operators, which map the set of coding vectors to the Hilbert space. Two results bounding the expected recon…
Study shows zero-shot super-resolution in neural operators is impossible in many cases.
problem Understanding the theoretical limits of zero-shot super-resolution in neural operators.
method Systematic theoretical study including information-theoretic and generalization bounds analysis.
result Zero-shot super-resolution is information-theoretically impossible in many settings.
Calculates spectral flow bounds for reducible solutions to Vafa-Witten equations.
problem Bounding spectral flow between diverging reducible solutions.
method Localization and excision techniques to calculate spectral flow.
result Bounds on spectral flow are given for reducible solutions.
New bounds for adaptive control in high dimensions without fixed state space.
problem Adaptive control of linear systems in high or infinite dimensions.
method Novel perturbation bound for certainty equivalence, scaling with prediction error.
result First regret bounds for LQR in infinite dimensional systems, independent of ambient dimension.
Paper develops a method to learn optimal sparsity-promoting regularizers for linear inverse problems.
problem Solving linear inverse problems with sparse solutions.
method Bilevel optimization framework to select an optimal synthesis operator B. result Established well-posedness and theoretical guarantees for the learning process.
The paper provides a concentration result and sample complexity for linear Monge mapping estimation and its application in domain adaptation.
problem Estimating the linear Monge mapping between distributions and its application in domain adaptation.
method The approach involves proving a concentration result and sample complexity for the linear mapping operator, and using it to derive a generalization bound for domain adaptation with optimal transport.
result The method achieves a sample complexity of n−1/2 and approaches the performance of theoretical Bayes predictor under mild conditions. Uniform elliptic theory for Dirac operators on orbifold resolutions.
problem Analyzing Dirac operators on orbifold resolutions.
method Viewing orbifolds as conically fibred singular spaces and resolving them by gluing asymptotically conical fibrations.
result Uniform index formula for Dirac operators on orbifold resolutions.
Study the Bochner-Schrödinger operator's trace in semiclassical limit.
problem Trace formula for Bochner-Schrödinger operator on tensor powers of line and vector bundles.
method Semiclassical analysis of the Bochner-Schrödinger operator Hp on tensor powers of a Hermitian line bundle and vector bundle. result Complete asymptotic expansion of the trace of φ(Hp) in the semiclassical limit po∞. We show how the discovery of robust scalable numerical solvers for arbitrary bounded linear operators can be automated as a Game Theory problem by reformulating the process of computing with partial information and limited resources as that of playing underlying hierarchies of adversarial information games. When the so…
The paper develops a method for Gaussian Process regression with linear operator constraints.
problem Modeling functions with multiple linear constraints in high-consequence engineering systems.
method Develops a method for constrained Gaussian Process regression using linear operator constraints.
result Derives the exact posterior for a conjugate likelihood under linear operator constraints.
We generalize the Omori-Yau almost maximum principle of the Laplace-Beltrami operator on a complete Riemannian manifold M to a second-order linear semi-elliptic operator L with bounded coefficients and no zeroth order term. Using this result, we prove some Liouville-type theorems for a real-valued C2 function …
Data-driven control of robotic systems using Koopman operators with error bounds.
problem Real-time control of nonlinear robotic systems with unknown dynamics.
method Constructing a Koopman operator-based linear representation using higher-order derivatives of nonlinear dynamics, with error bounds derived from Taylor series accuracy analysis.
result The Koopman model provides marginally better performance than competing nonlinear modeling methods and can be efficiently controlled using linear control design tools.
Devoted to multi-task learning and structured output learning, operator-valued kernels provide a flexible tool to build vector-valued functions in the context of Reproducing Kernel Hilbert Spaces. To scale up these methods, we extend the celebrated Random Fourier Feature methodology to get an approximation of operator-…
We compute estimates for eigenvalues of a class of linear second-order elliptic differential operators in divergence form (with Dirichlet boundary condition) on a bounded domain in a complete Riemannian manifold. Our estimates are based upon the Weyl's asymptotic formula. As an application, we find a lower bound for th…
Study variance-reduced method for estimating fixed points in Banach spaces.
problem Estimating fixed points of contractive operators in Banach spaces with noisy evaluations.
method Variance-reduced stochastic approximation scheme in Banach spaces.
result Establish non-asymptotic bounds for operator defect and estimation error.
We relate canonical algebraic curvature tensors that are built from a self-adjoint (RAS) or skew adjoint (RAΛ) linear operator A. Several authors have proven that any algebraic curvature tensor R may be expressed as a sum of RAS, or as a sum of RAΛ. This motivates our interest in relating them as well…
The developments of deep neural networks (DNN) in recent years have ushered a brand new era of artificial intelligence. DNNs are proved to be excellent in solving very complex problems, e.g., visual recognition and text understanding, to the extent of competing with or even surpassing people. Despite inspiring and enco…
We exhibit a knot P in the solid torus, representing a generator of first homology, such that for any knot K in the 3-sphere, the satellite knot with pattern P and companion K is not smoothly slice in any homology 4-ball. As a consequence, we obtain a knot in a homology 3-sphere that does not bound a piecewise-…
Linear Q-learning converges to a bounded set without divergence.
problem Proving linear Q-learning does not diverge and converges to a bounded set.
method No modifications to the original linear Q-learning algorithm, no Bellman completeness or near-optimality assumptions, only an ε-softmax behavior policy with adaptive temperature.
result First L2 convergence rate of linear Q-learning iterates to a bounded set.