arXiv research

Analyzes learning dynamics of RNNs under locality constraints.

Local LMO optimizes constrained problems using local linear minimization.

Local Linear embedding (LLE) is a popular dimension reduction method. In this paper, we first show LLE with nonnegative constraint is equivalent to the widely used Laplacian embedding. We further propose to iterate the two steps in LLE repeatedly to improve the results. Thirdly, we relax the kNN constraint of LLE and p…

New method for private linear regression under privacy constraints, achieving optimal rates.

Method solves nonconvex constrained optimization problems with a new augmented Lagrangian approach.

FedCONST adapts update magnitudes to enhance feature generalization in FL.

Enhances sequence labeling with embedded-state latent CRFs.

S.Bauer and M.Furuta defined a stable cohomotopy refinement of the Seiberg-Witten invariants. In this paper, we prove a vanishing theorem of Bauer-Furuta invariants for 4-manifolds with smooth Z/2-actions. As an application, we give a constraint on smooth Z/2-actions on homotopy K3#K3, and construct a nonsmoothable loc…

This paper achieves optimal regret bounds for locally private linear contextual bandit.

Proposes a matrix completion method for medical records with long time intervals.

It is known that fixed points of loopy belief propagation (BP) correspond to stationary points of the Bethe variational problem, where we minimize the Bethe free energy subject to normalization and marginalization constraints. Unfortunately, this does not entirely explain BP because BP is a dual rather than primal algo…

Local Linear Forests improve random forests for smooth signals and causal inference.

Data processing inequalities link Fisher information to local differential privacy constraints.

Optimizes stochastic convex optimization with local minimax theory and adaptive methods.

MVRSM optimizes expensive functions with mixed variables, outperforming state-of-the-art methods.

Holistic GLMs add constraints for better model quality.

This work is a further study on the Generalized Constraint Neural Network (GCNN) model [1], [2]. Two challenges are encountered in the study, that is, to embed any type of prior information and to select its imposing schemes. The work focuses on the second challenge and studies a new constraint imposing scheme for equa…

Locally Linear Embedding improves psychiatric diagnosis accuracy from fMRI data.

A large number of problems in optimization, machine learning, signal processing can be effectively addressed by suitable semidefinite programming (SDP) relaxations. Unfortunately, generic SDP solvers hardly scale beyond instances with a few hundreds variables (in the underlying combinatorial problem). On the other hand…

Paper proposes a privacy-preserving RL algorithm for linear MDPs with theoretical guarantees.

The paper improves SVR with linear constraints for better model properties.

Paper proposes a new sparse group k-max regularization for sparsity constraints.

Graphical notation simplifies complex polynomial constraints in linear models.

Regularized least-squares approaches have been successfully applied to linear system identification. Recent approaches use quadratic penalty terms on the unknown impulse response defined by stable spline kernels, which control model space complexity by leveraging regularity and bounded-input bounded-output stability. T…

New method for linear connections in ODEs with constraints.

In structured prediction problems where we have indirect supervision of the output, maximum marginal likelihood faces two computational obstacles: non-convexity of the objective and intractability of even a single gradient computation. In this paper, we bypass both obstacles for a class of what we call linear indirectl…

Controller-Augmented Hidden Markov Models (CHMMs) are a framework for constrained sequential inference.

Generative classifiers' properties are linked to linear constraints.

Proximal splitting methods solve rank-constrained convex problems locally.

We study the projected gradient descent method on low-rank matrix problems with a strongly convex objective. We use the Burer-Monteiro factorization approach to implicitly enforce low-rankness; such factorization introduces non-convexity in the objective. We focus on constraint sets that include both positive semi-defi…

New algorithms protect user data while optimizing personalized decisions.

FOSC-X: An extended framework for extracting multiple optimal flat clusterings from hierarchical cluster trees

The paper optimizes policies constrained to Schur stabilizing controllers using a Newton-type algorithm.

ACOL learns constraints from human preferences in driving simulations.

Unified approach tackles logical constraints in mixed-integer optimization.

On a Möbius surface, as defined by D. Calderbank, we study a variant of the Einstein-Weyl (EW) equation which we call scalar-flat Möbius EW (sf-MEW). This is a conformally invariant, finite type, overdetermined system of semi-linear partial differential equations. We derive local algebraic constraints for this equation…

Unified framework for integrating linear constraints in time series forecasting.

The paper gauges non-linear sigma models using Lie algebroid actions.

We identify linear dynamical systems under convex constraints with fewer samples.

This paper extends forecast reconciliation to non-linearly constrained time series.

We analyze the performance of a class of manifold-learning algorithms that find their output by minimizing a quadratic form under some normalization constraints. This class consists of Locally Linear Embedding (LLE), Laplacian Eigenmap, Local Tangent Space Alignment (LTSA), Hessian Eigenmaps (HLLE), and Diffusion maps.…

Safe algorithm for linear bandits with safety constraints, matching previous results.

New method parameterizes solutions to linearized vacuum constraints on Einstein manifolds.

We consider a model of linear market impact, and address the problem of replicating a contingent claim in this framework. We derive a non-linear Black-Scholes Equation that provides an exact replication strategy. This equation is fully non-linear and singular, but we show that it is well posed, and we prove existence o…

LP algorithm optimizes neural networks with architectural constraints.

This paper proposes a new method for GLM estimation using distance penalties to handle constraints.

New algorithm solves minimax games with linear constraints.

Generalized Linear Models (GLMs) and Single Index Models (SIMs) provide powerful generalizations of linear regression, where the target variable is assumed to be a (possibly unknown) 1-dimensional function of a linear predictor. In general, these problems entail non-convex estimation procedures, and, in practice, itera…