Optimization

Freya PAGE achieves optimal time complexity for large-scale nonconvex finite-sum optimization using asynchronous and heterogeneous computations, overcoming limitations of prior methods.

First-order methods conquer linearly constrained bilevel optimization, achieving near-optimal convergence rates and enhancing high-dimensional applicability.

FERERO, a novel framework, tackles multi-objective learning by efficiently finding preference-guided Pareto solutions using flexible preference modeling and convergent algorithms.

Brain-inspired recurrent neural networks learn efficiently by using feedback control to approximate optimal gradients, enabling rapid movement corrections and efficient adaptation to persistent errors…

Faster primal-dual algorithms achieve order-optimal complexity for convex optimization with strongly convex constraints, improving convergence rates and solving large-scale problems efficiently.

Fast Tree-Field Integrators (FTFIs) revolutionize graph processing by enabling polylog-linear time computation for integrating tensor fields on trees, providing significant speedups for various machin…

Fast T2T: Optimization Consistency Boosts Diffusion-Based Combinatorial Optimization!

This paper introduces a novel approach for fast rates in online convex optimization by exploiting the curvature of feasible sets, achieving logarithmic regret bounds under specific conditions.

Forgetful algorithms are essential for fast last-iterate convergence in learning games; otherwise, even popular methods like OMWU fail.

Accelerate iterative hard thresholding (IHT) up to 73x by safely pruning unnecessary gradient computations without accuracy loss.

Polar codes revolutionize channel simulation, offering scalable, high-performance schemes that significantly outperform existing methods.

VIJI, a novel second-order algorithm, achieves optimal convergence rates for variational inequalities even with inexact Jacobian information, bridging the gap between theory and practice in machine le…

ENOT, a new Neural Optimal Transport training method, achieves 3x quality and 10x speed improvements by using expectile regularization to stabilize the learning process.

Exact Gauss-Newton optimization in deep reversible networks surprisingly reveals poor generalization, despite faster training, challenging existing deep learning optimization theories.

New framework uses rough path theory to enable gradient-based training of SSNNs driven by rough signals, allowing for noise in spike timing and network dynamics.

New consistent estimators precisely track generalization error during robust regression’s iterative model training, enabling optimal stopping iteration for minimized error.

This research paper presents a novel theoretical error analysis for the spherically constrained least squares (SCLS) method used to solve Stackelberg prediction games (SPGs). SPGs model strategic int…

This paper provides novel entrywise error bounds for low-rank kernel matrix approximations, showing how many data points are needed to get statistically consistent results for low-rank approximations.

This paper presents near-optimal algorithms for entropy identity testing, significantly improving Bayesian network testing efficiency.

New algorithm approximates continuous Entropic Optimal Transport (EOT) barycenters for any cost function, offering quality bounds and seamless EBM integration.