Yinyu Ye
YOU?
Author Swipe
View article: Gradient Methods with Online Scaling Part II. Practical Aspects
Gradient Methods with Online Scaling Part II. Practical Aspects Open
Part I of this work [Gao25] establishes online scaled gradient methods (OSGM), a framework that utilizes online convex optimization to adapt stepsizes in gradient methods. This paper focuses on the practical aspects of OSGM. We leverage th…
View article: Adaptively Robust LLM Inference Optimization under Prediction Uncertainty
Adaptively Robust LLM Inference Optimization under Prediction Uncertainty Open
We study the problem of optimizing Large Language Model (LLM) inference scheduling to minimize total latency. LLM inference is an online and multi-task service process and also heavily energy consuming by which a pre-trained LLM processes …
View article: Real-Time, Population-Based Reconstruction of 3D Bone Models via Very-Low-Dose Protocols
Real-Time, Population-Based Reconstruction of 3D Bone Models via Very-Low-Dose Protocols Open
Patient-specific bone models are essential for designing surgical guides and preoperative planning, as they enable the visualization of intricate anatomical structures. However, traditional CT-based approaches for creating bone models are …
View article: LLM Serving Optimization with Variable Prefill and Decode Lengths
LLM Serving Optimization with Variable Prefill and Decode Lengths Open
We study the problem of serving LLM (Large Language Model) requests where each request has heterogeneous prefill and decode lengths. In LLM serving, the prefill length corresponds to the input prompt length, which determines the initial me…
View article: The Second-Order Tâtonnement: Decentralized Interior-Point Methods for Market Equilibrium
The Second-Order Tâtonnement: Decentralized Interior-Point Methods for Market Equilibrium Open
The tâtonnement process and Smale's process are two classical approaches to compute market equilibrium in exchange economies. While the tâtonnement process can be seen as a first-order method, Smale's process, being second-order, is less p…
View article: Adjoint-Based Aerodynamic Shape Optimization with a Manifold Constraint Learned by Diffusion Models
Adjoint-Based Aerodynamic Shape Optimization with a Manifold Constraint Learned by Diffusion Models Open
We introduce an adjoint-based aerodynamic shape optimization framework that integrates a diffusion model trained on existing designs to learn a smooth manifold of aerodynamically viable shapes. This manifold is enforced as an equality cons…
View article: Quantum Algorithms for Bandits with Knapsacks with Improved Regret and Time Complexities
Quantum Algorithms for Bandits with Knapsacks with Improved Regret and Time Complexities Open
Bandits with knapsacks (BwK) constitute a fundamental model that combines aspects of stochastic integer programming with online learning. Classical algorithms for BwK with a time horizon $T$ achieve a problem-independent regret bound of ${…
View article: PDHCG: A Scalable First-Order Method for Large-Scale Competitive Market Equilibrium Computation
PDHCG: A Scalable First-Order Method for Large-Scale Competitive Market Equilibrium Computation Open
Large-scale competitive market equilibrium problems arise in a wide range of important applications, including economic decision-making and intelligent manufacturing. Traditional solution methods, such as interior-point algorithms and cert…
View article: Gradient Methods with Online Scaling Part I. Theoretical Foundations
Gradient Methods with Online Scaling Part I. Theoretical Foundations Open
This paper establishes the theoretical foundations of the online scaled gradient methods (OSGM), a framework that utilizes online learning to adapt stepsizes and provably accelerate first-order methods. OSGM quantifies the effectiveness of…
View article: Adaptive Resolving Methods for Reinforcement Learning with Function Approximations
Adaptive Resolving Methods for Reinforcement Learning with Function Approximations Open
Reinforcement learning (RL) problems are fundamental in online decision-making and have been instrumental in finding an optimal policy for Markov decision processes (MDPs). Function approximations are usually deployed to handle large or in…
View article: Algorithm 1055: HDSDP: Software for Semidefinite Programming
Algorithm 1055: HDSDP: Software for Semidefinite Programming Open
HDSDP is a numerical software solving semidefinite programming problems. The main framework of HDSDP resembles the dual-scaling interior point solver DSDP and several new features, including a dual method based on the simplified homogeneou…
View article: Provable and Practical Online Learning Rate Adaptation with Hypergradient Descent
Provable and Practical Online Learning Rate Adaptation with Hypergradient Descent Open
This paper investigates the convergence properties of the hypergradient descent method (HDM), a 25-year-old heuristic originally proposed for adaptive stepsize selection in stochastic first-order methods. We provide the first rigorous conv…
View article: Wait-Less Offline Tuning and Re-solving for Online Decision Making
Wait-Less Offline Tuning and Re-solving for Online Decision Making Open
Online linear programming (OLP) has found broad applications in revenue management and resource allocation. State-of-the-art OLP algorithms achieve low regret by repeatedly solving linear programming (LP) subproblems that incorporate updat…
View article: When Does Primal Interior Point Method Beat Primal-dual in Linear Optimization?
When Does Primal Interior Point Method Beat Primal-dual in Linear Optimization? Open
The primal-dual interior point method (IPM) is widely regarded as the most efficient IPM variant for linear optimization. In this paper, we demonstrate that the improved stability of the pure primal IPM can allow speedups relative to a pri…
View article: Gradient Methods with Online Scaling
Gradient Methods with Online Scaling Open
We introduce a framework to accelerate the convergence of gradient-based methods with online learning. The framework learns to scale the gradient at each iteration through an online learning algorithm and provably accelerates gradient-base…
View article: Algorithm 1053: SOLNP+: A Derivative-Free Solver for Constrained Nonlinear Optimization
Algorithm 1053: SOLNP+: A Derivative-Free Solver for Constrained Nonlinear Optimization Open
SOLNP \(+\) is a derivative-free solver for constrained nonlinear optimization. It starts from SOLve Nonlinear Programming (SOLNP) proposed in 1989 by Ye. The main ideas are to use finite difference to approximate the gradient of the objec…
View article: Online Linear Programming with Batching
Online Linear Programming with Batching Open
We study Online Linear Programming (OLP) with batching. The planning horizon is cut into $K$ batches, and the decisions on customers arriving within a batch can be delayed to the end of their associated batch. Compared with OLP without bat…
View article: Accelerating Low-Rank Factorization-Based Semidefinite Programming Algorithms on GPU
Accelerating Low-Rank Factorization-Based Semidefinite Programming Algorithms on GPU Open
In this paper, we address a long-standing challenge: how to achieve both efficiency and scalability in solving semidefinite programming problems. We propose breakthrough acceleration techniques for a wide range of low-rank factorization-ba…
View article: Computationally Efficient Estimation of Large Probit Models
Computationally Efficient Estimation of Large Probit Models Open
Probit models are useful for modeling correlated discrete responses in many disciplines, including consumer choice data in economics and marketing. However, the Gaussian latent variable feature of probit models coupled with identification …
View article: Data-driven aerodynamic shape design with distributionally robust optimization approaches
Data-driven aerodynamic shape design with distributionally robust optimization approaches Open
We formulate and solve data-driven aerodynamic shape design problems with distributionally robust optimization (DRO) approaches. DRO aims to minimize the worst-case expected performance in a set of distributions that is informed by observe…
View article: A Single-Loop Robust Policy Gradient Method for Robust Markov Decision Processes
A Single-Loop Robust Policy Gradient Method for Robust Markov Decision Processes Open
Robust Markov Decision Processes (RMDPs) have recently been recognized as a valuable and promising approach to discovering a policy with creditable performance, particularly in the presence of a dynamic environment and estimation errors in…
View article: Sketched Newton Value Iteration for Large-Scale Markov Decision Processes
Sketched Newton Value Iteration for Large-Scale Markov Decision Processes Open
Value Iteration (VI) is one of the most classic algorithms for solving Markov Decision Processes (MDPs), which lays the foundations for various more advanced reinforcement learning algorithms, such as Q-learning. VI may take a large number…
View article: Trust Region Methods for Nonconvex Stochastic Optimization beyond Lipschitz Smoothness
Trust Region Methods for Nonconvex Stochastic Optimization beyond Lipschitz Smoothness Open
In many important machine learning applications, the standard assumption of having a globally Lipschitz continuous gradient may fail to hold. This paper delves into a more general (L0, L1)-smoothness setting, which gains particular signifi…
View article: A Low-Rank ADMM Splitting Approach for Semidefinite Programming
A Low-Rank ADMM Splitting Approach for Semidefinite Programming Open
We introduce a new first-order method for solving general semidefinite programming problems, based on the alternating direction method of multipliers (ADMM) and a matrix-splitting technique. Our algorithm has an advantage over the Burer-Mo…
View article: Achieving Instance-dependent Sample Complexity for Constrained Markov Decision Process
Achieving Instance-dependent Sample Complexity for Constrained Markov Decision Process Open
We consider the reinforcement learning problem for the constrained Markov decision process (CMDP), which plays a central role in satisfying safety or resource constraints in sequential learning and decision-making. In this problem, we are …
View article: TriAug: Out-of-Distribution Detection for Imbalanced Breast Lesion in Ultrasound
TriAug: Out-of-Distribution Detection for Imbalanced Breast Lesion in Ultrasound Open
Different diseases, such as histological subtypes of breast lesions, have severely varying incidence rates. Even trained with substantial amount of in-distribution (ID) data, models often encounter out-of-distribution (OOD) samples belongi…
View article: Decoupling Learning and Decision-Making: Breaking the $\mathcal{O}(\sqrt{T})$ Barrier in Online Resource Allocation with First-Order Methods
Decoupling Learning and Decision-Making: Breaking the $\mathcal{O}(\sqrt{T})$ Barrier in Online Resource Allocation with First-Order Methods Open
Online linear programming plays an important role in both revenue management and resource allocation, and recent research has focused on developing efficient first-order online learning algorithms. Despite the empirical success of first-or…
View article: A Tuning-Free Primal-Dual Splitting Algorithm for Large-Scale Semidefinite Programming
A Tuning-Free Primal-Dual Splitting Algorithm for Large-Scale Semidefinite Programming Open
This paper proposes and analyzes a tuning-free variant of Primal-Dual Hybrid Gradient (PDHG), and investigates its effectiveness for solving large-scale semidefinite programming (SDP). The core idea is based on the combination of two seemi…
View article: Scalable Approximate Optimal Diagonal Preconditioning
Scalable Approximate Optimal Diagonal Preconditioning Open
We consider the problem of finding the optimal diagonal preconditioner for a positive definite matrix. Although this problem has been shown to be solvable and various methods have been proposed, none of the existing approaches are scalable…