← Search

Jian-Feng Cai

17 accepted papers

2026

Column Thresholding for Sparse Spiked Wigner Models: Improved Signal Strength Requirements

ICML 2026poster

We study the sparse spiked Wigner model, where the goal is to recover an $s$-sparse unit vector $\symbfit{u} \in \mathbb{R}^d$ from a noisy observation $\symbfit{Y} = \beta \symbfit{u} \symbfit{u}^\top + \symbfit{W}$. While the information-theoretic threshold is $\beta = \widetilde{\Omega}(\sqrt{s})…

Cited by 0SourceScholar
2026

Fast Spectrally Sparse Signal Reconstruction via Jacobi-Preconditioned Gradient Descent

ICML 2026spotlight

Spectrally sparse signal reconstruction arises in a wide range of applications and can be formulated as a low-rank Hankel matrix completion problem. We develop a Jacobi-preconditioned gradient descent method that preserves the low per-iteration complexity of first-order algorithms while achieving li…

Cited by 0SourceScholar
2026

Improving Classifier-Free Guidance of Flow Matching via Manifold Projection

ICML 2026poster

Classifier-free guidance (CFG) is a widely used technique for controllable generation in diffusion and flow-based models. Despite its empirical success, CFG relies on a heuristic linear extrapolation that is often sensitive to the guidance scale. In this work, we provide a principled interpretation …

Cited by 0SourceScholar
2026

Online Tensor Learning: Computational and Statistical Trade-offs, Adaptivity and Optimal Regret

ICML 2026poster

Large tensor learning algorithms are typically computationally expensive and require storing a vast amount of data. In this paper, we propose a unified online Riemannian gradient descent (oRGrad) algorithm for tensor learning, which is computationally efficient, consumes much less memory, and can ha…

Cited by 0SourceScholar
2025

Fast and Provable Algorithms for Sparse PCA with Improved Sample Complexity

ICML 2025poster

We explore the single-spiked covariance model within the context of sparse principal component analysis (PCA), which aims to recover a sparse unit vector from noisy samples. From an information-theoretic perspective, $O(k \log p)$ observations are sufficient to recover a $k$-sparse $p$-dimensional v…

Cited by 0SourcePDFScholar
2025

Finding Low-Rank Matrix Weights in DNNs via Riemannian Optimization: RAdaGrad and RAdamW

NeurIPS 2025poster

Finding low-rank matrix weights is a key technique for addressing the high memory usage and computational demands of large models. Most existing algorithms rely on the factorization of the low-rank matrix weights, which is non-unique and redundant. Their convergence is slow especially when the targe…

Cited by 0SourceScholar
2025

Preconditioned Riemannian Gradient Descent Algorithm for Low-Multilinear-Rank Tensor Completion

ICML 2025poster

Tensors play a crucial role in numerous scientific and engineering fields. This paper addresses the low-multilinear-rank tensor completion problem, a fundamental task in tensor-related applications. By exploiting the manifold structure inherent to the fixed-multilinear-rank tensor set, we introduce…

Cited by 0SourcePDFScholar
2024

On the Convergence of Projected Bures-Wasserstein Gradient Descent under Euclidean Strong Convexity

ICML 2024poster

The Bures-Wasserstein (BW) gradient descent method has gained considerable attention in various domains, including Gaussian barycenter, matrix recovery and variational inference problems, due to its alignment with the Wasserstein geometry of normal distributions. Despite its popularity, existing con…

Cited by 0SourcePDFScholar
2024

RL in Markov Games with Independent Function Approximation: Improved Sample Complexity Bound under the Local Access Model

AISTATS 2024poster

Efficiently learning equilibria with large state and action spaces in general-sum Markov games while overcoming the curse of multi-agency is a challenging problem. Recent works have attempted to solve this problem by employing independent linear function classes to approximate the marginal $Q$-value…

Cited by 2SourcePDFScholar
2023

Fast Projected Newton-like Method for Precision Matrix Estimation under Total Positivity

NeurIPS 2023poster

We study the problem of estimating precision matrices in Gaussian distributions that are multivariate totally positive of order two ($\mathrm{MTP}_2$). The precision matrix in such a distribution is an M-matrix. This problem can be formulated as a sign-constrained log-determinant program. Current al…

Cited by 5SourcePDFScholar
2019

Fast Single Image Reflection Suppression via Convex Optimization

CVPR 2019poster

Removing undesired reflections from images taken through the glass is of great importance in computer vision. It serves as a means to enhance the image quality for aesthetic purposes as well as to preprocess images in machine learning and pattern recognition applications. We propose a convex model t…

Cited by 79PDFcodeScholar
2016

Fast alternating projected gradient descent algorithms for recovering spectrally sparse signals

ICASSP 2016accepted

We propose fast algorithms that speed up or improve the performance of recovering spectrally sparse signals from un-derdetermined measurements. Our algorithms are based on a non-convex approach of using alternating projected gradient descent for structured matrix recovery. We apply this approach to…

Cited by 0SourceScholar