no code implementations • ICML 2020 • Peng Wang, Zirui Zhou, Anthony Man-Cho So
In this paper, we focus on the problem of exactly recovering the communities in a binary symmetric SBM, where a graph of $n$ vertices is partitioned into two equal-sized communities and the vertices are connected with probability $p = \alpha\log(n)/n$ within communities and $q = \beta\log(n)/n$ across communities for some $\alpha>\beta>0$.
no code implementations • 11 Apr 2024 • He Chen, Jiajin Li, Anthony Man-Cho So
Despite the considerable success of Bregman proximal-type algorithms, such as mirror descent, in machine learning, a critical question remains: Can existing stationarity measures, often based on Bregman divergence, reliably distinguish between stationary and non-stationary points?
no code implementations • 11 Mar 2024 • Junbin Liu, Ya Liu, Wing-Kin Ma, Mingjie Shao, Anthony Man-Cho So
This study develops a framework for a class of constant modulus (CM) optimization problems, which covers binary constraints, discrete phase constraints, semi-orthogonal matrix constraints, non-negative semi-orthogonal matrix constraints, and several types of binary assignment constraints.
no code implementations • 11 Mar 2024 • Junbin Liu, Ya Liu, Wing-Kin Ma, Mingjie Shao, Anthony Man-Cho So
In the first part of this study, a convex-constrained penalized formulation was studied for a class of constant modulus (CM) problems.
no code implementations • 6 Dec 2023 • Jinxin Wang, Chonghe Jiang, Huikang Liu, Anthony Man-Cho So
The heteroscedastic probabilistic principal component analysis (PCA) technique, a variant of the classic PCA that considers data heterogeneity, is receiving more and more attention in the data science and signal processing communities.
1 code implementation • NeurIPS 2023 • Huikang Liu, Xiao Li, Anthony Man-Cho So
This work presents ReSync, a Riemannian subgradient-based algorithm for solving the robust rotation synchronization problem, which arises in various engineering applications.
no code implementations • 23 May 2023 • Xiao Li, Lei Zhao, Daoli Zhu, Anthony Man-Cho So
In particular, when $f$ is convex, we show $\mathcal{O}(\log(k)/\sqrt{k})$ rate of convergence in terms of the suboptimality gap.
no code implementations • 8 Apr 2023 • Zihao Fu, Wai Lam, Qian Yu, Anthony Man-Cho So, Shengding Hu, Zhiyuan Liu, Nigel Collier
Grounded on our analysis, we propose a novel partial attention language model to solve the attention degeneration problem.
no code implementations • 31 Mar 2023 • Jinxin Wang, Jiang Hu, Shixiang Chen, Zengde Deng, Anthony Man-Cho So
We focus on a class of non-smooth optimization problems over the Stiefel manifold in the decentralized setting, where a connected network of $n$ agents cooperatively minimize a finite-sum objective function with each component being weakly convex in the ambient Euclidean space.
2 code implementations • 12 Mar 2023 • Jiajin Li, Jianheng Tang, Lemin Kong, Huikang Liu, Jia Li, Anthony Man-Cho So, Jose Blanchet
This observation allows us to provide an approximation bound for the distance between the fixed-point set of BAPG and the critical point set of GW.
no code implementations • 23 Feb 2023 • Lai Tian, Anthony Man-Cho So
This implies that testing a certain stationarity concept for a modern nonsmooth neural network is in general computationally intractable.
1 code implementation • NeurIPS 2023 • Lemin Kong, Jiajin Li, Jianheng Tang, Anthony Man-Cho So
Gromov-Wasserstein (GW) distance is a powerful tool for comparing and aligning probability distributions supported on different metric spaces.
no code implementations • 24 Jan 2023 • Zihao Fu, Anthony Man-Cho So, Nigel Collier
The theoretical bounds explain why and how several existing methods can stabilize the fine-tuning procedure.
1 code implementation • 28 Nov 2022 • Zihao Fu, Haoran Yang, Anthony Man-Cho So, Wai Lam, Lidong Bing, Nigel Collier
How to choose the tunable parameters?
no code implementations • 22 Sep 2022 • Jiajin Li, Linglingzhi Zhu, Anthony Man-Cho So
Specifically, we consider the setting where the primal function has a nonsmooth composite structure and the dual function possesses the Kurdyka-Lojasiewicz (KL) property with exponent $\theta \in [0, 1)$.
no code implementations • 15 Jul 2022 • Jiang Hu, Ruicheng Ao, Anthony Man-Cho So, MingHan Yang, Zaiwen Wen
Moreover, we show that if the loss function satisfies certain convexity and smoothness conditions and the input-output map satisfies a Riemannian Jacobian stability condition, then our proposed method enjoys a local linear -- or, under the Lipschitz continuity of the Riemannian Jacobian of the input-output map, even quadratic -- rate of convergence.
1 code implementation • 11 Jun 2022 • Peng Wang, Huikang Liu, Anthony Man-Cho So, Laura Balzano
The K-subspaces (KSS) method is a generalization of the K-means method for subspace clustering.
no code implementations • 17 May 2022 • Jiajin Li, Jianheng Tang, Lemin Kong, Huikang Liu, Jia Li, Anthony Man-Cho So, Jose Blanchet
In this paper, we study the design and analysis of a class of efficient algorithms for computing the Gromov-Wasserstein (GW) distance tailored to large-scale graph learning tasks.
no code implementations • 22 Feb 2022 • Xiaolu Wang, Peng Wang, Anthony Man-Cho So
Signed graphs encode similarity and dissimilarity relationships among different entities with positive and negative edges.
no code implementations • 28 Dec 2021 • Sijin Chen, Xiwei Cheng, Anthony Man-Cho So
This paper proposes a Generalized Power Method (GPM) to tackle the problem of community detection and group synchronization simultaneously in a direct non-convex manner.
no code implementations • 13 Dec 2021 • Linglingzhi Zhu, Jinxin Wang, Anthony Man-Cho So
In this paper, we focus on the orthogonal group synchronization problem with general additive noise models under incomplete measurements, which is much more general than the commonly considered setting of complete measurements.
no code implementations • 30 Sep 2021 • Kaiwen Zhou, Anthony Man-Cho So, James Cheng
We show that stochastic acceleration can be achieved under the perturbed iterate framework (Mania et al., 2017) in asynchronous lock-free optimization, which leads to the optimal incremental gradient complexity for finite-sum objectives.
no code implementations • 1 Jul 2021 • Chujun Huang, Mingjie Shao, Wing-Kin Ma, Anthony Man-Cho So
By establishing associations between the SISAL algorithm and a line-search-based proximal gradient method, we confirm that SISAL can indeed guarantee convergence to a stationary point.
no code implementations • NeurIPS 2021 • Kaiwen Zhou, Lai Tian, Anthony Man-Cho So, James Cheng
In convex optimization, the problem of finding near-stationary points has not been adequately studied yet, unlike other optimality measures such as the function value.
no code implementations • 12 May 2021 • Xiaolu Wang, Yuen-Man Pun, Anthony Man-Cho So
To address this issue, we propose a novel graph learning model based on the distributionally robust optimization methodology, which aims to identify a graph that not only provides a smooth representation of but is also robust against uncertainties in the observed signals.
no code implementations • 18 Mar 2021 • Ruiyuan Wu, Wing-Kin Ma, Yuening Li, Anthony Man-Cho So, Nicholas D. Sidiropoulos
PRISM uses a simple probabilistic model, namely, uniform simplex data distribution and additive Gaussian noise, and it carries out inference by maximum likelihood.
1 code implementation • 29 Dec 2020 • Zihao Fu, Wai Lam, Anthony Man-Cho So, Bei Shi
The experimental results show that our theoretical framework is applicable in general generation models and our proposed rebalanced encoding approach alleviates the repetition problem significantly.
1 code implementation • NeurIPS 2020 • Jiajin Li, Caihua Chen, Anthony Man-Cho So
In this paper, we focus on a family of Wasserstein distributionally robust support vector machine (DRSVM) problems and propose two novel epigraphical projection-based incremental algorithms to solve them.
no code implementations • 16 Sep 2020 • Huikang Liu, Man-Chung Yue, Anthony Man-Cho So
In this paper, we consider the class of synchronization problems in which the group is a closed subgroup of the orthogonal group.
no code implementations • 29 Aug 2020 • Jinxin Wang, Zengde Deng, Taoli Zheng, Anthony Man-Cho So
Optimization for the MVSKC model is of great difficulty in two parts.
1 code implementation • 29 Jun 2020 • Peng Wang, Zirui Zhou, Anthony Man-Cho So
Community detection in graphs that are generated according to stochastic block models (SBMs) has received much attention lately.
no code implementations • 26 Jun 2020 • Jiajin Li, Anthony Man-Cho So, Wing-Kin Ma
Many contemporary applications in signal processing and machine learning give rise to structured non-convex non-smooth optimization problems that can often be tackled by simple iterative methods quite effectively.
1 code implementation • NeurIPS 2020 • Kaiwen Zhou, Anthony Man-Cho So, James Cheng
Specifically, instead of tackling the original objective directly, we construct a shifted objective function that has the same minimizer as the original objective and encodes both the smoothness and strong convexity of the original objective in an interpolation condition.
no code implementations • 5 May 2020 • Shixiang Chen, Zengde Deng, Shiqian Ma, Anthony Man-Cho So
Second, we propose a stochastic variant of ManPPA called StManPPA, which is well suited for large-scale computation, and establish its sublinear convergence rate.
1 code implementation • NeurIPS 2019 • Jiajin Li, Sen Huang, Anthony Man-Cho So
In this paper, we take a first step towards resolving the above difficulty by developing a first-order algorithmic framework for tackling a class of Wasserstein distance-based distributionally robust logistic regression (DRLR) problem.
1 code implementation • 28 Oct 2019 • Jiajin Li, Sen Huang, Anthony Man-Cho So
In this paper, we take a first step towards resolving the above difficulty by developing a first-order algorithmic framework for tackling a class of Wasserstein distance-based distributionally robust logistic regression (DRLR) problem.
no code implementations • 2 Jul 2019 • Yue Xu, Zengde Deng, Mengdi Wang, Wenjun Xu, Anthony Man-Cho So, Shuguang Cui
The recent success of single-agent reinforcement learning (RL) in Internet of things (IoT) systems motivates the study of multi-agent reinforcement learning (MARL), which is more challenging but more useful in large-scale IoT.
no code implementations • 12 Mar 2019 • Zengde Deng, Anthony Man-Cho So
Variable selection is one of the most important tasks in statistics and machine learning.
no code implementations • 2 Nov 2018 • Shixiang Chen, Shiqian Ma, Anthony Man-Cho So, Tong Zhang
We prove that the proposed method globally converges to a stationary point.
no code implementations • 24 Sep 2018 • Xiao Li, Zhihui Zhu, Anthony Man-Cho So, Rene Vidal
In this paper we study the problem of recovering a low-rank matrix from a number of random linear measurements that are corrupted by outliers taking arbitrary values.
Information Theory Information Theory
no code implementations • 31 May 2016 • Xiao Fu, Kejun Huang, Mingyi Hong, Nicholas D. Sidiropoulos, Anthony Man-Cho So
Generalized canonical correlation analysis (GCCA) aims at finding latent low-dimensional common structure from multiple views (feature vectors in different domains) of the same entities.
no code implementations • 11 Dec 2015 • Zirui Zhou, Anthony Man-Cho So
In this paper, we present a new framework for establishing error bounds for a class of structured convex optimization problems, in which the objective function is the sum of a smooth convex function and a general closed proper convex function.
no code implementations • 5 Oct 2015 • Huikang Liu, Weijie Wu, Anthony Man-Cho So
To determine the convergence rate of these methods, we give an explicit estimate of the exponent in a Lojasiewicz inequality for the (non-convex) set of critical points of the aforementioned class of problems.
no code implementations • NeurIPS 2013 • Ke Hou, Zirui Zhou, Anthony Man-Cho So, Zhi-Quan Luo
Motivated by various applications in machine learning, the problem of minimizing a convex smooth loss function with trace norm regularization has received much attention lately.
no code implementations • 31 Aug 2013 • Anthony Man-Cho So
In this paper, we combine the best of these two types of results and establish---under the standard assumption that the gradient approximation errors decrease linearly to zero---the non-asymptotic linear convergence of IGMs when applied to a class of structured convex optimization problems.