no code implementations • 6 May 2024 • Zhiyao Tan, Ling Zhou, Huazhen Lin
In particular, we first provide both theoretical and practical evidence to validate the existence of an optimal input dimension (OID) that minimizes the generalization error.
no code implementations • 16 Oct 2023 • Ling Zhou, Mingpei Wang, Xiaohua Huang, Wenming Zheng, Qirong Mao, Guoying Zhao
Micro-expression recognition (MER) in low-resolution (LR) scenarios presents an important and complex challenge, particularly for practical applications such as group MER in crowded environments.
no code implementations • 14 Dec 2022 • Luis Scoccola, Hitesh Gakhar, Johnathan Bush, Nikolas Schonsheck, Tatum Rask, Ling Zhou, Jose A. Perea
The circular coordinates algorithm of de Silva, Morozov, and Vejdemo-Johansson takes as input a dataset together with a cohomology class representing a $1$-dimensional hole in the data; the output is a map from the data into the circle that captures this hole, and that is of minimum energy in a suitable sense.
no code implementations • 4 Jan 2022 • Wen Wang, Shihao Wu, Ziwei Zhu, Ling Zhou, Peter X. -K. Song
Fusing regression coefficients into homogenous groups can unveil those coefficients that share a common value within each group.
no code implementations • 13 Jul 2021 • Qirong Mao, Ling Zhou, Wenming Zheng, Xiuyan Shao, Xiaohua Huang
More specifically, the backbone network aims at extracting feature representations from different facial regions, RI module computing an adaptive weight from the region itself based on attention mechanism with respect to the unobstructedness and importance for suppressing the influence of occlusion, and RR module exploiting the progressive interactions among these regions by performing graph convolutions.
1 code implementation • 10 Mar 2021 • Xiaoqing Tan, Chung-Chou H. Chang, Ling Zhou, Lu Tang
We propose a tree-based model averaging approach to improve the estimation accuracy of conditional average treatment effects (CATE) at a target site by leveraging models derived from other potentially heterogeneous sites, without them sharing subject-level data.
no code implementations • 13 Jan 2021 • Ling Zhou, Qirong Mao, Xiaohua Huang, Feifei Zhang, Zhihong Zhang
It aims to obtain salient and discriminative features for specific expressions and also predict expression by fusing the expression-specific features.
Micro Expression Recognition Micro-Expression Recognition +1
no code implementations • 24 Dec 2020 • Ling Zhou, Qirong Mao, Ming Dong
Specifically, we propose two new strategies in our AU detection module for more effective AU feature learning: the attention mechanism and the balanced detection loss function.
no code implementations • CVPR 2020 • Ling Zhou, Zhen Cui, Chunyan Xu, Zhenyu Zhang, Chaoqun Wang, Tong Zhang, Jian Yang
Inspired by the observation that pattern structures high-frequently recur within intra-task also across tasks, we propose a pattern-structure diffusion (PSD) framework to mine and propagate task-specific and task-across pattern structures in the task-level space for joint depth estimation, segmentation and surface normal prediction.
Ranked #12 on Semantic Segmentation on SUN-RGBD (using extra training data)
1 code implementation • 28 Dec 2019 • Facundo Mémoli, Ling Zhou
We pay particular attention to the case of fundamental groups, for which we obtain a more precise description.
Algebraic Topology Computational Geometry 53C23, 51F99, 55N35
no code implementations • 4 Aug 2019 • Fei Wang, Ling Zhou, Lu Tang, Peter X. -K. Song
To establish a simultaneous post-model selection inference, we propose a method of contraction and expansion (MOCE) along the line of debiasing estimation that enables us to balance the bias-and-variance trade-off so that the super-sparsity assumption may be relaxed.
no code implementations • 11 Feb 2019 • Xiao Dong, Ling Zhou
This can be regarded as a strong support of our proposal that geometrization is not only the bible for physics, it is also the key idea to understand deep learning systems.
no code implementations • 6 Jan 2019 • Xiao Dong, Ling Zhou
By comparing the geometry of image matching and deep networks, we show that geometrization of deep networks can be used to understand existing deep learning systems and it may also help to solve the interpretability problem of deep learning systems.
1 code implementation • 22 May 2018 • Tong Qin, Ling Zhou, Dongbin Xiu
For neural networks (NNs) with rectified linear unit (ReLU) or binary activation functions, we show that their training can be accomplished in a reduced parameter space.
no code implementations • 24 Nov 2017 • Xiao Dong, Jiasong Wu, Ling Zhou
The astonishing success of AlphaGo Zero\cite{Silver_AlphaGo} invokes a worldwide discussion of the future of our human society with a mixed mood of hope, anxiousness, excitement and fear.
no code implementations • 30 Oct 2017 • Xiao Dong, Jiasong Wu, Ling Zhou
Why and how that deep learning works well on different tasks remains a mystery from a theoretical perspective.