Author name cluster

Zimeng Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

ICML Conference 2025 Conference Paper

Approximation to Smooth Functions by Low-Rank Swish Networks

Zimeng Li
Hongjun Li
Jingyuan Wang
Ke Tang

While deep learning has witnessed remarkable achievements in a wide range of applications, its substantial computational cost imposes limitations on application scenarios of neural networks. To alleviate this problem, low-rank compression is proposed as a class of efficient and hardware-friendly network compression methods, which reduce computation by replacing large matrices in neural networks with products of two small ones. In this paper, we implement low-rank networks by inserting a sufficiently narrow linear layer without bias between each of two adjacent nonlinear layers. We prove that low-rank Swish networks with a fixed depth are capable of approximating any function from the Hölder ball $\mathcal{C}^{\beta, R}([0, 1]^d)$ within an arbitrarily small error where $\beta$ is the smooth parameter and $R$ is the radius. Our proposed constructive approximation ensures that the width of linear hidden layers required for approximation is no more than one-third of the width of nonlinear layers, which implies that the computational cost can be decreased by at least one-third compared with a network with the same depth and width of nonlinear layers but without narrow linear hidden layers. Our theoretical finding can offer a theoretical basis for low-rank compression from the perspective of universal approximation theory.

Details

AAAI Conference 2024 Conference Paper

Full Bayesian Significance Testing for Neural Networks

Zehua Liu
Zimeng Li
Jingyuan Wang
Yue He

Significance testing aims to determine whether a proposition about the population distribution is the truth or not given observations. However, traditional significance testing often needs to derive the distribution of the testing statistic, failing to deal with complex nonlinear relationships. In this paper, we propose to conduct Full Bayesian Significance Testing for neural networks, called nFBST, to overcome the limitation in relationship characterization of traditional approaches. A Bayesian neural network is utilized to fit the nonlinear and multi-dimensional relationships with small errors and avoid hard theoretical derivation by computing the evidence value. Besides, nFBST can test not only global significance but also local and instance-wise significance, which previous testing methods don't focus on. Moreover, nFBST is a general framework that can be extended based on the measures selected, such as Grad-nFBST, LRP-nFBST, DeepLIFT-nFBST, LIME-nFBST. A range of experiments on both simulated and real data are conducted to show the advantages of our method.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

Full Bayesian Significance Testing for Neural Networks in Traffic Forecasting

Zehua Liu
Jingyuan Wang
Zimeng Li
Yue He

Due to the complex and dynamic traffic contexts, the interpretability and uncertainty of traffic forecasting have gained increasing attention. Significance testing is a powerful tool in statistics used to determine whether a hypothesis is valid, facilitating the identification of pivotal features that predominantly contribute to the true relationship. However, existing works mainly regard traffic forecasting as a deterministic problem, making it challenging to perform effective significance testing. To fill this gap, we propose to conduct Full Bayesian Significance Testing for Neural Networks in Traffic Forecasting, namely ST-nFBST. A Bayesian neural network is utilized to capture the complicated traffic relationships through an optimization function resolved in the context of aleatoric uncertainty and epistemic uncertainty. Thereupon, ST-nFBST can achieve the significance testing by means of a delicate grad-based evidence value, further capturing the inherent traffic schema for better spatiotemporal modeling. Extensive experiments are conducted on METR-LA and PEMS-BAY to verify the advantages of our method in terms of uncertainty analysis and significance testing, helping the interpretability and promotion of traffic forecasting.

PDF Details DOI

JBHI Journal 2024 Journal Article

PEB-DDI: A Task-Specific Dual-View Substructural Learning Framework for Drug–Drug Interaction Prediction

Xiangzhen Shen
Zimeng Li
Yuansheng Liu
Bosheng Song
Xiangxiang Zeng

Adverse drug-drug interactions (DDIs) pose potential risks in polypharmacy due to unknown physicochemical incompatibilities between co-administered drugs. Recent studies have utilized multi-layer graph neural network architectures to model hierarchical molecular substructures of drugs, achieving excellent DDI prediction performance. While extant substructural frameworks effectively encode interactions from atom-level features, they overlook valuable chemical bond representations within molecular graphs. More critically, given the multifaceted nature of DDI prediction tasks involving both known and novel drug combinations, previous methods lack tailored strategies to address these distinct scenarios. The resulting lack of adaptability impedes further improvements to model performance. To tackle these challenges, we propose PEB-DDI, a DDI prediction learning framework with enhanced substructure extraction. First, the information of chemical bonds is integrated and synchronously updated with the atomic nodes. Then, different dual-view strategies are selected based on whether novel drugs are present in the prediction task. Particularly, we constructed Molecular fingerprint–Molecular graph view for transductive task, and Bipartite graph–Molecular graph view for inductive task. Rigorous evaluations on benchmark datasets underscore PEB-DDI's superior performance. Notably, on DrugBank, it achieves an outstanding accuracy rate of 98. 18% when predicting previously unknown interactions among approved drugs. Even when faced with novel drugs, PEB-DDI consistently exhibits outstanding generalization capabilities with an accuracy rate of 88. 06%, attributing to the proper migrating of molecular basic structure learning.

Details DOI

AAAI Conference 2023 Conference Paper

WIERT: Web Information Extraction via Render Tree

Zimeng Li
Bo Shao
Linjun Shou
Ming Gong
Gen Li
Daxin Jiang

Web information extraction (WIE) is a fundamental problem in web document understanding, with a significant impact on various applications. Visual information plays a crucial role in WIE tasks as the nodes containing relevant information are often visually distinct, such as being in a larger font size or having a brighter color, from the other nodes. However, rendering visual information of a web page can be computationally expensive. Previous works have mainly focused on the Document Object Model (DOM) tree, which lacks visual information. To efficiently exploit visual information, we propose leveraging the render tree, which combines the DOM tree and Cascading Style Sheets Object Model (CSSOM) tree, and contains not only content and layout information but also rich visual information at a little additional acquisition cost compared to the DOM tree. In this paper, we present WIERT, a method that effectively utilizes the render tree of a web page based on a pretrained language model. We evaluate WIERT on the Klarna product page dataset, a manually labeled dataset of renderable e-commerce web pages, demonstrating its effectiveness and robustness.

PDF Details DOI