paper-with-me

홈 › Papers

Tighter Truncated Rectangular Prism Approximation for RNN Robustness Verification

2025-11-12 · Xingqi Lin, Liangyu Chen, Min Wu, Min Zhang, Zhenbing Zeng arxiv

Robustness verification is a promising technique for rigorously proving Recurrent Neural Networks (RNNs) robustly. A key challenge is to over-approximate the nonlinear activation functions with linear constraints, which can transform the verification problem into an efficiently solvable linear programming problem. Existing methods over-approximate the nonlinear parts with linear bounding planes individually, which may cause significant over-estimation and lead to lower verification accuracy. In this paper, in order to tightly enclose the three-dimensional nonlinear surface generated by the Hadamard product, we propose a novel truncated rectangular prism formed by two linear relaxation planes and a refinement-driven method to minimize both its volume and surface area for tighter over-approximation. Based on this approximation, we implement a prototype DeepPrism for RNN robustness verification. The experimental results demonstrate that \emph{DeepPrism} has significant improvement compared with the state-of-the-art approaches in various tasks of image classification, speech recognition and sentiment analysis.

📄 PDF Abstract BibTeX arXiv:2511.11699

Code (0)

등록된 구현이 없습니다.

Tasks

Image ClassificationSpeech RecognitionSentiment Analysis

Similar Papers 제목 키워드 기반

Truncated Cauchy Non-negative Matrix Factorization

2019-06-02 · Naiyang Guan, Tongliang Liu, Yangmuzi Zhang, DaCheng Tao 외

Non-negative matrix factorization (NMF) minimizes the Euclidean distance between the data matrix and its low rank approximation, and it fails when applied to corrupted data because the loss function is sensitive to outli…

ClusteringImage Clustering

Approximation of the truncated Zeta distribution and Zipf's law

2015-11-04 · Maurizio Naldi

Zipf's law appears in many application areas but does not have a closed form expression, which may make its use cumbersome. Since it coincides with the truncated version of the Zeta distribution, in this paper we propose…

Form

Concatenated Matrix SVD: Compression Bounds, Incremental Approximation, and Error-Constrained Clustering

2026-01-12 · Maksym Shamrai arxiv

Large collections of matrices arise throughout modern machine learning, signal processing, and scientific computing, where they are commonly compressed by concatenation followed by truncated singular value decomposition …

Neural Network Compression

Tighter Bounds on the Information Bottleneck with Application to Deep Learning

2024-02-12 · Nir Weingarten, Zohar Yakhini, Moshe Butman, Ran Gilad-Bachrach

Deep Neural Nets (DNNs) learn latent representations induced by their downstream task, objective function, and other parameters. The quality of the learned representations impacts the DNN's generalization ability and the…

Adversarial RobustnessDeep Learning

Dual Formulation for Non-Rectangular Lp Robust Markov Decision Processes

2025-02-13 · Navdeep Kumar, Adarsh Gupta, Maxence Mohamed Elfatihi, Giorgia Ramponi 외

We study robust Markov decision processes (RMDPs) with non-rectangular uncertainty sets, which capture interdependencies across states unlike traditional rectangular models. While non-rectangular robust policy evaluation…