paper-with-me

Papers

STR-Cert: Robustness Certification for Deep Text Recognition on Deep Learning Pipelines and Vision Transformers

2023-11-28 · Daqian Shao, Lukas Fesser, Marta Kwiatkowska

Robustness certification, which aims to formally certify the predictions of neural networks against adversarial inputs, has become an integral part of important tool for safety-critical applications. Despite considerable progress, existing certification methods are limited to elementary architectures, such as convolutional networks, recurrent networks and recently Transformers, on benchmark datasets such as MNIST. In this paper, we focus on the robustness certification of scene text recognition (STR), which is a complex and extensively deployed image-based sequence prediction problem. We tackle three types of STR model architectures, including the standard STR pipelines and the Vision Transformer. We propose STR-Cert, the first certification method for STR models, by significantly extending the DeepPoly polyhedral verification framework via deriving novel polyhedral bounds and algorithms for key STR model components. Finally, we certify and compare STR models on six datasets, demonstrating the efficiency and scalability of robustness certification, particularly for the Vision Transformer.

📄 PDF Abstract BibTeX arXiv:2401.05338

Code (0)

등록된 구현이 없습니다.

Tasks

Scene Text Recognition

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Certification of Speaker Recognition Models to Additive Perturbations

2024-04-29 · Dmitrii Korzh, Elvir Karimov, Mikhail Pautov, Oleg Y. Rogov 외

Speaker recognition technology is applied to various tasks, from personal virtual assistants to secure access systems. However, the robustness of these systems against adversarial attacks, particularly to additive pertur…

Few-Shot LearningSpeaker Recognition

UniT: A Unified Look at Certified Robust Training against Text Adversarial Perturbation

2023-09-21 · NeurIPS 2023 11

Recent years have witnessed a surge of certified robust training pipelines against text adversarial perturbation constructed by synonym substitutions. Given a base model, existing pipelines provide prediction certificate…

Incremental Randomized Smoothing Certification

2023-05-31 · Shubham Ugare, Tarun Suresh, Debangshu Banerjee, Gagandeep Singh 외

Randomized smoothing-based certification is an effective approach for obtaining robustness certificates of deep neural networks (DNNs) against adversarial attacks. This method constructs a smoothed DNN model and certifie…

Et Tu Certifications: Robustness Certificates Yield Better Adversarial Examples

2023-02-09 · Andrew C. Cullen, Shijie Liu, Paul Montague, Sarah M. Erfani 외

In guaranteeing the absence of adversarial examples in an instance's neighbourhood, certification mechanisms play an important role in demonstrating neural net robustness. In this paper, we ask if these certifications ca…

A Tool for Neural Network Global Robustness Certification and Training

2022-08-15 · Zhilu Wang, YiXuan Wang, Feisi Fu, Ruochen Jiao 외

With the increment of interest in leveraging machine learning technology in safety-critical systems, the robustness of neural networks under external disturbance receives more and more concerns. Global robustness is a ro…

GPU