International Joint Conference on Artificial Intelligence最新文献_第9页

Character As Pixels: A Controllable Prompt Adversarial Attacking Framework for Black-Box Text Guided Image Generation Models 字符作为像素:黑箱文本引导图像生成模型的可控提示对抗攻击框架

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/109

Ziyi Kou, Shichao Pei, Yijun Tian, Xiangliang Zhang

{"title":"Character As Pixels: A Controllable Prompt Adversarial Attacking Framework for Black-Box Text Guided Image Generation Models","authors":"Ziyi Kou, Shichao Pei, Yijun Tian, Xiangliang Zhang","doi":"10.24963/ijcai.2023/109","DOIUrl":"https://doi.org/10.24963/ijcai.2023/109","url":null,"abstract":"In this paper, we study a controllable prompt adversarial attacking problem for text guided image generation (Text2Image) models in the black-box scenario, where the goal is to attack specific visual subjects (e.g., changing a brown dog to white) in a generated image by slightly, if not imperceptibly, perturbing the characters of the driven prompt (e.g., ``brown'' to ``br0wn''). Our study is motivated by the limitations of current Text2Image attacking approaches that still rely on manual trials to create adversarial prompts. To address such limitations, we develop CharGrad, a character-level gradient based attacking framework that replaces specific characters of a prompt with pixel-level similar ones by interactively learning the perturbation direction for the prompt and updating the attacking examiner for the generated image based on a novel proxy perturbation representation for characters. We evaluate CharGrad using the texts from two public image captioning datasets. Results demonstrate that CharGrad outperforms existing text adversarial attacking approaches on attacking various subjects of generated images by black-box Text2Image models in a more effective and efficient way with less perturbation on the characters of the prompts.","PeriodicalId":394530,"journal":{"name":"International Joint Conference on Artificial Intelligence","volume":"317 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115947210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Counting and Sampling Models in First-Order Logic 一阶逻辑中的计数和抽样模型

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/801

Ondřej Kuželka

引用次数: 0

Finding an ϵ-Close Minimal Variation of Parameters in Bayesian Networks 在贝叶斯网络中寻找ϵ-Close最小参数变化

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/635

Bahar Salmani, J. Katoen

引用次数: 0

What Lies beyond the Pareto Front? A Survey on Decision-Support Methods for Multi-Objective Optimization 帕累托前线之外是什么?多目标优化决策支持方法综述

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/755

Zuzanna Osika, J. Z. Salazar, Diederik M. Roijers, F. Oliehoek, P. Murukannaiah

引用次数: 0

Contrastive Learning and Reward Smoothing for Deep Portfolio Management 深度投资组合管理的对比学习与奖励平滑

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/441

Yun-Hsuan Lien, Yuan-kui Li, Yu-Shuen Wang

引用次数: 0

A New ANN-SNN Conversion Method with High Accuracy, Low Latency and Good Robustness 一种高精度、低时延、鲁棒性好的ANN-SNN转换新方法

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/342

Bingsen Wang, Jian Cao, Jue Chen, Shuo Feng, Yuan Wang

{"title":"A New ANN-SNN Conversion Method with High Accuracy, Low Latency and Good Robustness","authors":"Bingsen Wang, Jian Cao, Jue Chen, Shuo Feng, Yuan Wang","doi":"10.24963/ijcai.2023/342","DOIUrl":"https://doi.org/10.24963/ijcai.2023/342","url":null,"abstract":"Due to the advantages of low energy consumption, high robustness and fast inference speed, Spiking Neural Networks (SNNs), with good biological interpretability and the potential to be applied on neuromorphic hardware, are regarded as the third generation of Artificial Neural Networks (ANNs). Despite having so many advantages, the biggest challenge encountered by spiking neural networks is training difficulty caused by the non-differentiability of spike signals. ANN-SNN conversion is an effective method that solves the training difficulty by converting parameters in ANNs to those in SNNs through a specific algorithm. However, the ANN-SNN conversion method also suffers from accuracy degradation and long inference time. In this paper, we reanalyzed the relationship between Integrate-and-Fire (IF) neuron model and ReLU activation function, proposed a StepReLU activation function more suitable for SNNs under membrane potential encoding, and used it to train ANNs. Then we converted the ANNs to SNNs with extremely small conversion error and introduced leakage mechanism to the SNNs and get the final models, which have high accuracy, low latency and good robustness, and have achieved the state-of-the-art performance on various datasets such as CIFAR and ImageNet.","PeriodicalId":394530,"journal":{"name":"International Joint Conference on Artificial Intelligence","volume":"195 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122521538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generalization Bounds for Adversarial Metric Learning 对抗性度量学习的泛化界限

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/489

Wen Wen, Han Li, H. Chen, Rui Wu, Lingjuan Wu, Liangxuan Zhu

引用次数: 0

Learning to Binarize Continuous Features for Neuro-Rule Networks 学习神经规则网络的连续特征二值化

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/510

Wei Zhang, Y. Liu, Zhuo Wang, Jianyong Wang

{"title":"Learning to Binarize Continuous Features for Neuro-Rule Networks","authors":"Wei Zhang, Y. Liu, Zhuo Wang, Jianyong Wang","doi":"10.24963/ijcai.2023/510","DOIUrl":"https://doi.org/10.24963/ijcai.2023/510","url":null,"abstract":"Neuro-Rule Networks (NRNs) emerge as a promising neuro-symbolic method, enjoyed by the ability to equate fully-connected neural networks with logic rules. To support learning logic rules consisting of boolean variables, converting input features into binary representations is required. Different from discrete features that could be directly transformed by one-hot encodings, continuous features need to be binarized based on some numerical intervals. Existing studies usually select the bound values of intervals based on empirical strategies (e.g., equal-width interval). However, it is not optimal since the bounds are fixed and cannot be optimized to accommodate the ultimate training target. In this paper, we propose AutoInt, an approach that automatically binarizes continuous features and enables the intervals to be optimized with NRNs in an end-to-end fashion. Specifically, AutoInt automatically selects an interval for a given continuous feature in a soft manner to enable a differentiable learning procedure of interval-related parameters. Moreover, it introduces an additional soft K-means clustering loss to make the interval centres approach the original feature value distribution, thus reducing the risk of overfitting intervals. We conduct comprehensive experiments on public datasets and demonstrate the effectiveness of AutoInt in boosting the performance of NRNs.","PeriodicalId":394530,"journal":{"name":"International Joint Conference on Artificial Intelligence","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122723225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Vision Language Navigation with Knowledge-driven Environmental Dreamer 视觉语言导航与知识驱动的环境梦想家

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/204

Fengda Zhu, Vincent CS Lee, Xiaojun Chang, Xiaodan Liang

{"title":"Vision Language Navigation with Knowledge-driven Environmental Dreamer","authors":"Fengda Zhu, Vincent CS Lee, Xiaojun Chang, Xiaodan Liang","doi":"10.24963/ijcai.2023/204","DOIUrl":"https://doi.org/10.24963/ijcai.2023/204","url":null,"abstract":"Vision-language navigation (VLN) requires an agent to perceive visual observation in a house scene and navigate step-by-step following natural language instruction. Due to the high cost of data annotation and data collection, current VLN datasets provide limited instruction-trajectory data samples. Learning vision-language alignment for VLN from limited data is challenging since visual observation and language instruction are both complex and diverse. Previous works only generate augmented data based on original scenes while failing to generate data samples from unseen scenes, which limits the generalization ability of the navigation agent. In this paper, we introduce the Knowledge-driven Environmental Dreamer (KED), a method that leverages the knowledge of the embodied environment and generates unseen scenes for a navigation agent to learn. Generating an unseen environment with texture consistency and structure consistency is challenging. To address this problem, we incorporate three knowledge-driven regularization objectives into the KED and adopt a reweighting mechanism for self-adaptive optimization. Our KED method is able to generate unseen embodied environments without extra annotations. We use KED to successfully generate 270 houses and 500K instruction-trajectory pairs. The navigation agent with the KED method outperforms the state-of-the-art methods on various VLN benchmarks, such as R2R, R4R, and RxR. Both qualitative and quantitative experiments prove that our proposed KED method is able to high-quality augmentation data with texture consistency and structure consistency.","PeriodicalId":394530,"journal":{"name":"International Joint Conference on Artificial Intelligence","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122932507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Measuring a Priori Voting Power in Liquid Democracy 衡量流动民主中的先验投票权

International Joint Conference on Artificial Intelligence Pub Date : 2023-08-01 DOI: 10.24963/ijcai.2023/290

Rachael Colley, Théo Delemazure, Hugo Gilbert

引用次数: 0