检索结果-内蒙古大学图书馆

Adaptive Mirror Descent Bilevel Optimization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Huang, Feihu College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing China

In the paper, we propose a class of efficient adaptive bilevel methods based on mirror descent for nonconvex bilevel optimization, where its upper-level problem is nonconvex possibly with nonsmooth regularization, and its lower-level problem is also nonconvex while satisfies Polyak-Lojasiewicz (PL) condition. To solve these deterministic bilevel problems, we present an efficient adaptive projection-aid gradient (i.e., AdaPAG) method based on mirror descent, and prove that it obtains the best known gradient complexity of O(ǫ−1) for finding an ǫ-stationary solution of nonconvex bilevel problems. To solve these stochastic bilevel problems, we propose an efficient adaptive stochastic projection-aid gradient (i.e., AdaVSPAG) methods based on mirror descent and variance-reduced techniques, and prove that it obtains the best known gradient complexity of O(ǫ−3/2) for finding an ǫ-stationary solution. Since the PL condition relaxes the strongly convex, our algorithms can be used to nonconvex strongly-convex bilevel optimization. Theoretically, we provide a useful convergence analysis framework for our methods under some mild conditions, and prove that our methods have a fast convergence rate of O(1/T ), where T denotes the number of iterations. © 2023, CC BY-NC-SA.

关键词： Mirrors

Class-Aware Feature Perturbation for Long-Tailed Visual Recognition

学校读者我要写书评

暂无评论

Class-Aware Feature Perturbation for Long-Tailed Visual Reco...

IEEE International Conference on Multimedia and Expo (ICME)

作者： Xicheng Chen Haibo Ye Fangyu Zhou College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing China Collaborative Innovation Center of Novel Software Technology and Industrialization

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

The distribution of data in the real world is often imbalanced, with a small number of classes having a large number of instances, while most other classes have relatively few sample instances, resulting in a long-tailed distribution. Supervised contrastive learning still tends to favor classes with more samples, making it challenging to learn discriminative and robust features in imbalanced settings. In our work, we theoretically analyze the advantages of hard samples in contrastive learning and propose an end-to-end training framework based on class-aware perturbation. Specifically, we introduce different levels of perturbation based on class priors to construct hard samples, improve the disadvantaged position of tail classes, balance the feature space distribution, and dynamically adjust the classifier output with a combination of local batch priors and global priors. Experimental results show that our method achieves competitive performance across multiple long-tailed benchmark datasets.

关键词： Training Visualization Technological innovation Perturbation methods Contrastive learning Tail Benchmark testing

GBCI: Adaptive Frequency Band Learning for Gender Recognition in Brain-Computer Interfaces 1st

学校读者我要写书评

暂无评论

GBCI: Adaptive Frequency Band Learning for Gender Recogniti...

1st CAAI International Conference on Artificial intelligence, CICAI 2021

作者： Wang, Pengpai Zhou, Yueying Li, Zhongnian Zhang, Daoqiang College of Computer Science and Technology MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing University of Aeronautics and Astronautics Nanjing211106 China

ISBN: (纸本)9783030930455

In recent years, with the rapid development of brain-computer interface (BCI) technology, various applications based on BCI have generated significant interest. A motivating application of BCI is predicting human gender by electroencephalogram (EEG) analysis. However, most of the recent researches only identify the gender of EEG in the resting state. In real-life practice, it is difficult for the subjects to be at a full resting state when it mixes with motor imagery (MI). Therefore, we acquired two kinds of EEG activities data including 9 subjects (5 males and 4 females) with two state (i.e., resting state and MI) based on Chinese sign language. In this paper, to recognize gender form two state, an improved adaptive variational mode decomposition with long short term memory (AVMD-LSTM) network is developed to construct a hybrid state learning framework. Besides, the sample entropy (SE) is employed to select the channel of EEG data as the input of AVMD-LSTM. The recognition accuracy of gender classification was 89.21 %. Comparing with the task of complete resting state, the model is validated with more robustness. Furthermore, the proposed algorithm has many applications, including biometrics, healthcare, and online entertainment advertising. © 2021, Springer Nature Switzerland AG.

关键词： Variational mode decomposition

Filter, Obstruct and Dilute: Defending Against Backdoor Attacks on Semi-Supervised Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Xinrui Geng, Chuanxing Wan, Wenhai Li, Shao-Yuan Chen, Songcan College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China MIIT Key Laboratory of Pattern Analysis and Machine Intelligence China School of Computer Science and Technology Huazhong University of Science and Technology China

Recent studies have verified that semi-supervised learning (SSL) is vulnerable to data poisoning backdoor attacks. Even a tiny fraction of contaminated training data is sufficient for adversaries to manipulate up to 90% of the test outputs in existing SSL methods. Given the emerging threat of backdoor attacks designed for SSL, this work aims to protect SSL against such risks, marking it as one of the few known efforts in this area. Specifically, we begin by identifying that the spurious correlations between the backdoor triggers and the target class implanted by adversaries are the primary cause of manipulated model predictions during the test phase. To disrupt these correlations, we utilize three key techniques: Gaussian Filter, complementary learning and trigger mix-up, which collectively filter, obstruct and dilute the influence of backdoor attacks in both data pre-processing and feature learning. Experimental results demonstrate that our proposed method, Backdoor Invalidator (BI), significantly reduces the average attack success rate from 84.7% to 1.8% across different state-of-the-art backdoor attacks. It is also worth mentioning that BI does not sacrifice accuracy on clean data and is supported by a theoretical guarantee of its generalization capability. © 2025, CC BY.

关键词： Adversarial machine learning

Generative Visual Commonsense Answering and Explaining with Generative Scene Graph Constructing

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Yuan, Fan Fang, Xiaoyuan Quan, Rong Li, Jing Bi, Wei Xu, Xiaogang Li, Piji College of Artificial Intelligence Nanjing University of Aeronautics and Astronautics MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing211106 China College of Artificial Intelligence Nanjing University of Aeronautics and Astronautics Key Laboratory of Brain-Machine Intelligence Technology Ministry of Education Nanjing211106 China Department of Computing The Hong Kong Polytechnic University Hong Kong The Chinese University of Hong Kong Hong Kong

Visual Commonsense Reasoning, which is regarded as one challenging task to pursue advanced visual scene comprehension, has been used to diagnose the reasoning ability of AI systems. However, reliable reasoning requires a good grasp of the scene’s details. Existing work fails to effectively exploit the real-world object relationship information present within the scene, and instead overly relies on knowledge from training memory. Based on these observations, we propose a novel scene-graph-enhanced visual commonsense reasoning generation method named G2, which first utilizes the image patches and LLMs to construct a location-free scene graph, and then answer and explain based on the scene graph’s information. We also propose automatic scene graph filtering and selection strategies to absorb valuable scene graph information during training. Extensive experiments are conducted on the tasks and datasets of scene graph constructing and visual commonsense answering and explaining, respectively. Experimental results and ablation analysis demonstrate the effectiveness of our proposed framework. Copyright © 2025, The Authors. All rights reserved.

关键词： Question answering

Unlocking Better Closed-Set Alignment Based on Neural Collapse for Open-Set Recognition 39

学校读者我要写书评

暂无评论

Unlocking Better Closed-Set Alignment Based on Neural Collap...

39th Annual AAAI Conference on Artificial intelligence, AAAI 2025

作者： Li, Chaohua Zhang, Enhao Geng, Chuanxing Chen, Songcan College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China MIIT Key Laboratory of Pattern Analysis and Machine Intelligence China Department of Computer Science Hong Kong Baptist University Hong Kong

ISBN: (纸本)157735897X

In recent Open-set Recognition (OSR) community, a prevailing belief is that enhancing the discriminative boundaries of closed-set classes can improve the robustness of Deep Neural Networks (DNNs) against open data during testing. Typical studies validate this implicitly by empirical evidence, without a formalized understanding of how DNNs help the closed-set features obtain more discriminative boundaries? For this, we provide an answer from the Neural Collapse (NC) perspective: DNNs align the closed-set with a Simplex Equiangular Tight Frame (ETF) structure that has geometric and mathematical interpretability. Regrettably, although NC naturally occurs in DNNs, we discover that typical studies cannot guarantee the features being learned to strictly align with the ETF. Thus, we introduce a novel concept, Fixed ETF Template (FiT), which holds an ideal structure associated with closed-set classes. To force class means and classifier vectors to align with FiT, we further design a Dual ETF (DEF) loss involving two components. Specifically, F-DEF loss is designed to align class means with FiT strictly, yielding optimal inter-class separability. Meanwhile, we extend a dual form to classifier vectors, termed C-DEF loss, which guides class means and classifier vectors to satisfy self-duality. Our theoretical analysis proves the validity of the proposed approach, and extensive experiments demonstrate that DEF achieves comparable or superior results with reduced computational resources on standard OSR benchmarks. Copyright © 2025, Association for the Advancement of Artificial intelligence (***). All rights reserved.

关键词： Deep neural networks

Characteristic AI Agents via Large Language Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Xi Dai, Hongliang Gao, Shen Li, Piji College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing China School of Computer Science and Technology Shandong University China

The advancement of Large Language Models (LLMs) has led to significant enhancements in the performance of chatbot systems. Many researchers have dedicated their efforts to the development of bringing characteristics to chatbots. While there have been commercial products for developing role-driven chatbots using LLMs, it is worth noting that academic research in this area remains relatively scarce. Our research focuses on investigating the performance of LLMs in constructing Characteristic AI Agents by simulating real-life individuals across different settings. Current investigations have primarily focused on act on roles with simple profiles. In response to this research gap, we create a benchmark for the characteristic AI agents task, including dataset, techniques, and evaluation metrics. A dataset called "Character100" is built for this benchmark, comprising the most-visited people on Wikipedia for language models to role-play. With the constructed dataset, we conduct comprehensive assessment of LLMs across various settings. In addition, we devise a set of automatic metrics for quantitative performance evaluation. The experimental results underscore the potential directions for further improvement in the capabilities of LLMs in constructing characteristic AI agents. The benchmark is available at https://***/nuaa-nlp/Character100. © 2024, CC0.

关键词： Computational linguistics

PIE: Physics-inspired Low-light Enhancement

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liang, Dong Xu, Zhengyan Li, Ling Wei, Mingqiang Chen, Songcan MIIT Key Laboratory of Pattern Analysis and Machine Intelligence College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China Nanjing Universily of Aeronautics Astronautics Shenzhen Research Institute China

In this paper, we propose a physics-inspired contrastive learning paradigm for low-light enhancement, called PIE. PIE primarily addresses three issues: (i) To resolve the problem of existing learning-based methods often training a LLE model with strict pixel-correspondence image pairs, we eliminate the need for pixel-correspondence paired training data and instead train with unpaired images. (ii) To address the disregard for negative samples and the inadequacy of their generation in existing methods, we incorporate physics-inspired contrastive learning for LLE and design the Bag of Curves (BoC) method to generate more reasonable negative samples that closely adhere to the underlying physical imaging principle. (iii) To overcome the reliance on semantic ground truths in existing methods, we propose an unsupervised regional segmentation module, ensuring regional brightness consistency while eliminating the dependency on semantic ground truths. Overall, the proposed PIE can effectively learn from unpaired positive/negative samples and smoothly realize non-semantic regional enhancement, which is clearly different from existing LLE efforts. Besides the novel architecture of PIE, we explore the gain of PIE on downstream tasks such as semantic segmentation and face detection. Training on readily available open data and extensive experiments demonstrate that our method surpasses the state-of-the-art LLE models over six independent cross-scenes datasets. PIE runs fast with reasonable GFLOPs in test time, making it easy to use on mobile devices. Code available. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

Class-distribution-aware pseudo-labeling for semi-supervised multi-label learning 23

学校读者我要写书评

暂无评论

Class-distribution-aware pseudo-labeling for semi-supervised...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Ming-Kun Xie Jia-Hao Xiao Hao-Zhe Liu Gang Niu Masashi Sugiyama Sheng-Jun Huang Nanjing University of Aeronautics and Astronautics and MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing China RIKEN Center for Advanced Intelligence Project RIKEN Center for Advanced Intelligence Project and The University of Tokyo Tokyo Japan

Pseudo-labeling has emerged as a popular and effective approach for utilizing unlabeled data. However, in the context of semi-supervised multi-label learning (SSMLL), conventional pseudo-labeling methods encounter difficulties when dealing with instances associated with multiple labels and an unknown label count. These limitations often result in the introduction of false positive labels or the neglect of true positive ones. To overcome these challenges, this paper proposes a novel solution called Class-Aware Pseudo-Labeling (CAP) that performs pseudo-labeling in a class-aware manner. The proposed approach introduces a regularized learning framework incorporating class-aware thresholds, which effectively control the assignment of positive and negative pseudo-labels for each class. Notably, even with a small proportion of labeled examples, our observations demonstrate that the estimated class distribution serves as a reliable approximation. Motivated by this finding, we develop a class-distribution-aware thresholding strategy to ensure the alignment of pseudo-label distribution with the true distribution. The correctness of the estimated class distribution is theoretically verified, and a generalization error bound is provided for our proposed method. Extensive experiments on multiple benchmark datasets confirm the efficacy of CAP in addressing the challenges of SSMLL problems. The implementation is available at https://***/milkxie/SSMLL-CAP.

关键词：