检索结果-内蒙古大学图书馆

Recent Advances in Out-of-Distribution Detection with CLIP-Like Models: A Survey

学校读者我要写书评

暂无评论

arXiv 2025年

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

Out-of-distribution detection (OOD) is a pivotal task for real-world applications that trains models to identify samples distributionally different from the in-distribution (ID) data during testing. Recent advances in AI, particularly Vision-Language Models (VLMs) like CLIP, have revolutionized OOD detection by shifting from traditional unimodal image detectors to multimodal image-text detectors. This shift has inspired extensive research, however, existing categorization scheme (e.g., few/zero-shot types) still rely solely on the availability of ID images, following a unimodal paradigm. To better align with CLIP’s cross-modal nature, we propose a new categorization scheme rooted in both image and text modalities. Specifically, we categorize existing methods guided by how the visual and textual information of OOD data is separately utilized in image + text modalities, further dividing them into four groups: {OOD Images (i.e., outliers) Seen/Unseen + OOD Texts (i.e., learnable vectors or class names) Known/Unknown} across two training strategies (e.g., train-free/required). More importantly, we further discuss the open problems of CLIP-like OOD detection, and highlight potential directions for future research, including cross-domain integration, practical applications, and theoretical understanding. Copyright © 2025, The Authors. All rights reserved.

关键词： Surveying

Dense Face Detection via High-level Context Mining

学校读者我要写书评

暂无评论

Dense Face Detection via High-level Context Mining

International Conference on Automatic Face and Gesture Recognition

作者： Qixiang Geng Dong Liang Huiyu Zhou Liyan Zhang Han Sun Ningzhong Liu MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Collaborative Innovation Center of Novel Software Technology and Industrialization College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics School of Informatics University of Leicester

ISBN: (纸本)9781665431774

The appearance degradation caused by low resolution is the core problem of small face detection. Therefore, a natural approach is to assemble information from the context. This paper focuses on how to use high-level contextual information to improve the abilities of anchor-based detectors to detect dense and degenerate faces. We tap the spatial contextual information on the overall view based on the density map, and propose the prior of face co-occurrence for inferred bounding-boxes coordination. We also propose score-size-specific non-maximum suppression to replace the traditional non-maximum suppression at the end of anchor-based detectors. According to the inferred face boxes' quantity, score and size, the proposed synthetical solution reduces false positives and increases true positives. Our method does not require additional training, which is model-independent and can be embedded into existing face detectors. We also propose a dataset - Crowd Face for face detection, which is full of challenges. We expect to supply enough samples to highlight the difficulties of detecting dense and degenerate faces. We embed our proposed methods into state-of-the-art face detectors on massively benchmarked face datasets. Compared with the prior art on the WIDER FACE hard set, our method increase an Average Precision of 0.1 %-1.3%. On Crowd Face, it increases an Average Precision of 1 % – 6%. Dataset is available on: https://***/QxGeng/Crowd-Face.

关键词： Training Degradation Face recognition Conferences Detectors Gesture recognition Benchmark testing

Generative locally linear embedding

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ghojogh, Benyamin Ghodsi, Ali Karray, Fakhri Crowley, Mark The Machine Learning Laboratory Department of Electrical and Computer Engineering University of Waterloo WaterlooON Canada The Department of Statistics and Actuarial Science University of Waterloo WaterlooON Canada The Centre for Pattern Analysis and Machine Intelligence Department of Electrical and Computer Engineering University of Waterloo WaterlooON Canada

Locally Linear Embedding (LLE) is a nonlinear spectral dimensionality reduction and manifold learning method. It has two main steps which are linear reconstruction and linear embedding of points in the input space and embedding space, respectively. In this work, we propose two novel generative versions of LLE, named Generative LLE (GLLE), whose linear reconstruction steps are stochastic rather than deterministic. GLLE assumes that every data point is caused by its linear reconstruction weights as latent factors. The proposed GLLE algorithms can generate various LLE embeddings stochastically while all the generated embeddings relate to the original LLE embedding. We propose two versions for stochastic linear reconstruction, one using expectation maximization and another with direct sampling from a derived distribution by optimization. The proposed GLLE methods are closely related to and inspired by variational inference, factor analysis, and probabilistic principal component analysis. Our simulations show that the proposed GLLE methods work effectively in unfolding and generating submanifolds of data. Copyright © 2021, The Authors. All rights reserved.

关键词： Embeddings

Better safe than sorry: Preventing delusive adversaries with adversarial training

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Tao, Lue Feng, Lei Yi, Jinfeng Huang, Sheng-Jun 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 College of Computer Science Chongqing University China JD AI Research China

Delusive attacks aim to substantially deteriorate the test accuracy of the learning model by slightly perturbing the features of correctly labeled training examples. By formalizing this malicious attack as finding the worst-case training data within a specific ∞-Wasserstein ball, we show that minimizing adversarial risk on the perturbed data is equivalent to optimizing an upper bound of natural risk on the original data. This implies that adversarial training can serve as a principled defense against delusive attacks. Thus, the test accuracy decreased by delusive attacks can be largely recovered by adversarial training. To further understand the internal mechanism of the defense, we disclose that adversarial training can resist the delusive perturbations by preventing the learner from overly relying on non-robust features in a natural setting. Finally, we complement our theoretical findings with a set of experiments on popular benchmark datasets, which show that the defense withstands six different practical attacks. Both theoretical and empirical results vote for adversarial training when confronted with delusive adversaries. Copyright © 2021, The Authors. All rights reserved.

关键词： Network security

AdaGDA: Faster Adaptive Gradient Descent Ascent Methods for Minimax Optimization

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, Feihu Wu, Xidong Hu, Zhengmian College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China Miit Key Laboratory of Pattern Analysis and Machine Intelligence China Department of Electrical and Computer Engineering University of Pittsburgh Pittsburgh United States

In the paper, we propose a class of faster adaptive Gradient Descent Ascent (GDA) methods for solving the nonconvex-strongly-concave minimax problems by using the unified adaptive matrices, which include almost all existing coordinate-wise and global adaptive learning rates. In particular, we provide an effective convergence analysis framework for our adaptive GDA methods. Specifically, we propose a fast Adaptive Gradient Descent Ascent (AdaGDA) method based on the basic momentum technique, which reaches a lower gradient complexity of Õ(κ4ϵ-4) for finding an ϵ-stationary point without large batches, which improves the existing results of the adaptive GDA methods by a factor of O(√κ). Moreover, we propose an accelerated version of AdaGDA (VR-AdaGDA) method based on the momentum-based variance reduced technique, which achieves a lower gradient complexity of Õ(κ 4.5ϵ-3) for finding an ϵ-stationary point without large batches, which improves the existing results of the adaptive GDA methods by a factor of O(ϵ-1). Moreover, we prove that our VR-AdaGDA method can reach the best known gradient complexity of Õ(κ3ϵ-3) with the minibatch size O(κ3). The experiments on policy evaluation and fair classifier learning tasks are conducted to verify the efficiency of our new algorithms. © 2021, CC BY-NC-SA.

关键词： Gradient methods

Learning twofold heterogeneous multi-task by sharing similar convolution kernel pairs

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Feng, Quan Chen, Songcan College of Computer Science & Technology Nanjing University of Aeronautics & Astronautics NanjingJiangsu211106 China MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing University of Aeronautics & Astronautics NanjingJiangsu211106 China

Heterogeneous multi-task learning (HMTL) is an important topic in multi-task learning (MTL). Most existing HMTL methods usually solve either scenario where all tasks reside in the same input (feature) space yet unnecessarily the consistent output (label) space or scenario where their input (feature) spaces are heterogeneous while the output (label) space is consistent. However, to the best of our knowledge, there is limited study on twofold heterogeneous MTL (THMTL) scenario where the input and the output spaces are both inconsistent or heterogeneous. In order to handle this complicated scenario, in this paper, we design a simple and effective multi-task adaptive learning (MTAL) network to learn multiple tasks in such THMTL setting. Specifically, we explore and utilize the inherent relationship between tasks for knowledge sharing from similar convolution kernels in individual layers of the MTAL network. Then in order to realize the sharing, we weightedly aggregate any pair of convolutional kernels with their similarity greater than some threshold ρ, consequently, our model effectively performs cross-task learning while suppresses the intra-redundancy of the entire network. Finally, we conduct end-to-end training. Our experimental results demonstrate the effectiveness of our method in comparison with the state-of-the-art counterparts. Copyright © 2021, The Authors. All rights reserved.

关键词： Linearization

BiAdam: Fast Adaptive Bilevel Optimization Methods

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, Feihu Li, Junyi Gao, Shangqian 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 Department of Electrical and Computer Engineering University of Pittsburgh Pittsburgh United States

Bilevel optimization recently has attracted increased interest in machine learning due to its many applications such as hyper-parameter optimization and meta learning. Although many bilevel methods recently have been proposed, these methods do not consider using adaptive learning rates. It is well known that adaptive learning rates can accelerate optimization algorithms. To fill this gap, in the paper, we propose a novel fast adaptive bilevel framework to solve stochastic bilevel optimization problems that the outer problem is possibly nonconvex and the inner problem is strongly convex. Our framework uses unified adaptive matrices including many types of adaptive learning rates, and can flexibly use the momentum and variance reduced techniques. In particular, we provide a useful convergence analysis framework for the bilevel optimization. Specifically, we propose a fast single-loop adaptive bilevel optimization (BiAdam) algorithm, which achieves a sample complexity of Õ(ϵ-4) for finding an ϵ-stationary solution. Meanwhile, we propose an accelerated version of BiAdam algorithm (VR-BiAdam), which reaches the best known sample complexity of Õ(ϵ-3). To the best of our knowledge, we first study the adaptive bilevel optimization methods with adaptive learning rates. Experimental results on data hyper-cleaning and hyper-representation learning tasks demonstrate the efficiency of our algorithms. © 2021, CC BY-NC-SA.

关键词： Stochastic systems

Quantile-quantile embedding for distribution transformation and manifold embedding with ability to choose the embedding distribution

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ghojogh, Benyamin Karray, Fakhri Crowley, Mark Machine Learning Laboratory Department of ECE University of Waterloo Canada Centre for Pattern Analysis and Machine Intelligence Department of ECE University of Waterloo Canada

We propose a new embedding method, named Quantile-Quantile Embedding (QQE), for distribution transformation and manifold embedding with the ability to choose the embedding distribution. QQE, which uses the concept of quantile-quantile plot from visual statistical tests, can transform the distribution of data to any theoretical desired distribution or empirical reference sample. Moreover, QQE gives the user a choice of embedding distribution in embedding the manifold of data into the low dimensional embedding space. It can also be used for modifying the embedding distribution of other dimensionality reduction methods, such as PCA, t-SNE, and deep metric learning, for better representation or visualization of data. We propose QQE in both unsupervised and supervised forms. QQE can also transform a distribution to either an exact reference distribution or its shape. We show that QQE allows for better discrimination of classes in some cases. Our experiments on different synthetic and image datasets show the effectiveness of the proposed embedding method. Copyright © 2020, The Authors. All rights reserved.

关键词： Embeddings

Learning from crowds with sparse and imbalanced annotations

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Shi, Ye Li, Shao-Yuan Huang, Sheng-Jun Ministry of Industry and Information Technology Key Laboratory of Pattern Analysis and Machine Intelligence College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing211106 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China

Traditional supervised learning requires ground truth labels for the training data, whose collection can be difficult in many cases. Recently, crowdsourcing has established itself as an efficient labeling solution through resorting to non-expert crowds. To reduce the labeling error effects, one common practice is to distribute each instance to multiple workers, whereas each worker only annotates a subset of data, resulting in the sparse annotation phenomenon. In this paper, we note that when meeting with class-imbalance, i.e., when the ground truth labels are class-imbalanced, the sparse annotations are prone to be skewly distributed, which thus can severely bias the learning algorithm. To combat this issue, we propose one self-training based approach named Self-Crowd by progressively adding confident pseudo-annotations and rebalancing the annotation distribution. Specifically, we propose one distribution aware confidence measure to select confident pseudo-annotations, which adopts the resampling strategy to oversample the minority annotations and undersample the majority annotations. On one real-world crowdsourcing image classification task, we show that the proposed method yields more balanced annotations throughout training than the distribution agnostic methods and substantially improves the learning performance at different annotation sparsity levels. Copyright © 2021, The Authors. All rights reserved.

关键词： Crowdsourcing