检索结果-内蒙古大学图书馆

Towards the Identifiability in Noisy Label Learning: A Multinomial Mixture Approach

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Nguyen, Cuong Do, Thanh-Toan Carneiro, Gustavo School of Computer Science University of Adelaide Australia Department of Data Science and AI Monash University Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom

Learning from noisy labels (LNL) plays a crucial role in deep learning. The most promising LNL methods rely on identifying clean-label samples from a dataset with noisy annotations. Such an identification is challenging because the conventional LNL problem, which assumes a single noisy label per instance, is non-identifiable, i.e., clean labels cannot be estimated theoretically without additional heuristics. In this paper, we aim to formally investigate this identifiability issue using multinomial mixture models to determine the constraints that make the problem identifiable. Specifically, we discover that the LNL problem becomes identifiable if there are at least 2C − 1 noisy labels per instance, where C is the number of classes. To meet this requirement without relying on additional 2C − 2 manual annotations per instance, we propose a method that automatically generates additional noisy labels by estimating the noisy label distribution based on nearest neighbours. These additional noisy labels enable us to apply the Expectation-Maximisation algorithm to estimate the posterior probabilities of clean labels, which are then used to train the model of interest. We empirically demonstrate that our proposed method is capable of estimating clean labels without any heuristics in several label noise benchmarks, including synthetic, web-controlled, and real-world label noises. Furthermore, our method performs competitively with many state-of-the-art methods. © 2023, CC BY.

关键词： Deep learning

Task Weighting in Meta-learning with Trajectory Optimisation

学校读者我要写书评

暂无评论

arXiv 2023年

Developing meta-learning algorithms that are un-biased toward a subset of training tasks often requires hand-designed criteria to weight tasks, potentially resulting in sub-optimal solutions. In this paper, we introduce a new principled and fully-automated task-weighting algorithm for meta-learning methods. By considering the weights of tasks within the same mini-batch as an action, and the meta-parameter of interest as the system state, we cast the task-weighting meta-learning problem to a trajectory optimisation and employ the iterative linear quadratic regulator to determine the optimal action or weights of tasks. We theoretically show that the proposed algorithm converges to an ∊0-stationary point, and empirically demonstrate that the proposed approach out-performs common hand-engineering weighting methods in two few-shot learning benchmarks. © 2023, CC BY.

关键词： Learning algorithms

AEON: Adaptive Estimation of Instance-Dependent In-Distribution and Out-of-Distribution Label Noise for Robust Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Garg, Arpit Nguyen, Cuong Felix, Rafael Liu, Yuyuan Do, Thanh-Toan Carneiro, Gustavo Australian Institute for Machine Learning University of Adelaide Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom Department of Engineering Science University of Oxford United Kingdom Department of Data Science and AI Monash University Australia

Robust training with noisy labels is a critical challenge in image classification, offering the potential to reduce reliance on costly clean-label datasets. Real-world datasets often contain a mix of in-distribution (ID) and out-of-distribution (OOD) instance-dependent label noise, a challenge that is rarely addressed simultaneously by existing methods and is further compounded by the lack of comprehensive benchmarking datasets. Furthermore, even though current noisy-label learning approaches attempt to find noisy-label samples during training, these methods do not aim to estimate ID and OOD noise rates to promote their effectiveness in the selection of such noisy-label samples, and they are often represented by inefficient multi-stage learning algorithms. We propose the Adaptive Estimation of Instance-Dependent In-Distribution and Out-of-Distribution Label Noise (AEON) approach to address these research gaps. AEON is an efficient one-stage noisy-label learning methodology that dynamically estimates instance-dependent ID and OOD label noise rates to enhance robustness to complex noise settings. Additionally, we introduce a new benchmark reflecting real-world ID and OOD noise scenarios. Experiments demonstrate that AEON achieves state-of-the-art performance on both synthetic and real-world datasets*. © 2025, CC BY-NC-ND.

关键词： Contrastive Learning

SSFam: Scribble Supervised Salient Object Detection Family

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Zhengyi Deng, Sheng Wang, Xinrui Wang, Linbo Fang, Xianyong Tang, Bin Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University Hefei China School of Artificial Intelligence and Big Data Hefei University Hefei China

Scribble supervised salient object detection (SSSOD) constructs segmentation ability of attractive objects from surroundings under the supervision of sparse scribble labels. For the better segmentation, depth and thermal infrared modalities serve as the supplement to RGB images in the complex scenes. Existing methods specifically design various feature extraction and multi-modal fusion strategies for RGB, RGB-Depth, RGB-Thermal, and Visual-Depth-Thermal image input respectively, leading to similar model flood. As the recently proposed Segment Anything Model (SAM) possesses extraordinary segmentation and prompt interactive capability, we propose an SSSOD family based on SAM, named SSFam, for the combination input with different modalities. Firstly, different modal-aware modulators are designed to attain modal-specific knowledge which cooperates with modal-agnostic information extracted from the frozen SAM encoder for the better feature ensemble. Secondly, a siamese decoder is tailored to bridge the gap between the training with scribble prompt and the testing with no prompt for the stronger decoding ability. Our model demonstrates the remarkable performance among combinations of different modalities and refreshes the highest level of scribble supervised methods and comes close to the ones of fully supervised methods. https://***/liuzywen/SSFam Copyright © 2024, The Authors. All rights reserved.

关键词： Image segmentation

Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Pham, Cuong Nguyen, Cuong C. Le, Trung Phung, Dinh Carneiro, Gustavo Do, Thanh-Toan Department of Data Science and AI Monash University Australia Australian Institute for Machine Learning University of Adelaide Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom VinAI Viet Nam

Bayesian Neural Networks (BNNs) offer probability distributions for model parameters, enabling uncertainty quantification in predictions. However, they often underperform compared to deterministic neural networks. Utilizing mutual learning can effectively enhance the performance of peer BNNs. In this paper, we propose a novel approach to improve BNNs performance through deep mutual learning. The proposed approaches aim to increase diversity in both network parameter distributions and feature distributions, promoting peer networks to acquire distinct features that capture different characteristics of the input, which enhances the effectiveness of mutual learning. Experimental results demonstrate significant improvements in the classification accuracy, negative log-likelihood, and expected calibration error when compared to traditional mutual learning for BNNs. © 2024, CC BY.

关键词： Probability distributions

Set a Thief to Catch a Thief: Combating Label Noise through Noisy Meta Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Hanxuan Lu, Na Zhao, Xueying Yan, Yuxuan Ma, Kaipeng Keong, Kwoh Chee Carneiro, Gustavo Department of Automation Science and Engineering Xi’an Jiaotong University China School of Electrical Engineering Xi’an Jiaotong University China College of Computing and Data Science Nanyang Technological University Singapore Centre for Vision Speech and Signal Processing University of Surrey United Kingdom

Learning from noisy labels (LNL) aims to train high-performance deep models using noisy datasets. Meta learning based label correction methods have demonstrated remarkable performance in LNL by designing various meta label rectification tasks. However, extra clean validation set is a prerequisite for these methods to perform label correction, requiring extra labor and greatly limiting their practicality. To tackle this issue, we propose a novel noisy meta label correction framework STCT, which counterintuitively uses noisy data to correct label noise, borrowing the spirit in the saying "Set a Thief to Catch a Thief". The core idea of STCT is to leverage noisy data which is i.i.d. with the training data as a validation set to evaluate model performance and perform label correction in a meta learning framework, eliminating the need for extra clean data. By decoupling the complex bi-level optimization in meta learning into representation learning and label correction, STCT is solved through an alternating training strategy between noisy meta correction and semi-supervised representation learning. Extensive experiments on synthetic and real-world datasets demonstrate the outstanding performance of STCT, particularly in high noise rate scenarios. STCT achieves 96.9% label correction and 95.2% classification performance on CIFAR-10 with 80% symmetric noise, significantly surpassing the current state-of-the-art. © 2025, CC BY.

关键词：

Analyzing Cognitive Patterns in Gifted Children Using MRI and Morphometric Similarity Networks

学校读者我要写书评

暂无评论

Analyzing Cognitive Patterns in Gifted Children Using MRI an...

18th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2025

作者： Han, Shuning Duan, Feng Vilaseca, Gemma Vilaró, Núria Caiafa, Cesar F. Sun, Zhe Solé-Casals, Jordi Data and Signal Processing Research Group University of Vic-Central University of Catalonia Vic Catalonia Spain Image Processing Research Group RIKEN Center for Advanced Photonics RIKEN Saitama Wako-Shi Japan Tianjin Key Laboratory of Brain Science and Intelligent Rehabilitation Nankai University Tianjin China Psychological Department Oms and Prat school Fundació Catalunya- La Pedrera Catalonia Manresa Spain Oms Foundation Catalonia Manresa Spain Instituto Argentino de Radioastronomıa-CCT La Plata CONICET / CIC-PBA / UNLP Argentina Faculty of Health Data Science Juntendo University Chiba Urayasu Japan Department of Psychiatry University of Cambridge Cambridge United Kingdom

Advances in non-invasive neuroimaging, such as structural magnetic resonance imaging (sMRI), have en abled the construction of structural brain networks (SBNs), allowing in vivo mapping of anatomical connec tions. This study investigates brain network structural differences linked to different intelligence levels in children by individual morphometric similarity networks (MSNs) derived from sMRI data. Through group and individual-level analyses, we aim to uncover key topological features associated with cognitive perfor mance and to identify a suitable connection density for SBN analysis. Connection density strongly affects global and nodal topological features, with a range of p = 0.05 to 0.15 recommended for stable and optimal results. Gifted individuals exhibit stronger intra-hemispheric and intra-modular connectivity, a more balanced distribution of left-to-right intra-hemispheric connections, and lower mean versatility, supporting efficient and stable cognitive processing. Moreover, anatomical modularity analyses based on von Economo indicate that higher cognitive performance is linked to enhanced connectivity in specific areas (such as secondary sensory area, motor to association area and secondary sensory to limbic area), alongside selective reduction in cer tain modular connections (such as motor to insular area, association to secondary sensory area and motor to secondary sensory area). Furthermore, topological features, including participation coefficient and local effi ciency, are linked to cognitive performance. These findings provide valuable insights into the SBNs underlying cognitive levels in children. © 2025 by SCITEPRESS– science and Technology Publications, Lda.

关键词： Brain mapping

Model and feature diversity for bayesian neural networks in mutual learning 23

学校读者我要写书评

暂无评论

Model and feature diversity for bayesian neural networks in ...

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

作者： Cuong Pham Cuong C. Nguyen Trung Le Dinh Phung Gustavo Carneiro Thanh-Toan Do Department of Data Science and AI Monash University Australia Australian Institute for Machine Learning University of Adelaide Australia Department of Data Science and AI Monash University Australia and VinAI Vietnam Centre for Vision Speech and Signal Processing University of Surrey United Kingdom

关键词：

Instance-dependent Noisy-label Learning with Graphical Model Based Noise-rate Estimation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Garg, Arpit Nguyen, Cuong Felix, Rafael Do, Thanh-Toan Carneiro, Gustavo Australian Institute for Machine Learning University of Adelaide Australia Department of Data Science and AI Monash University Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom

Deep learning faces a formidable challenge when handling noisy labels, as models tend to overfit samples affected by label noise. This challenge is further compounded by the presence of instance-dependent noise (IDN), a realistic form of label noise arising from ambiguous sample information. To address IDN, Label Noise Learning (LNL) incorporates a sample selection stage to differentiate clean and noisy-label samples. This stage uses an arbitrary criterion and a pre-defined curriculum that initially selects most samples as noisy and gradually decreases this selection rate during training. Such curriculum is sub-optimal since it does not consider the actual label noise rate in the training set. This paper addresses this issue with a new noise-rate estimation method that is easily integrated with most state-of-the-art (SOTA) LNL methods to produce a more effective curriculum. Synthetic and real-world benchmarks’ results demonstrate that integrating our approach with SOTA LNL methods improves accuracy in most cases. © 2023, CC BY-NC-SA.

关键词：