检索结果-内蒙古大学图书馆

A Closer Look at Benchmarking Self-Supervised Pre-training with Image Classification

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Marks, Markus Knott, Manuel Kondapaneni, Neehar Cole, Elijah Defraeye, Thijs Perez-Cruz, Fernando Perona, Pietro California Institute of Technology United States ETH Zurich Institute for Machine Learning Department of Computer Science Switzerland Swiss Data Science Center ETH Zurich and EPFL Switzerland Empa Swiss Federal Laboratories for Materials Science and Technology Switzerland Altos Labs Switzerland

Self-supervised learning (SSL) is a machine learning approach where the data itself provides supervision, eliminating the need for external labels. The model is forced to learn about the data's inherent structure or context by solving a pretext task. With SSL, models can learn from abundant and cheap unlabeled data, significantly reducing the cost of training models where labels are expensive or inaccessible. In Computer Vision, SSL is widely used as pre-training followed by a downstream task, such as supervised transfer, few-shot learning on smaller labeled data sets, and/or unsupervised clustering. Unfortunately, it is infeasible to evaluate SSL methods on all possible downstream tasks and objectively measure the quality of the learned representation. Instead, SSL methods are evaluated using in-domain evaluation protocols, such as fine-tuning, linear probing, and k-nearest neighbors (kNN). However, it is not well understood how well these evaluation protocols estimate the representation quality of a pre-trained model for different downstream tasks under different conditions, such as dataset, metric, and model architecture. In this work, we study how classification-based evaluation protocols for SSL correlate and how well they predict downstream performance on different dataset types. Our study includes eleven common image datasets and 26 models that were pre-trained with different SSL methods or have different model backbones. We find that in-domain linear/kNN probing protocols are, on average, the best general predictors for out-of-domain performance. We further investigate the importance of batch normalization for the various protocols and evaluate how robust correlations are for different kinds of dataset domain shifts. In addition, we challenge assumptions about the relationship between discriminative and generative self-supervised methods, finding that most of their performance differences can be explained by changes to model backbones. Copyright © 2024, The

关键词： Nearest neighbor search

Segment Together: A Versatile Paradigm for Semi-Supervised Medical Image Segmentation

学校读者我要写书评

暂无评论

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Zeng, Qingjie Xie, Yutong Lu, Zilin Lu, Mengkang Wu, Yicheng Xia, Yong Northwestern Polytechnical University National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Xi’an710072 China The University of Adelaide Australian Institute for Machine Learning AdelaideSA5000 Australia Monash University Faculty of Information Technology Department of Data Science and AI Australia

The scarcity of annotations has become a significant obstacle in training powerful deep-learning models for medical image segmentation, limiting their clinical application. To overcome this, semi-supervised learning that leverages abundant unlabeled data is highly desirable to enhance model training. However, most existing works still focus on specific medical tasks and underestimate the potential of learning across diverse tasks and datasets. In this paper, we propose a Versatile Semi-supervised framework (VerSemi) to present a new perspective that integrates various SSL tasks into a unified model with an extensive label space, exploiting more unlabeled data for semi-supervised medical image segmentation. Specifically, we introduce a dynamic task-prompted design to segment various targets from different datasets. Next, this unified model is used to identify the foreground regions from all labeled data, capturing cross-dataset semantics. Particularly, we create a synthetic task with a CutMix strategy to augment foreground targets within the expanded label space. To effectively utilize unlabeled data, we introduce a consistency constraint that aligns aggregated predictions from various tasks with those from the synthetic task, further guiding the model to accurately segment foreground regions during training. We evaluated our VerSemi framework against seven established SSL methods on four public benchmarking datasets. Our results suggest that VerSemi consistently outperforms all competing methods, beating the second-best method with a 2.69% average Dice gain on four datasets and setting a new state of the art for semi-supervised medical image segmentation. Code is available at https://***/maxwell0027/VerSemi. © 1982-2012 IEEE.

关键词： Self-supervised learning

Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Yenho Mudrik, Noga Johnsen, Kyle A. Alagapan, Sankaraleengam Charles, Adam S. Rozell, Christopher J. Machine Learning Center Georgia Institute of Technology United States School of Electrical and Computer Engineering Georgia Institute of Technology United States Coulter Dept. of Biomedical Engineering Emory University Georgia Institute of Technology United States Department of Biomedical Engineering Mathematical Institute for Data Science Center for Imaging Science Kavli Neuroscience Discovery Institute Johns Hopkins University United States

Time-varying linear state-space models are powerful tools for obtaining mathematically interpretable representations of neural signals. For example, switching and decomposed models describe complex systems using latent variables that evolve according to simple locally linear dynamics. However, existing methods for latent variable estimation are not robust to dynamical noise and system nonlinearity due to noise-sensitive inference procedures and limited model formulations. This can lead to inconsistent results on signals with similar dynamics, limiting the model’s ability to provide scientific insight. In this work, we address these limitations and propose a probabilistic approach to latent variable estimation in decomposed models that improves robustness against dynamical noise. Additionally, we introduce an extended latent dynamics model to improve robustness against system nonlinearities. We evaluate our approach on several synthetic dynamical systems, including an empirically-derived brain-computer interface experiment, and demonstrate more accurate latent variable inference in nonlinear systems with diverse noise conditions. Furthermore, we apply our method to a real-world clinical neurophysiology dataset, illustrating the ability to identify interpretable and coherent structure where previous models cannot. © 2024, CC BY-SA.

关键词： State space methods

Robust Autonomous Vehicle Pursuit without Expert Steering Labels

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Pan, Jiaxin Zhou, Changyao Gladkova, Mariia Khan, Qadeer Cremers, Daniel The Technical University of Munich TUM Germany Munich Center for Machine Learning MCML Germany The University of Oxford United Kingdom The Munich Data Science Institute Germany

In this work, we present a learning method for both lateral and longitudinal motion control of an ego-vehicle for the task of vehicle pursuit. The car being controlled does not have a pre-defined route, rather it reactively adapts to follow a target vehicle while maintaining a safety distance. To train our model, we do not rely on steering labels recorded from an expert driver, but effectively leverage a classical controller as an offline label generation tool. In addition, we account for the errors in the predicted control values, which can lead to a loss of tracking and catastrophic crashes of the controlled vehicle. To this end, we propose an effective data augmentation approach, which allows to train a network that is capable of handling different views of the target vehicle. During the pursuit, the target vehicle is firstly localized using a Convolutional Neural Network. The network takes a single RGB image along with cars’ velocities and estimates target vehicle’s pose with respect to the ego-vehicle. This information is then fed to a Multi-Layer Perceptron, which regresses the control commands for the ego-vehicle, namely throttle and steering angle. We extensively validate our approach using the CARLA simulator on a wide range of terrains. Our method demonstrates real-time performance, robustness to different scenarios including unseen trajectories and high route completion. Project page containing code and multimedia can be publicly accessed here: https://***/Autonomous-Vehicle-Pursuit/. Copyright © 2023, The Authors. All rights reserved.

关键词： Autonomous vehicles

How robust is randomized blind deconvolution via nuclear norm minimization against adversarial noise?

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Kostin, Julia Krahmer, Felix Stöger, Dominik Technical University of Munich Department of Mathematics Germany Munich Center for Machine Learning Germany Technical University of Munich Munich Data Science Institute Germany Germany

In this paper, we study the problem of recovering two unknown signals from their convolution, which is commonly referred to as blind deconvolution. Reformulation of blind deconvolution as a low-rank recovery problem has led to multiple theoretical recovery guarantees in the past decade due to the success of the nuclear norm minimization heuristic. In particular, in the absence of noise, exact recovery has been established for sufficiently incoherent signals contained in lower-dimensional subspaces. However, if the convolution is corrupted by additive bounded noise, the stability of the recovery problem remains much less understood. In particular, existing reconstruction bounds involve large dimension factors and therefore fail to explain the empirical evidence for dimension-independent robustness of nuclear norm minimization. Recently, theoretical evidence has emerged for ill-posed behavior of low-rank matrix recovery for sufficiently small noise levels. In this work, we develop improved recovery guarantees for blind deconvolution with adversarial noise which exhibit square-root scaling in the noise level. Hence, our results are consistent with existing counterexamples which speak against linear scaling in the noise level as demonstrated for related low-rank matrix recovery problems. Copyright © 2023, The Authors. All rights reserved.

关键词： Recovery

Ethical and methodological challenges in building morally informed AI systems

学校读者我要写书评

暂无评论

AI and Ethics 2022年第2期3卷 553-566页

作者： Hagendorff, Thilo Danks, David Cluster of Excellence “Machine Learning: New Perspectives for Science” University of Tuebingen Tuebingen Germany Halicioğlu Data Science Institute University of California San Diego USA

Recent progress in large language models has led to applications that can (at least) simulate possession of full moral agency due to their capacity to report context-sensitive moral assessments in open-domain conversations. However, automating moral decision-making faces several methodological as well as ethical challenges. They arise in the fields of bias mitigation, missing ground truth for moral “correctness”, effects of bounded ethicality in machines, changes in moral norms over time, risks of using morally informed AI systems as actual advice, as well as societal implications an increasing importance of algorithmic moral decision-making would have. This paper comments on all these challenges and provides critical considerations for future research on full artificial moral agency. Importantly, some of the adduced challenges can be met by more careful technology design, but others necessarily require engagement with core problems of meta-ethics.

关键词：

A Two-Stage Minimum Cost Multicut Approach to Self-supervised Multiple Person Tracking 15th

学校读者我要写书评

暂无评论

A Two-Stage Minimum Cost Multicut Approach to Self-supervise...

15th Asian Conference on Computer Vision, ACCV 2020

作者： Ho, Kalun Kardoost, Amirhossein Pfreundt, Franz-Josef Keuper, Janis Keuper, Margret Fraunhofer Center Machine Learning Sankt Augustin Germany CC-HPC Fraunhofer ITWM Kaiserslautern Germany Data and Web Science Group University of Mannheim Mannheim Germany Institute for Machine Learning and Analytics Offenburg University Offenburg Germany

ISBN: (纸本)9783030695316

Multiple Object Tracking (MOT) is a long-standing task in computer vision. Current approaches based on the tracking by detection paradigm either require some sort of domain knowledge or supervision to associate data correctly into tracks. In this work, we present a self-supervised multiple object tracking approach based on visual features and minimum cost lifted multicuts. Our method is based on straight-forward spatio-temporal cues that can be extracted from neighboring frames in an image sequences without supervision. Clustering based on these cues enables us to learn the required appearance invariances for the tracking task at hand and train an AutoEncoder to generate suitable latent representations. Thus, the resulting latent representations can serve as robust appearance cues for tracking even over large temporal distances where no reliable spatio-temporal features can be extracted. We show that, despite being trained without using the provided annotations, our model provides competitive results on the challenging MOT Benchmark for pedestrian tracking. © 2021, Springer Nature Switzerland AG.

关键词： Computer vision

Optimal bounds for ℓp sensitivity sampling via ℓ2 augmentation 24

学校读者我要写书评

暂无评论

Optimal bounds for ℓp sensitivity sampling via ℓ2 augmenta...

Proceedings of the 41st International Conference on machine learning

作者： Alexander Munteanu Simon Omlor Dortmund Data Science Center Faculties of Statistics and Computer Science TU Dortmund University Dortmund Germany Faculty of Statistics TU Dortmund University Dortmund Germany and Lamarr-Institute for Machine Learning and Artificial Intelligence Dortmund Germany

data subsampling is one of the most natural methods to approximate a massively large data set by a small representative proxy. In particular, sensitivity sampling received a lot of attention, which samples points proportional to an individual importance measure called sensitivity. This framework reduces in very general settings the size of data to roughly the VC dimension d times the total sensitivity S while providing strong (1±ε) guarantees on the quality of approximation. The recent work of Woodruff & Yasuda (2023c) improved substantially over the general $\tilde O(\varepsilon^{-2}\mathfrak Sd)$ bound for the important problem of ℓp subspace embeddings to $\tilde O(\varepsilon^{-2}\mathfrak S^{2/p})$ for p ∈ [1, 2]. Their result was subsumed by an earlier $\tilde O(\varepsilon^{-2}\mathfrak Sd^{1-p/2})$ bound which was implicitly given in the work of Chen & Derezinski (2021). We show that their result is tight when sampling according to plain ℓp sensitivities. We observe that by augmenting the ℓp sensitivities by ℓ2 sensitivities, we obtain better bounds improving over the aforementioned results to optimal linear $\tilde O(\varepsilon^{-2}(\mathfrak S+d)) = \tilde O(\varepsilon^{-2}d)$ sampling complexity for all p ∈ [1, 2]. In particular, this resolves an open question of Woodruff & Yasuda (2023c) in the affirmative for p ∈ [1, 2] and brings sensitivity subsampling into the regime that was previously only known to be possible using Lewis weights (Cohen & Peng, 2015). As an application of our main result, we also obtain an Õ(ε-2µd) sensitivity sampling bound for logistic regression, where µ is a natural complexity measure for this problem. This improves over the previous Õ (ε-2µ2d) bound of Mai et al. (2021) which was based on Lewis weights subsampling.

关键词：

The Sharpness Disparity Principle in Transformers for Accelerating Language Model Pre-Training

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Jinbo Wang, Mingze Zhou, Zhanpeng Yan, Junchi Weinan, E. Wu, Lei School of Mathematical Sciences Peking University China Center for Machine Learning Research Peking University China Sch. of Computer Science Sch. of Artificial Intelligence Shanghai Jiao Tong University China AI for Science Institute Beijing China

Transformers consist of diverse building blocks, such as embedding layers, normalization layers, self-attention mechanisms, and point-wise feedforward networks. Thus, understanding the differences and interactions among these blocks is important. In this paper, we uncover a clear sharpness disparity across these blocks, which emerges early in training and intriguingly persists throughout the training process. Motivated by this finding, we propose Blockwise learning Rate (LR), a strategy that tailors the LR to each block’s sharpness, accelerating large language model (LLM) pre-training. By integrating Blockwise LR into AdamW, we consistently achieve lower terminal loss and nearly 2× speedup compared to vanilla AdamW. We demonstrate this acceleration across GPT-2 and LLaMA, with model sizes ranging from 0.12B to 1.1B and datasets of OpenWebText and MiniPile. Finally, we incorporate Blockwise LR into Adam-mini (Zhang et al., 2024c), a recently proposed memory-efficient variant of Adam, achieving a combined 2× speedup and 2× memory saving. These results underscore the potential of exploiting the sharpness disparity to improve LLM training. Copyright © 2025, The Authors. All rights reserved.

关键词： Modeling languages