Generalized Transfer Subspace Learning Through Low-Rank Constraint

It is expensive to obtain labeled real-world visual data for use in training of supervised algorithms. Therefore, it is valuable to leverage existing databases of labeled data. However, the data in the source databases is often obtained under conditions that differ from those in the new task. Transfer learning provides techniques for transferring learned knowledge from a source domain to a target domain by finding a mapping between them. In this paper, we discuss a method for projecting both source and target data to a generalized subspace where each target sample can be represented by some combination of source samples. By employing a low-rank constraint during this transfer, the structure of source and target domains are preserved. This approach has three benefits. First, good alignment between the domains is ensured through the use of only relevant data in some subspace of the source domain in reconstructing the data in the target domain. Second, the discriminative power of the source domain is naturally passed on to the target domain. Third, noisy information will be filtered out during knowledge transfer. Extensive experiments on synthetic data, and important computer vision problems such as face recognition application and visual domain adaptation for object recognition demonstrate the superiority of the proposed approach over the existing, well-established methods.     

基于自动弱标注数据的跨领域命名实体识别

近年来,在大规模标注语料上训练的神经网络模型大大提升了命名实体识别任务的性能.但是,新领域人工标注数据获取代价高昂,如何快速、低成本地进行领域迁移就显得非常重要.在目标领域仅给定无标注数据的情况下,该文尝试自动构建目标领域的弱标注语料并对其建模.首先,采用两种不同的方法对无标注数据进行自动标注;然后,采用留"同"去"异...     

Toward few-shot domain adaptation with perturbation-invariant representation and transferable prototypes

Domain adaptation (DA) for semantic segmentation aims to reduce the annotation burden for the dense pixel-level prediction task. It focuses on tackling the domain gap problem and manages to transfer knowledge learned from abundant source data to new target scenes. Although recent works have achieved rapid progress in this field, they still underperform fully supervised models with a large margin due to the absence of any available hints in the target domain. Considering that few-shot labels are cheap to obtain in practical applications, we attempt to leverage them to mitigate the performance gap between DA and fully supervised methods. The key to this problem is to leverage the few-shot labels to learn robust domain-invariant predictions effectively. To this end, we first design a data perturbation strategy to enhance the robustness of the representations. Furthermore, a transferable prototype module is proposed to bridge the domain gap based on the source data and few-shot targets. By means of these proposed methods, our approach can perform on par with the fully supervised models to some extent. We conduct extensive experiments to demonstrate the effectiveness of the proposed methods and report the state-of-the-art performance on two popular DA tasks, i.e., from GTA5 to Cityscapes and SYNTHIA to Cityscapes.     

Transferring knowledge across robots: A risk sensitive approach

One of the most impressive characteristics of human perception is its domain adaptation capability. Humans can recognize objects and places simply by transferring knowledge from their past experience. Inspired by that, current research in robotics is addressing a great challenge: building robots able to sense and interpret the surrounding world by reusing information previously collected, gathered by other robots or obtained from the web. But, how can a robot automatically understand what is useful among a large amount of information and perform knowledge transfer? In this paper we address the domain adaptation problem in the context of visual place recognition. We consider the scenario where a robot equipped with a monocular camera explores a new environment. In this situation traditional approaches based on supervised learning perform poorly, as no annotated data are provided in the new environment and the models learned from data collected in other places are inappropriate due to the large variability of visual information. To overcome these problems we introduce a novel transfer learning approach. With our algorithm the robot is given only some training data (annotated images collected in different environments by other robots) and is able to decide whether, and how much, this knowledge is useful in the current scenario. At the base of our approach there is a transfer risk measure which quantifies the similarity between the given and the new visual data. To improve the performance, we also extend our framework to take into account multiple visual cues. Our experiments on three publicly available datasets demonstrate the effectiveness of the proposed approach.     

译文语序的领域性思考:一种融合主题信息的领域自适应调序模型

领域自适应研究的目标是建立一种动态调整翻译模型,使翻译模型对目标领域的语言特征具备较强的学习和处理能力,借以保证翻译系统在不同领域获得平衡可靠的翻译能力。现有翻译模型的自适应研究已经取得显著进展,但调序过程的领域适应性研究相对较少。在该文前期工作中通过对大规模源语言和目标语言的真实互译样本统计发现,在语义等价的短语级互译对子中,36.17%的样本在不同领域中的语序存在显著差异。针对这一问题,该文从主题角度出发,探索不同主题分布下的短语调序差异,提出一种融合主题信息的领域自适应调序模型。实验结果显示,嵌入调序适应性模型的翻译系统取得了较为明显的性能优势。     

Learning Kernels for Unsupervised Domain Adaptation with Applications to Visual Object Recognition

Domain adaptation aims to correct the mismatch in statistical properties between the source domain on which a classifier is trained and the target domain to which the classifier is to be applied. In this paper, we address the challenging scenario of unsupervised domain adaptation, where the target domain does not provide any annotated data to assist in adapting the classifier. Our strategy is to learn robust features which are resilient to the mismatch across domains and then use them to construct classifiers that will perform well on the target domain. To this end, we propose novel kernel learning approaches to infer such features for adaptation. Concretely, we explore two closely related directions. In the first direction, we propose unsupervised learning of a geodesic flow kernel (GFK). The GFK summarizes the inner products in an infinite sequence of feature subspaces that smoothly interpolates between the source and target domains. In the second direction, we propose supervised learning of a kernel that discriminatively combines multiple base GFKs. Those base kernels model the source and the target domains at fine-grained granularities. In particular, each base kernel pivots on a different set of landmarks—the most useful data instances that reveal the similarity between the source and the target domains, thus bridging them to achieve adaptation. Our approaches are computationally convenient, automatically infer important hyper-parameters, and are capable of learning features and classifiers discriminatively without demanding labeled data from the target domain. In extensive empirical studies on standard benchmark recognition datasets, our appraches yield state-of-the-art results compared to a variety of competing methods.     

通过域适应实现人脸识别

当源域和目的域数据分布不同时,大多数机器学习方法的性能会降低。为了解决这一问题,基于域适应的思想,提出了一种新的人脸识别方法。首先计算源域样本的相对权值,删除与目的域样本相差很大的样本,降低两域之间的差异性。然后采用基于正规化的Bregman Divergence获得公共子空间,获得两域之间的共性。最后利用目的域样本目标化源域样本,充分利用目的域的特有信息。在此基础上建立的分类模型能够充分利用两域之间的共性和目的域的特性,实现对目的域的准确分类。为了评估本方法的性能,在多个数据集上测试实验。实验结果证明,该方法的性能与其他几种方法相比均有所提高。     

加权解耦语义表达的多源领域自适应方法

近年来,深度学习受到越来越多研究者的重视并成功应用于许多领域.虽然深度学习在这些领域获得了巨大的成功,但是数据采集和标注成本高,严重限制了深度学习的推广应用.迁移学习不仅可以打破训练集数据和测试集数据独立同分布的假设,而且可以利用有标签的迁移源数据和没有标签的迁移目标数据训练得到具有良好泛化能力的模型,是扩展深度学习应用场景的重要研究方向.在众多的迁移学习方法中,多源领域自适应方法可以充分利用多个迁移源的信息,具有重要的实际价值.从数据的因果生成机制出发,假设观测数据由语义隐变量和领域隐变量这两组独立的隐变量同时生成.基于上述假设,提出了一种基于多种距离度量框架和加权解耦语义表达的多源领域自适应方法.该方法利用了双重对抗网络来提取解耦的语义信息和领域信息;另一方面,采用了3种不同的语义信息聚合策略获得领域不变的语义表达;最后使用领域不变的语义表达进行图片分类.在多个多源领域自适应数据上的对比及鲁棒性分析实验中,充分地验证了所提出方法的有效性.     

Cluster-based dynamic variance adaptation for interconnecting speech enhancement pre-processor and speech recognizer

A conventional approach to noise robust speech recognition consists of employing a speech enhancement pre-processor prior to recognition. However, such a pre-processor usually introduces artifacts that limit recognition performance improvement. In this paper we discuss a framework for improving the interconnection between speech enhancement pre-processors and a recognizer. The framework relies on recent proposals for increasing robustness by replacing the point estimate of the enhanced features with a distribution with a dynamic (i.e. time varying) feature variance. We have recently proposed a model for the dynamic feature variance consisting of a dynamic feature variance root obtained from the pre-processor, which is multiplied by a weight representing the pre-processor uncertainty, and that uses adaptation data to optimize the pre-processor uncertainty weight. The formulation of the method is general and could be used with any speech enhancement pre-processor. However, we observed that in case of noise reduction based on spectral subtraction or related approaches, adaptation could fail because the proposed model is weak at representing well the actual dynamic feature variance. The dynamic feature variance changes according to the level of speech sound, which varies with the HMM states. Therefore, we propose improving the model by introducing HMM state dependency. We achieve this by using a cluster-based representation, i.e. the Gaussians of the acoustic model are grouped into clusters and a different pre-processor uncertainty weight is associated with each cluster. Experiments with various pre-processors and recognition tasks prove the generality of the proposed integration scheme and show that the proposed extension improves the performance with various speech enhancement pre-processors.     

Multi-source unsupervised domain adaptation for object detection

Domain adaptation for object detection has been extensively studied in recent years. Most existing approaches focus on single-source unsupervised domain adaptive object detection. However, a more practical scenario is that the labeled source data is collected from multiple domains with different feature distributions. The conventional approaches do not work very well since multiple domain gaps exist. We propose a Multi-source domain Knowledge Transfer (MKT) method to handle this situation. First, the low-level features from multiple domains are aligned by learning a shallow feature extraction network. Then, the high-level features from each pair of source and target domains are aligned by the followed multi-branch network. After that, we perform two parts of information fusion: (1) We train a detection network shared by all branches based on the transferability of each source sample feature. The transferability of a source sample feature means the indistinguishable degree to the target domain sample features. (2) For using our model, the target sample features output by the multi-branch network are fused based on the average transferability of each domain. Moreover, we leverage both image-level and instance-level attention to promote positive cross-domain transfer and suppress negative transfer. Our main contributions are the two-stage feature alignments and information fusion. Extensive experimental results on various transfer scenarios show that our method achieves the state-of-the-art performance.     

Finding and removing Clever Hans: Using explanation methods to debug and improve deep models

Contemporary learning models for computer vision are typically trained on very large (benchmark) datasets with millions of samples. These may, however, contain biases, artifacts, or errors that have gone unnoticed and are exploitable by the model. In the worst case, the trained model does not learn a valid and generalizable strategy to solve the problem it was trained for, and becomes a “Clever Hans” predictor that bases its decisions on spurious correlations in the training data, potentially yielding an unrepresentative or unfair, and possibly even hazardous predictor. In this paper, we contribute by providing a comprehensive analysis framework based on a scalable statistical analysis of attributions from explanation methods for large data corpora. Based on a recent technique — Spectral Relevance Analysis — we propose the following technical contributions and resulting findings: (a) a scalable quantification of artifactual and poisoned classes where the machine learning models under study exhibit Clever Hans behavior, (b) several approaches we collectively denote as Class Artifact Compensation, which are able to effectively and significantly reduce a model’s Clever Hans behavior, i.e., we are able to un-Hans models trained on (poisoned) datasets, such as the popular ImageNet data corpus. We demonstrate that Class Artifact Compensation, defined in a simple theoretical framework, may be implemented as part of a neural network’s training or fine-tuning process, or in a post-hoc manner by injecting additional layers, preventing any further propagation of undesired Clever Hans features, into the network architecture. Using our proposed methods, we provide qualitative and quantitative analyses of the biases and artifacts in, e.g., the ImageNet dataset, the Adience benchmark dataset of unfiltered faces, and the ISIC 2019 skin lesion analysis dataset. We demonstrate that these insights can give rise to improved, more representative, and fairer models operating on implicitly cleaned data corpora.     

核分布一致局部领域适应学习

针对领域适应学习(Domain adaptation learning, DAL)问题,提出一种核分布一致局部领域适应学习机(Kernel distribution consistency based local domain adaptation classifier, KDC-LDAC),在某个通用再生核Hilbert空间(Universally reproduced kernel Hilbert space, URKHS),基于结构风险最小化模型, KDC-LDAC首先学习一个核分布一致正则化支持向量机(Support vector machine, SVM),对目标数据进行初始划分; 然后,基于核局部学习思想,对目标数据类别信息进行局部回归重构; 最后,利用学习获得的类别信息,在目标领域训练学习一个适于目标判别的分类器.人 造和实际数据集实验结果显示,所提方法具有优化或可比较的领域适应学习性能.     

伪标签不确定性估计的源域无关鲁棒域自适应

无监督域自适应是解决训练集(源域)和测试集(目标域)分布不一致的有效途径之一.现有的无监督域自适应的理论和方法在相对封闭、静态的环境下取得了一定成功,但面向开放动态任务环境时,在隐私保护、数据孤岛等限制条件下,源域数据往往不可直接获取,现有无监督域自适应方法的鲁棒性将面临严峻的挑战.鉴于此,研究了一个更具挑战性却又未被...     

多层校正的无监督领域自适应方法

目的 目前深度神经网络已成功应用于众多机器学习任务,并展现出惊人的性能提升效果。然而传统的深度网络和机器学习算法都假定训练数据和测试数据服从的是同一分布,而这种假设在实际应用中往往是不成立的。如果训练数据和测试数据的分布差异很大,那么由传统机器学习算法训练出来的分类器的性能将会大大降低。为了解决此类问题,提出了一种基于多层校正的无监督领域自适应方法。方法 首先利用多层校正来调整现有的深度网络,利用加法叠加来完美对齐源域和目标域的数据表示;然后采用多层权值最大均值差异来适应目标域,增加网络的表示能力;最后提取学习获得的域不变特征来进行分类,得到目标图像的识别效果。结果 本文算法在Office-31图像数据集等4个数字数据集上分别进行了测试实验,以对比不同算法在图像识别和分类方面的性能差异,并进行准确度测量。测试结果显示,与同领域算法相比,本文算法在准确率上至少提高了5%,在应对照明变化、复杂背景和图像质量不佳等干扰情况时,亦能获得较好的分类效果,体现出更强的鲁棒性。结论 在领域自适应相关数据集上的实验结果表明,本文方法具备一定的泛化能力,可以实现较高的分类性能,并且优于其他现有的无监督领域自适应方法。     

关联子域对齐网络的跨域高光谱图像分类

目的 近年来,深度网络成功应用于高光谱图像分类。然而,难以获取充足的标记数据大大限制了深度网络的充分训练,进而导致网络对高光谱图像的分类能力下降。为解决以上困难,提出一种关联子域对齐网络的高光谱图像迁移分类方法。方法 基于深度迁移学习方法,通过对两域分布进行多角度、全面领域适应的同时将两域分类器进行差异适配。一方面,利用关联对齐从整体上对齐了两域的二阶统计量信息,适配了两域的全局分布;另一方面,利用局部最大均值差异对齐了相关子域的一阶统计量信息,适配了两域的局部分布。另外,构造一种分类器适配模块并将其加入所提网络中,通过对两域分类器差异进行适配,进一步增强网络的领域适应效果。结果 从4组真实高光谱数据集上的实验结果可看出：在分别采集于不同区域的高光谱图像数据对上,所提方法的精度比排名第2的分类方法高出1.01%、0.42%、0.73%和0.64%。本文方法的Kappa系数也取得最优结果。结论 与现有主流算法相比较,所提网络能够在整体和局部、一阶和二阶统计量上分别对两域进行有效对齐,进而充分利用在源域上训练好的分类器完成对目标域高光谱数据的跨域分类。     

Using deep learning and social network analysis to understand and manage extreme flooding

Combining machine learning with social network analysis (SNA) can leverage vast amounts of social media data to better respond to crises. We present a case study using Twitter data from the March 2019 Nebraska floods in the United States, which caused over $1 billion in damage in the state and widespread evacuations of residents. We use a subset of machine learning, deep learning (DL), to classify text content of 11,982 tweets, and we integrate that with SNA to understand the structure of tweet interactions. Our DL approach pre‐trains our model with a DL language technique, BERT, and then trains the model using the standard training dataset to sort a dataset of tweets into classes tailored to crisis events. Several performance measures demonstrate that our two‐tiered trained model improves domain adaptation and generalization across different extreme weather event types. This approach identifies the role of Twitter during the damage containment stage of the flood. Our SNA identifies accounts that function as primary sources of information on Twitter. Together, these two approaches help crisis managers filter large volumes of data and overcome challenges faced by simple statistical models and other computational techniques to provide useful information during crises like flooding.     

A survey of multi-source domain adaptation

In many machine learning algorithms, a major assumption is that the training and the test samples are in the same feature space and have the same distribution. However, for many real applications this assumption does not hold. In this paper, we survey the problem where the training samples and the test samples are from different distributions. This problem can be referred as domain adaptation. The training samples, always with labels, are obtained from what is called source domains, while the test samples, which usually have no labels or only a few labels, are obtained from what is called target domains. The source domains and the target domains are different but related to some extent; the learners can learn some information from the source domains for the learning of the target domains. We focus on the multi-source domain adaptation problem where there is more than one source domain available together with only one target domain. A key issue is how to select good sources and samples for the adaptation. In this survey, we review some theoretical results and well developed algorithms for the multi-source domain adaptation problem. We also discuss some open problems which can be explored in future work.     

基于弱标注数据的汉语分词领域移植

近年来,基于神经网络的分词模型在封闭领域文本上取得了很高的性能。然而,在领域移植场景下,即测试数据与训练数据的领域差异较大时,分词的性能会显著下降。该文尝试利用自动获取的弱标注数据来提升领域移植场景下的分词性能。首先,对目前性能最好的BiLSTM-CRF分词模型进行扩展,引入适用于弱标注数据的损失函数;进而提出一种简单有效的数据筛选方法,从海量弱标注数据中筛选和目前领域更相关的数据;最后,该文发现数据预处理和在神经网络中引入传统特征均可以有效提高分词性能。在SIGHAN Bakeoff 2010和ZhuXian标注测试集上的实验结果表明,该文所提方法可有效提升汉语分词领域移植性能,平均F值提高了3.6%。     

基于改进残差卷积自编码网络的类自适应旋转机械故障诊断

针对旋转机械传感器信号样本有限影响深层网络模型训练学习的问题,提出一种结合改进残差卷积自编码网络与类自适应方法的故障诊断模型应对小样本数据。首先将少量已标记的源域数据和目标域数据创建为成对样本,并设计一种改进的一维残差卷积自编码网络对两种不同分布的原始振动信号进行特征提取;其次,利用最大均值差异（MMD）减小分布差异,并将两个域同一故障类别的数据空间映射到一个共同的特征空间,最终实现准确的故障诊断。实验结果表明,与微调、域自适应等方法相比,所提模型能够有效提高不同工况、微量已标记的目标域振动数据下的故障诊断准确率。