Improving meta-learning model via meta-contrastive loss

Recently, addressing the few-shot learning issue with meta-learning framework achieves great success. As we know, regularization is a powerful technique and widely used to improve machine learning algorithms. However, rare research focuses on designing appropriate meta-regularizations to further improve the generalization of meta-learning models in few-shot learning. In this paper, we propose a novel meta-contrastive loss that can be regarded as a regularization to fill this gap. The motivation of our method depends on the thought that the limited data in few-shot learning is just a small part of data sampled from the whole data distribution, and could lead to various bias representations of the whole data because of the different sampling parts. Thus, the models trained by a few training data (support set) and test data (query set) might misalign in the model space, making the model learned on the support set can not generalize well on the query data. The proposed meta-contrastive loss is designed to align the models of support and query sets to overcome this problem. The performance of the meta-learning model in few-shot learning can be improved. Extensive experiments demonstrate that our method can improve the performance of different gradient-based meta-learning models in various learning problems, e.g., few-shot regression and classification.     

面向深度网络的小样本学习综述

深度学习以数据为驱动,被广泛应用于各个领域,但由于数据隐私、标记昂贵等导致样本少、数据不完备性等问题,同时小样本难于准确地表示数据分布,使得分类模型误差较大,且泛化能力差。为此,小样本学习被提出,旨在利用较少目标数据训练模型快速学习的能力。系统梳理了近几年来小样本学习领域的相关工作,主要整理和总结了基于数据增强、基于元学习和基于转导图小样本学习方法的研究进展。首先,从基于监督增强和基于无监督增强阐述数据增强的主要特点。其次,从基于度量学习和基于参数优化两方面对基于元学习的方法进行分析。接着,详细总结转导图小样本学习方法,介绍常用的小样本数据集,并通过实验阐述分析具有代表性的小样本学习模型。最后总结现有方法的局限性,并对小样本学习的未来研究方向进行展望。     

Augmented domain agreement for adaptable Meta-Learner on Few-Shot classification

Meta-learning approaches are recently introduced to obtain a proficient model in the problem of few-shot learning. The existing state-of-the-art methods can resolve the training difficulty to achieve the model generalization with limited instances by incorporating several tasks containing various drawn classes. However, during the fast-adaptation stage, the characteristics of instances in meta-training and meta-test sets are assumed to be similar. In contrast, most meta-learning algorithms are implemented in challenging settings where those are drawn from different populations. This critical assumption can cause the model to exhibit degraded performance. We propose an Augmented Domain Agreement for Adaptable Meta-Learner (AD2AML), which augments the domain adaptation framework in meta-learning to overcome this problem. We minimize the latent representation divergence of the inputs drawn from different distributions to enhance the model at obtaining more general features. Therefore, the trained network can be more transferable at shifted domain conditions. Furthermore, we extend our main idea by augmenting the image reconstruction network with cross-consistency loss to encourage the shared network to extract a similar input representation. We demonstrate our proposed method’s effectiveness on the benchmark datasets of few-shot classification and few-shot domain adaptation problems. Our experiment shows that our proposed idea can improve generalization performance. Moreover, the extension with image reconstruction and cross-consistency loss can stabilize domain loss minimization during training.

     

MAML2: meta reinforcement learning via meta-learning for task categories

Meta-learning has been widely applied to solving few-shot reinforcement learning problems, where we hope to obtain an agent that can learn quickly in a new task. However, these algorithms often ignore some isolated tasks in pursuit of the average performance, which may result in negative adaptation in these isolated tasks, and they usually need sufficient learning in a stationary task distribution. In this paper, our algorithm presents a hierarchical framework of double meta-learning, and the whole framework includes classification, meta-learning, and re-adaptation. Firstly, in the classification process, we classify tasks into several task subsets, considered as some categories of tasks, by learned parameters of each task, which can separate out some isolated tasks thereafter. Secondly, in the meta-learning process, we learn category parameters in all subsets via meta-learning. Simultaneously, based on the gradient of each category parameter in each subset, we use meta-learning again to learn a new meta-parameter related to the whole task set, which can be used as an initial parameter for the new task. Finally, in the re-adaption process, we adapt the parameter of the new task with two steps, by the meta-parameter and the appropriate category parameter successively. Experimentally, we demonstrate our algorithm prevents the agent from negative adaptation without losing the average performance for the whole task set. Additionally, our algorithm presents a more rapid adaptation process within re-adaptation. Moreover, we show the good performance of our algorithm with fewer samples as the agent is exposed to an online meta-learning setting.     

元迁移学习在少样本跨域图像分类中的研究

目的 现有基于元学习的主流少样本学习方法假设训练任务和测试任务服从相同或相似的分布,然而在分布差异较大的跨域任务上,这些方法面临泛化能力弱、分类精度差等挑战。同时,基于迁移学习的少样本学习方法没有考虑到训练和测试阶段样本类别不一致的情况,在训练阶段未能留下足够的特征嵌入空间。为了提升模型在有限标注样本困境下的跨域图像分类能力,提出简洁的元迁移学习（compressed meta transfer learning,CMTL）方法。方法 基于元学习,对目标域中的支持集使用数据增强策略,构建新的辅助任务微调元训练参数,促使分类模型更加适用于域差异较大的目标任务。基于迁移学习,使用自压缩损失函数训练分类模型,以压缩源域中基类数据所占据的特征嵌入空间,微调阶段引导与源域分布差异较大的新类数据有更合适的特征表示。最后,将以上两种策略的分类预测融合视为最终的分类结果。结果 使用mini-ImageNet作为源域数据集进行训练,分别在EuroSAT(EuropeanSatellite)、ISIC(InternationalSkinImagingCollaboration)、CropDiseas(Cr...     

基于多样真实任务生成的鲁棒小样本分类方法

随着大数据、计算机与互联网等技术的不断进步,以机器学习和深度学习为代表的人工智能技术取得了巨大成功,尤其是最近不断涌现的各种大模型,极大地加速了人工智能技术在各个领域的应用.但这些技术的成功离不开海量训练数据和充足的计算资源,大大限制了这些方法在一些数据或计算资源匮乏领域的应用.因此,如何利用少量样本进行学习,也就是小样本学习成为以人工智能技术引领新一轮产业变革中一个十分重要的研究问题.小样本学习中最常用的方法是基于元学习的方法,这类方法通过在一系列相似的训练任务上学习解决这类任务的元知识,在新的测试任务上利用元知识可以进行快速学习.虽然这类方法在小样本分类任务上取得了不错的效果,但是这类方法的一个潜在假设是训练任务和测试任务来自同一分布.这意味着训练任务需要足够多才能使模型学到的元知识泛化到不断变化的测试任务中.但是在一些真正数据匮乏的应用场景,训练任务的数量也是难以保证的.为此,提出一种基于多样真实任务生成的鲁棒小样本分类方法(DATG).该方法通过对已有少量任务进行Mixup,可以生成更多的训练任务帮助模型进行学习.通过约束生成任务的多样性和真实性,该方法可以有效提高小样本分类方...     

任务相关的图像小样本深度学习分类方法研究

传统基于度量学习的图像小样本分类方法与任务无关,这导致模型对新查询任务的泛化能力 较差。针对该问题,该研究提出一种任务相关的图像小样本深度学习方法——可以根据查询任务自适 应地调整支持集样本特征,从而有效形成任务相关的度量分类器。同时,该研究通过引入多种正则化 方法,解决了数据量严重不足所带来的过拟合问题。基于 miniImageNet 和 tieredImageNet 两个常用标 准数据集,在特征提取网络相同的前提下,所提出方法在 miniImageNe 中 1-shot 上获得了 66.05% 的 准确率,较目前最好的模型提高了 4.29%。     

多模态引导的局部特征选择小样本学习方法

深度学习模型取得了令人瞩目的成绩,但其训练依赖于大量的标注样本,在标注样本匮乏的场景下模型表现不尽人意.针对这一问题,近年来以研究如何从少量样本快速学习的小样本学习被提了出来,方法主要采用元学习方式对模型进行训练,取得了不错的学习效果.但现有方法:1)通常仅基于样本的视觉特征来识别新类别,信息源较为单一; 2)元学习的使用使得模型从大量相似的小样本任务中学习通用的、可迁移的知识,不可避免地导致模型特征空间趋于一般化,存在样本特征表达不充分、不准确的问题.为解决上述问题,将预训练技术和多模态学习技术引入小样本学习过程,提出基于多模态引导的局部特征选择小样本学习方法.所提方法首先在包含大量样本的已知类别上进行模型预训练,旨在提升模型的特征表达能力;而后在元学习阶段,方法利用元学习对模型进行进一步优化,旨在提升模型的迁移能力或对小样本环境的适应能力,所提方法同时基于样本的视觉特征和文本特征进行局部特征选择来提升样本特征的表达能力,以避免元学习过程中模型特征表达能力的大幅下降;最后所提方法利用选择后的样本特征进行小样本学习.在MiniImageNet、CIFAR-FS和FC-100这3个基准数...     

Learning dataset representation for automatic machine learning algorithm selection

The algorithm selection problem is defined as identifying the best-performing machine learning (ML) algorithm for a given combination of dataset, task, and evaluation measure. The human expertise required to evaluate the increasing number of ML algorithms available has resulted in the need to automate the algorithm selection task. Various approaches have emerged to handle the automatic algorithm selection challenge, including meta-learning. Meta-learning is a popular approach that leverages accumulated experience for future learning and typically involves dataset characterization. Existing meta-learning methods often represent a dataset using predefined features and thus cannot be generalized across different ML tasks, or alternatively, learn a dataset’s representation in a supervised manner and therefore are unable to deal with unsupervised tasks. In this study, we propose a novel learning-based task-agnostic method for producing dataset representations. Then, we introduce TRIO, a meta-learning approach, that utilizes the proposed dataset representations to accurately recommend top-performing algorithms for previously unseen datasets. TRIO first learns graphical representations for the datasets, using four tools to learn the latent interactions among dataset instances and then utilizes a graph convolutional neural network technique to extract embedding representations from the graphs obtained. We extensively evaluate the effectiveness of our approach on 337 datasets and 195 ML algorithms, demonstrating that TRIO significantly outperforms state-of-the-art methods for algorithm selection for both supervised (classification and regression) and unsupervised (clustering) tasks.

     

Self-Supervised Task Augmentation for Few-Shot Intent Detection

Few-shot intent detection is a practical challenge task, because new intents are frequently emerging and collecting large-scale data for them could be costly. Meta-learning, a promising technique for leveraging data from previous tasks to enable efficient learning of new tasks, has been a popular way to tackle this problem. However, the existing meta-learning models have been evidenced to be overfitting when the meta-training tasks are insufficient. To overcome this challenge, we present a novel self-supervised task augmentation with meta-learning framework, namely STAM. Firstly, we introduce the task augmentation, which explores two different strategies and combines them to extend meta-training tasks. Secondly, we devise two auxiliary losses for integrating self-supervised learning into meta-learning to learn more generalizable and transferable features. Experimental results show that STAM can achieve consistent and considerable performance improvement to existing state-of-the-art methods on four datasets.     

面向小样本数据的机器学习方法研究综述

小样本学习是面向小样本数据的机器学习,旨在利用较少的有监督样本数据去构建能够解决实际问题的机器学习模型。小样本学习能够解决传统机器学习方法在样本数据不充分时性能严重下降的问题,可以为新型小样本任务实现低成本和快速的模型部署,缩小人类智能与人工智能之间的距离,对推动发展通用型人工智能具有重要意义。从小样本学习的概念、基础模型和实际应用入手,系统梳理当前小样本学习的相关工作,将小样本学习方法分类为基于模型微调、基于数据增强、基于度量学习和基于元学习,并具体阐述这4大类方法的核心思想、基本模型、细分领域和最新研究进展,以及每一类方法在科学研究或实际应用中存在的问题,总结目前小样本学习研究的常用数据集和评价指标,整理基于部分典型小样本学习方法在Omniglot和Mini-ImageNet数据集上的实验结果。最后对各种小样本学习方法及其优缺点进行总结,分别从数据层面、理论研究和应用研究3个方面对小样本学习的未来研究方向进行展望。     

Multimodal Data Fusion for Few-shot Named Entity Recognition Method

As a crucial subtask in Natural Language Processing (NLP), Named Entity Recognition (NER) aims to extract import information from text, which can help many downstream tasks such as machine translation, text generation, knowledge graph construction, and multimodal data fusion to deeply understand the complex semantic information of the text and effectively complete these tasks. In practice, due to time and labor costs, NER suffers from annotated data scarcity, known as few-shot NER. Although few-shot NER methods based on text have achieved good generalization performance, the semantic information that the model can extract is still limited due to the few samples, which leads to the poor prediction effect of the model. To this end, in this paper we propose a few-shot NER model based on multimodal data fusion, which provides additional semantic information with multimodal data for the first time, to help the model prediction and can further effectively improve the effect of multimodal data fusion and modeling. This method converts image information into text information as auxiliary modality information, which effectively solves the problem of poor modality alignment caused by the inconsistent granularity of semantic information contained in text and images. In order to effectively consider the label dependencies in few-shot NER, we use the CRF framework and introduce the state-of-the-art meta-learning methods as the emission module and the transition module. To alleviate the negative impact of noise samples in the auxiliary modal samples, we propose a general denoising network based on the idea of meta-learning. The denoising network can measure the variability of the samples and evaluate the beneficial extent of each sample to the model. Finally, we conduct extensive experiments on real unimodal and multimodal datasets. The experimental results show the outstanding generalization performance of the proposed method, where our method outperforms the state-of-the-art methods by 10 F1 scores in the 1-shot scenario.     

融合多模态数据的小样本命名实体识别方法

作为自然语言处理领域的关键子任务,命名实体识别通过提取文本中的关键信息,帮助机器翻译、文本生成、知识图谱构建以及多模态数据融合等许多下游任务深度理解文本蕴含的复杂语义信息,有效地完成任务.在实际生活中,由于时间和人力等成本问题,命名实体识别任务常常受限于标注样本的稀缺.尽管基于文本的小样本命名实体识别方法已取得较好的泛化表现,但由于样本量有限,使得模型能提取的语义信息也十分受限,进而导致模型预测效果依然不佳.针对标注样本稀缺给基于文本的小样本命名实体识别方法带来的挑战,提出了一种融合多模态数据的小样本命名实体识别模型,借助多模态数据提供额外语义信息,帮助模型提升预测效果,进而可以有效提升多模态数据融合、建模效果.该方法将图像信息转化为文本信息作为辅助模态信息,有效地解决了由文本与图像蕴含语义信息粒度不一致导致的模态对齐效果不佳的问题.为了有效地考虑实体识别中的标签依赖关系,使用CRF框架并使用最先进的元学习方法分别作为发射模块和转移模块.为了缓解辅助模态中的噪声样本对模型的负面影响,提出一种基于元学习的通用去噪网络.该去噪网络在数据量十分有限的情况下,依然可以有效地评估辅助模态中不同样...     

Teachers cooperation: team-knowledge distillation for multiple cross-domain few-shot learning

Although few-shot learning (FSL) has achieved great progress, it is still an enormous challenge especially when the source and target set are from different domains, which is also known as cross-domain few-shot learning (CD-FSL). Utilizing more source domain data is an effective way to improve the performance of CD-FSL. However, knowledge from different source domains may entangle and confuse with each other, which hurts the performance on the target domain. Therefore, we propose team-knowledge distillation networks (TKD-Net) to tackle this problem, which explores a strategy to help the cooperation of multiple teachers. Specifically, we distill knowledge from the cooperation of teacher networks to a single student network in a meta-learning framework. It incorporates task-oriented knowledge distillation and multiple cooperation among teachers to train an efficient student with better generalization ability on unseen tasks. Moreover, our TKD-Net employs both response-based knowledge and relation-based knowledge to transfer more comprehensive and effective knowledge. Extensive experimental results on four fine-grained datasets have demonstrated the effectiveness and superiority of our proposed TKD-Net approach.     

A fusion spatial attention approach for few-shot learning

Few-shot learning is a challenging problem in computer vision that aims to learn a new visual concept from very limited data. A core issue is that there is a large amount of uncertainty introduced by the small training set. For example, the few images may include cluttered backgrounds or different scales of objects. Existing approaches mostly address this problem from either the original image space or the embedding space by using meta-learning. To the best of our knowledge, none of them tackle this problem from both spaces jointly. To this end, we propose a fusion spatial attention approach that performs spatial attention in both image and embedding spaces. In the image space, we employ a Saliency Object Detection (SOD) module to extract the saliency map of an image and provide it to the network as an additional channel. In the embedding space, we propose an Adaptive Pooling (Ada-P) module tailored to few-shot learning that introduces a meta-learner to adaptively fuse local features of the feature maps for each individual embedding. The fusion process assigns different pooling weights to the features at different spatial locations. Then, weighted pooling can be conducted over an embedding to fuse local information, which can avoid losing useful information by considering the spatial importance of the features. The SOD and Ada-P modules can be used within a plug-and-play module and incorporated into various existing few-shot learning approaches. We empirically demonstrate that designing spatial attention methods for few-shot learning is a nontrivial task and our method has proven effective for it. We evaluate our method using both shallow and deeper networks on three widely used few-shot learning benchmarks, miniImageNet, tieredImageNet and CUB, and demonstrate very competitive performance.     

自监督学习下小样本遥感图像场景分类

目的 卷积神经网络(convolutional neural network, CNN)在遥感场景图像分类中广泛应用,但缺乏训练数据依然是不容忽视的问题。小样本遥感场景分类是指模型只需利用少量样本训练即可完成遥感场景图像分类任务。虽然现有基于元学习的小样本遥感场景图像分类方法可以摆脱大数据训练的依赖,但模型的泛化能力依然较弱。为了解决这一问题,本文提出一种基于自监督学习的小样本遥感场景图像分类方法来增加模型的泛化能力。方法 本文方法分为两个阶段。首先,使用元学习训练老师网络直到收敛;然后,双学生网络和老师网络对同一个输入进行预测。老师网络的预测结果会通过蒸馏损失指导双学生网络的训练。另外,在图像特征进入分类器之前,自监督对比学习通过度量同类样本的类中心距离,使模型学习到更明确的类间边界。两种自监督机制能够使模型学习到更丰富的类间关系,从而提高模型的泛化能力。结果 本文在NWPU-RESISC45(North Western Polytechnical University-remote sensing image scene classification)、AID(aerial ima...     

结合记忆与迁移学习的小样本学习

小样本学习是视觉识别中的一个受关注的领域,旨在通过少量的数据来学习新的视觉概念。为了解决小样本问题,一些元学习方法提出从大量辅助任务中学习可迁移的知识并将其应用于目标任务上。为了更好地对知识进行迁移,提出了一种基于记忆的迁移学习方法。提出一种权重分解策略,将部分权重分解为冻结权重与可学习权重,在迁移学习中通过固定冻结权重,仅更新可学习权重的方式来减少模型需要学习的参数。通过一个额外的记忆模块来存储之前任务的经验,在学习新任务时,这些经验被用来初始化模型的参数状态,以此更好地进行迁移学习。通过在miniImageNet、tieredImageNet以及CUB数据集上的实验结果表明,相对于其他先进的方法,该方法在小样本分类任务上取得了具有竞争力甚至是更好的表现。     

基于元学习的小样本遥感图像分类

为改善传统分类算法在小样本遥感图像分类上效果差的缺陷,提升模型的快速学习能力,提出融合迁移学习和元学习的小样本分类算法.设计基于长短期记忆网络的元学习器,通过门控结构拟合网络参数更新方式最下化损失下界,具有自动学习分类器参数更新方式的机制,相比于传统方法,能够有效扩展优化算法的搜索空间;考虑样本的跨类别知识转移和训练时...     

基于深度神经网络的少样本学习综述

如何从少数训练样本中学习并识别新的类别对于深度神经网络来说是一个具有挑战性的问题。针对如何解决少样本学习的问题,全面总结了现有的基于深度神经网络的少样本学习方法,涵盖了方法所用模型、数据集及评估结果等各个方面。具体地,针对基于深度神经网络的少样本学习方法,提出将其分为数据增强方法、迁移学习方法、度量学习方法和元学习方法四种类别;对于每个类别,进一步将其分为几个子类别,并且在每个类别与方法之间进行一系列比较,以显示各种方法的优劣和各自的特点。最后强调了现有方法的局限性,并指出了少样本学习研究领域未来的研究方向。     

Unsupervised law article mining based on deep pre-trained language representation models with application to the Italian civil code

Modeling law search and retrieval as prediction problems has recently emerged as a predominant approach in law intelligence. Focusing on the law article retrieval task, we present a deep learning framework named LamBERTa, which is designed for civil-law codes, and specifically trained on the Italian civil code. To our knowledge, this is the first study proposing an advanced approach to law article prediction for the Italian legal system based on a BERT (Bidirectional Encoder Representations from Transformers) learning framework, which has recently attracted increased attention among deep learning approaches, showing outstanding effectiveness in several natural language processing and learning tasks. We define LamBERTa models by fine-tuning an Italian pre-trained BERT on the Italian civil code or its portions, for law article retrieval as a classification task. One key aspect of our LamBERTa framework is that we conceived it to address an extreme classification scenario, which is characterized by a high number of classes, the few-shot learning problem, and the lack of test query benchmarks for Italian legal prediction tasks. To solve such issues, we define different methods for the unsupervised labeling of the law articles, which can in principle be applied to any law article code system. We provide insights into the explainability and interpretability of our LamBERTa models, and we present an extensive experimental analysis over query sets of different type, for single-label as well as multi-label evaluation tasks. Empirical evidence has shown the effectiveness of LamBERTa, and also its superiority against widely used deep-learning text classifiers and a few-shot learner conceived for an attribute-aware prediction task.