基于小样本学习的SAR图像识别

深度学习已成为图像识别领域的一个研究热点.与传统图像识别方法不同,深度学习从大量数据中自动学习特征,并且具有强大的自学习能力和高效的特征表达能力.但在小样本条件下,传统的深度学习方法如卷积神经网络难以学习到有效的特征,造成图像识别的准确率较低.因此,提出一种新的小样本条件下的图像识别算法用于解决SAR图像的分类识别.该...     

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法的泛化性.具体来说,先对训练集中的基类进行聚类得到不同的簇,然后从不同的簇中选取任务进行Mixup以增加生成任务的多样性.此外,簇间任务Mixup策略可以减轻学习到与类别高度相关的伪判别特征.同时,为了避免生成的任务与真实分布太偏离,误导模型学习,通过最小化生成任务与真实任务之间的最大均值差异(MMD)来保证生成任务的真实性.最后,从理论上分析了为什么基于簇间任务Mixup的策略可以提高模型的泛化性能.多个数据集上的实验结果进一步证明了所提出的基于多样性和真实性任务扩充方法的有效性.     

Discovery as Autonomous Learning from the Environment

Discovery involves collaboration among many intelligent activities. However, little is known about how and in what form such collaboration occurs. In this article, a framework is proposed for autonomous systems that learn and discover from their environment. Within this framework, many intelligent activities such as perception, action, exploration, experimentation, learning, problem solving, and new term construction can be integrated in a coherent way. The framework is presented in detail through an implemented system called LIVE, and is evaluated through the performance of LIVE on several discovery tasks. The conclusion is that autonomous learning from the environment is a feasible approach for integrating the activities involved in a discovery process.     

风洞试验绳牵引并联支撑系统自适应滑模控制

针对绳牵引并联支撑系统在风洞试验中的应用,提出一种自适应滑模控制方法以提高飞行器模型动态试验的运动精度.首先,详细分析了系统不确定因素,并重点考虑了气动力与绳弹性变形的影响,重构了系统动力学方程;基于奇异摄动理论,提出一种复合控制律,其中对慢变状态量采用自适应连续非奇异终端滑模控制,对快变状态量采用微分控制;通过李雅普诺夫函数法对系统的稳定性进行了分析,确定了控制律中微分增益项的影响.最后,以两种典型的动态轨迹为例,考虑气动力建模,对所设计控制律进行多参数仿真分析.结果表明该复合控制律可以减小绳弹性以及气动力等不确定性参数对跟踪误差的影响,提高运动控制精度,因此该控制方法有效可行,可为绳牵引并联支撑的动态试验应用提供理论指导.     

融合三元卷积神经网络与关系网络的小样本食品图像识别

食品识别在食品健康和智能家居等领域获得了广泛关注.目前大部分的食品识别工作是基于大规模标记样本的深度神经网络,这些工作无法有效地识别只有少量样本的类别,因此小样本食品识别是一个亟待解决的问题.目前基于度量学习的小样本识别方法着重于探究样本之间的相似度信息,忽略了类内与类间更加细粒度的区分.学习类内与类间区分信息的主流方...     

面向多标签小样本学习的双流重构网络

多标签图像分类问题是计算机视觉领域的重要问题之一,它需要对图像中的所有标签进行预测.而一幅图像中待分类的标签个数往往不止一个,同时图像中对象的大小、位置和姿态的变化都会对模型的分类性能产生影响.因此,如何有效地提高图像特征的准确表达能力是一个亟需解决的难题.针对上述难题,文中提出了一个新颖的双流重构网络来对图像进行特征...     

Robot navigation in the real world:: Experiments with Manchester’s FortyTwo in unmodified, large environments

Mobile robot navigation under controlled laboratory conditions is, by now, state of the art and reliably achievable. To transfer navigation mechanisms used in such small-scale environments to applications in untreated, large environments, however, is not trivial, and typically requires modifications to the original navigation mechanism: scaling up is hard.In this paper, we discuss the difficulties of mobile robot navigation in general, the various options to achieve navigation in large environments, and experiments with Manchester’s FortyTwo, which investigate how scaling up of navigational competencies can be achieved. We were particularly interested in autonomous mobile robot navigation in unmodified, large and varied environments, without the aid of pre-installed maps or supplied CAD models of the environment. This paper presents a general approach to achieve this.FortyTwo regularly travels the corridors of the Department of Computer Science at Manchester University, using topological maps, landmarks, low level “enabling behaviours” and active exploitation of features of the environment. Experimental results obtained in these environments are given in this paper.     

基于自适应原型特征类矫正的小样本学习方法

针对小样本学习过程中样本数量不足导致的性能下降问题,基于原型网络(Prototype network, ProtoNet)的小样本学习方法通过实现查询样本与支持样本原型特征间的距离度量,从而达到很好的分类性能.然而,这种方法直接将支持集样本均值视为类原型,在一定程度上加剧了对样本数量稀少情况下的敏感性.针对此问题,提出了基于自适应原型特征类矫正的小样本学习方法 (Few-shot learning based on class rectification via adaptive prototype features, CRAPF),通过自适应生成原型特征来缓解方法对数据细微变化的过度响应,并同步实现类边界的精细化调整.首先,使用卷积神经网络构建自适应原型特征生成模块,该模块采用非线性映射获取更为稳健的原型特征,有助于减弱异常值对原型构建的影响;然后,通过对原型生成过程的优化,提升不同类间原型表示的区分度,进而强化原型特征对类别表征的整体效能;最后,在3个广泛使用的基准数据集上的实验结果显示,该方法提升了小样本学习任务的表现.     

基于BERT的社交电商文本分类算法

随着网络购物的高速发展,网络商家和购物者在网络交易活动中产生了大量的交易数据,其中蕴含着巨大的分析价值。针对社交电商商品文本的文本分类问题,为了更加高效准确地判断文本所描述商品的类别,提出了一种基于BERT模型的社交电商文本分类算法。首先,该算法采用BERT(Bidirectional Encoder Representations from Transformers)预训练语言模型来完成社交电商文本的句子层面的特征向量表示,随后有针对性地将获得的特征向量输入分类器进行分类,最后采用社交电商文本的数据集进行算法验证。实验结果表明,经过训练的模型在测试集上的分类结果F1值最高可达94.61%,高出BERT模型针对MRPC的分类任务6%。因此,所提社交电商文本分类算法能够较为高效准确地判断文本所描述商品的类别,有助于进一步分析网络交易数据,从海量数据中提取有价值的信息。     

强化学习中资格迹的作用

强化学习一词来自行为心理学,该学科把学习看作反复试验的过程,强化学习系统中的资格迹用来解决时间信度分配问题,文章介绍了资格迹的基本原理和实现方法。     

一种多层次选择性集成学习算法

选择性集成学习是为解决同一个问题而训练多个基分类器,并依据某种规则选取部分基分类器的结果进行整合的学习算法。通过选择性集成可以获得比单个学习器和全部集成学习更好的学习效果,可以显著地提高学习系统的泛化性能。提出了一种多层次选择性集成学习算法Ada_ens。试验结果表明,Ada_ens具有更好的学习效果和泛化性能。     

风洞通用压力扫描阀综合数据采集系统的设计与实现

为了满足风洞中常用压力扫描阀的数据采集要求,设计了一套风洞通用压力扫描阀综合数据采集系统。本文首先对风洞压力扫描阀数据采集系统的结构组成以及常用的压力扫描阀进行了概述。其次,详细的给出了本系统的设计思路、系统组成以及实现方法。最后,使用本系统分别对两台DSA 3217、两台PSI 9116、一台DSA 3217和一台PSI 9116三种情况进行压力测试,测试结果表明,三种情况的数据差异都比较小,分别为3.2Pa、1.6Pa、9.9Pa,同时,数据时间差最大值也很小,分别为5ms、8ms、30ms。由此,可以说明本系统具有很好的同步性与兼容性,能够实现风洞中的各类压力扫描阀的综合数据采集。     

An introduction to inductive programming

The research field of inductive programming is concerned with the design of algorithms for learning computer programs with complex flow of control (typically recursive calls) from incomplete specifications such as examples. We introduce a basic algorithmic approach for inductive programming and illustrate it with three systems: dialogs learns logic programs by combining inductive and abductive reasoning; the classical thesys system and its extension igor1 learn functional programs based on a recurrence detection mechanism in traces; igor2 learns functional programs over algebraic data-types making use of constructor-term rewriting systems. Furthermore, we give a short history of inductive programming, discuss related approaches, and give hints about current applications and possible future directions of research. A short, non-technical version of this paper appears in C. Sammut, editor, Encyclopedia of Machine Learning, Springer–Verlag, forthcoming. The paper was written while the first author was on sabbatical in 2006/2007 at Sabancı University in İstanbul, Turkey.     

Learning control and knowledge representation

Practising to operate an unknown system and observing the input and output of the system, in a sense, helps to optimally control that system. The acquired knowledge, is, in turn, used to solve future analogous control problems. This means that it is very important to know how to memorize the acquired knowledge and to utilize it for learning. In this paper, we propose a new knowledge representation and reasoning method and develop a learning machine (KBLC: Knowledge-Based Learning Controller) by using them. A simple implementation has been constructed that demonstrates the feasibility of building such a machine.     

Multiagent learning is not the answer. It is the question

The article by Shoham, Powers, and Grenager called “If multi-agent learning is the answer, what is the question?” does a great job of laying out the current state of the art and open issues at the intersection of game theory and artificial intelligence (AI). However, from the AI perspective, the term “multiagent learning” applies more broadly than can be usefully framed in game theoretic terms. In this larger context, how (and perhaps whether) multiagent learning can be usefully applied in complex domains is still a large open question.