Convolutional neural networks (CNNs) have constantly achieved better performance over years by introducing more complex topology, and enlarging the capacity towards deeper and wider CNNs. This makes the manual design of CNNs extremely difficult, so the automated design of CNNs has come into the research spotlight, which has obtained CNNs that outperform manually-designed CNNs. However, the computational cost is still the bottleneck of automatically designing CNNs. In this paper, inspired by transfer learning, a new evolutionary computation based framework is proposed to efficiently evolve CNNs without compromising the classification accuracy. The proposed framework leverages multi-source domains, which are smaller datasets than the target domain datasets, to evolve a generalised CNN block only once. And then, a new stacking method is proposed to both widen and deepen the evolved block, and a grid search method is proposed to find optimal stacking solutions. The experimental results show the proposed method acquires good CNNs faster than 15 peer competitors within less than 40 GPU-hours. Regarding the classification accuracy, the proposed method gains its strong competitiveness against the peer competitors, which achieves the best error rates of 3.46%, 18.36% and 1.76% for the CIFAR-10, CIFAR-100 and SVHN datasets, respectively.
translated by 谷歌翻译
Deep convolutional neural networks have proven their effectiveness, and have been acknowledged as the most dominant method for image classification. However, a severe drawback of deep convolutional neural networks is poor explainability. Unfortunately, in many real-world applications, users need to understand the rationale behind the predictions of deep convolutional neural networks when determining whether they should trust the predictions or not. To resolve this issue, a novel genetic algorithm-based method is proposed for the first time to automatically evolve local explanations that can assist users to assess the rationality of the predictions. Furthermore, the proposed method is model-agnostic, i.e., it can be utilised to explain any deep convolutional neural network models. In the experiments, ResNet is used as an example model to be explained, and the ImageNet dataset is selected as the benchmark dataset. DenseNet and MobileNet are further explained to demonstrate the model-agnostic characteristic of the proposed method. The evolved local explanations on four images, randomly selected from ImageNet, are presented, which show that the evolved local explanations are straightforward to be recognised by humans. Moreover, the evolved explanations can explain the predictions of deep convolutional neural networks on all four images very well by successfully capturing meaningful interpretable features of the sample images. Further analysis based on the 30 runs of the experiments exhibits that the evolved local explanations can also improve the probabilities/confidences of the deep convolutional neural network models in making the predictions. The proposed method can obtain local explanations within one minute, which is more than ten times faster than LIME (the state-of-the-art method).
translated by 谷歌翻译
数据有效的图像分类是一项具有挑战性的任务,旨在使用小型培训数据来解决图像分类。基于神经网络的深度学习方法对于图像分类很有效,但是它们通常需要大规模的培训数据,并且具有重大局限性,例如需要专业知识来设计网络架构和具有差的可解释性。进化深度学习是一个最近的热门话题,将进化计算与深度学习结合在一起。但是,大多数进化的深度学习方法都集中在神经网络的架构上,这些方法仍然遭受诸如不良解释性之类的局限性。为了解决这个问题,本文提出了一种新的基于基因编程的进化深度学习方法,以进行数据有效的图像分类。新方法可以使用来自图像和分类域的许多重要运算符自动发展可变长度模型。它可以从颜色或灰度图像中学习不同类型的图像特征,并构建有效而多样的合奏以进行图像分类。灵活的多层表示可以使新方法自动构建浅层或深模型/树以进行不同的任务,并通过多个内部节点对输入数据进行有效的转换。新方法用于解决具有不同训练集大小的五个图像分类任务。结果表明,在大多数情况下,它比深度学习方法的图像分类更好。深入的分析表明,新方法具有良好的收敛性,并演变具有高解释性,不同长度/尺寸/形状以及良好可传递性的模型。
translated by 谷歌翻译
在本文中,我们开发了损失功能学习的新兴主题,该主题旨在学习损失功能,从而显着提高在其下方训练的模型的性能。具体而言,我们提出了一个新的元学习框架,用于通过混合神经符号搜索方法来学习模型 - 不足的损失函数。该框架首先使用基于进化的方法来搜索原始数学操作的空间,以找到一组符号损耗函数。其次,随后通过基于端梯度的训练程序对学习的损失功能集进行了参数化和优化。拟议框架的多功能性在经验上得到了各种监督的学习任务的经验验证。结果表明,通过新提出的方法发现的元学习损失函数在各种神经网络体系结构和数据集上都超过了交叉渗透丢失和最新的损失函数学习方法。
translated by 谷歌翻译
计算机视觉(CV)是涵盖广泛应用的人工智能中的一个重要领域。图像分析是CV的主要任务,目的是提取,分析和理解图像的视觉内容。但是,由于许多因素,图像之间的较高变化,高维度,域专业知识要求和图像扭曲,因此与图像相关的任务非常具有挑战性。进化计算方法(EC)方法已被广泛用于图像分析,并取得了重大成就。但是,没有对现有的EC方法进行图像分析的全面调查。为了填补这一空白,本文提供了一项全面的调查,涵盖了重要的图像分析任务的所有基本EC方法,包括边缘检测,图像分割,图像特征分析,图像分类,对象检测等。这项调查旨在通过讨论不同方法的贡献并探讨如何以及为什么将EC用于简历和图像分析,以更好地了解进化计算机视觉(ECV)。还讨论并总结了与该研究领域相关的应用,挑战,问题和趋势,以提供进一步的指南和未来研究的机会。
translated by 谷歌翻译
近年来,行业和学术界的深度学习(DL)迅速发展。但是,找到DL模型的最佳超参数通常需要高计算成本和人类专业知识。为了减轻上述问题,进化计算(EC)作为一种强大的启发式搜索方法显示出在DL模型的自动设计中,所谓的进化深度学习(EDL)具有重要优势。本文旨在从自动化机器学习(AUTOML)的角度分析EDL。具体来说,我们首先从机器学习和EC阐明EDL,并将EDL视为优化问题。根据DL管道的说法,我们系统地介绍了EDL方法,从功能工程,模型生成到具有新的分类法的模型部署(即,什么以及如何发展/优化),专注于解决方案表示和搜索范式的讨论通过EC处理优化问题。最后,提出了关键的应用程序,开放问题以及可能有希望的未来研究线。这项调查回顾了EDL的最新发展,并为EDL的开发提供了有见地的指南。
translated by 谷歌翻译
我们引入了一个新的非线性降低框架的新框架,其中预测因子和响应都是分布数据,它们被建模为度量空间的成员。我们实现非线性足够尺寸降低的关键步骤是在度量空间上构建通用内核,从而导致繁殖Hilbert空间的预测变量和响应,这些空间足以表征有条件的独立性,以决定足够的尺寸减少。对于单变量分布,我们使用Wasserstein距离的众所周知的分位数来构建通用内核。对于多元分布,我们求助于最近开发的切成薄片的Wasserstein距离,以实现此目的。由于可以通过单变量瓦斯汀距离的分位数表示来计算切片的瓦斯坦距离,因此多变量瓦斯坦距离的计算保持在可管理的水平。该方法应用于几个数据集,包括生育能力和死亡率分布数据和卡尔加里温度数据。
translated by 谷歌翻译
基于添加条件独立性,我们为离散节点变量引入非参数图形模型。添加剂条件独立性是一种三种方式统计关系,其通过满足半石灰阳极公理来利用有条件独立性与有条件的独立性共享类似的性质。基于该关系,我们构建了一种用于离散变量的加性图形模型,其不受诸如诸如Ising模型的参数模型的限制。我们通过惩罚添加精度运算符的离散版本的惩罚估算来开发新的图形模型的估计,并在超高维设置下建立估计器的一致性。随着这些方法的发展,我们还利用离散随机变量的性质来揭示添加剂条件独立性与条件独立性之间的更深层次关系。新的图形模型在某些稀疏条件下减少了条件独立性图形模型。我们进行仿真实验和对HIV抗逆转录病毒治疗数据集的分析,以比较现有的新方法。
translated by 谷歌翻译
With the rapid development of data collection techniques, complex data objects that are not in the Euclidean space are frequently encountered in new statistical applications. Fr\'echet regression model (Peterson & M\"uller 2019) provides a promising framework for regression analysis with metric space-valued responses. In this paper, we introduce a flexible sufficient dimension reduction (SDR) method for Fr\'echet regression to achieve two purposes: to mitigate the curse of dimensionality caused by high-dimensional predictors and to provide a visual inspection tool for Fr\'echet regression. Our approach is flexible enough to turn any existing SDR method for Euclidean (X,Y) into one for Euclidean X and metric space-valued Y. The basic idea is to first map the metric-space valued random object $Y$ to a real-valued random variable $f(Y)$ using a class of functions, and then perform classical SDR to the transformed data. If the class of functions is sufficiently rich, then we are guaranteed to uncover the Fr\'echet SDR space. We showed that such a class, which we call an ensemble, can be generated by a universal kernel. We established the consistency and asymptotic convergence rate of the proposed methods. The finite-sample performance of the proposed methods is illustrated through simulation studies for several commonly encountered metric spaces that include Wasserstein space, the space of symmetric positive definite matrices, and the sphere. We illustrated the data visualization aspect of our method by exploring the human mortality distribution data across countries and by studying the distribution of hematoma density.
translated by 谷歌翻译
将对象检测和ID嵌入提取到统一网络的单次多对象跟踪,近年来取得了开创性的结果。然而,目前的单次追踪器仅依赖于单帧检测来预测候选界限盒,当面对灾难性的视觉下降时,例如运动模糊,闭塞时可能是不可靠的。一旦检测器错误地被错误地归类为背景,将不再维护其相应的ROCKLET的时间一致性。在本文中,我们首先通过提出重新检查网络恢复被错误分类为“假背景”的边界框。重新检查网络创新地扩展了ID从数据关联嵌入ID的角色,以通过有效地将先前的轨迹传播到具有小开销的当前帧的运动预测。请注意,传播结果由独立和有效的嵌入搜索产生,防止模型过度依赖于检测结果。最终,它有助于重新加载“假背景”并修复破碎的Tracklet。在强大的基线Cstrack上建立一个新的单次追踪器,分别通过70.7 $ 76.4,70.6 $ \右前场达到76.3美元的MOT17和MOT17。它还达到了新的最先进的Mota和IDF1性能。代码在https://github.com/judasdie/sots发布。
translated by 谷歌翻译