Abstract
We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overfitting in the fully-connected layers we employed a recently-developed regularization method called that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.
我们训练了一个大型的深度卷积神经网络，将ImageNet lsvprc -2010竞赛中的120万幅高分辨率图像分类为1000个不同的类。在测试数据上，我们实现了top-1和top-5的错误率，分别为37.5%和17.0%，这与前的最高水平相比有了很大的提高。该神经网络有6000万个参数和65万个神经元，由5个卷积层(其中一些后面接了最大池化层)和3个全连接层(最后的1000路softmax)组成。为了使训练更快，我们使用了非饱和神经元和一个非常高效的GPU实现卷积运算。为了减少全连通层的过拟合，我们采用了一种最近发展起来的正则化方法——dropout，结果显示它非常有效。我们还在ILSVRC-2012比赛中输入了该模型的一个变体，并获得了15.3%的top-5测试错误率，而第二名获得了26.2%的错误率.
1 Introduction
Current approaches to object recognition make essential use of machine learning methods. To improve their performance, we can collect larger datasets, learn more powerful models, and use better techniques for preventing overfitting. Until recently, datasets of labeled images were relatively small — on the order of tens of thousands of images (e.g., NORB [16], Caltech-101/256 [8, 9], and CIFAR-10/100 [12]). Simple recognition tasks can be solved quite well with datasets of this size, especially if they are augmented with label-preserving transformations. For example, the current best error rate on the MNIST digit-recognition task (<0.3%) approaches human performance [4]. But objects in realistic settings exhibit considerable variability, so to learn to recognize them it is necessary to use much larger training sets. And indeed, the shortcomings of small image datasets have been widely recognized (e.g., Pinto et al. [21]), but it has only recently become possible to collect labeled datasets with millions of images. The new larger datasets include LabelMe [23], which consists of hundreds of thousands of fully-segmented images, and ImageNet [6], which consists of over 15 million labeled high-resolution images in over 22,000 categories.
当前的物体识别方法主要利用机器学习方法。为了提高它们的性能，我们可以收集更大的数据集，学习更强大的模型，并使用更好的技术来防止过度拟合。直到最近，标记图像的数据集在成千上万的图像(例如，NORB [16]， Caltech-101/256 [8,9]， CIFAR-10/100[12])中相对较小。使用这种大小的数据集可以很好地解决简单的识别任务，特别是如果使用保存标签的转换来扩展它们。例如，MNIST数字识别任务的当前最佳错误率(<0.3%)接近人类性能[4]。但是现实环境中的物体表现出相当大的可变性，所以为了学会识别它们，有必要使用更大的训练集。的确，小图像数据集的缺点已经被广泛认识(例如，Pinto等人的[21])，但直到最近才有可能收集数百万张图像的标记数据集。新的更大的数据集包括LabelMe[23]，它由成千上万的全分段图像组成，和ImageNet[6]，它由超过22000个类别的超过1500万标记的高分辨率图像组成。
To learn about thousands of objects from millions of images, we need a model with a large learning capacity. However, the immense complexity of the object recognition task means that this problem cannot be specified even by a dataset as large as ImageNet, so our model should also have lots of prior knowledge to compensate for all the data we don’t have. Convolutional neural networks (CNNs) constitute one such class of models [16, 11, 13, 18, 15, 22, 26]. Their capacity can be controlled by varying their depth and breadth, and they also make strong and mostly correct assumptions about the nature of images (namely, stationarity of statistics and locality of pixel dependencies). Thus, compared to standard feedforward neural networks with similarly-sized layers, CNNs have much fewer connections and parameters and so they are easier to train, while their theoretically-best performance is likely to be only slightly worse.
要从数百万张图像中了解数千个物体，我们需要一个具有巨大学习能力的模型。
然而，对象识别任务的巨大复杂性意味着即使像ImageNet这样大的数据集也无法指定这个问题，因此我们的模型也应该具有大量的先验知识来补偿我们没有的所有数据。卷积神经网络(Convolutional neural networks, CNNs)就是这样一类模型[16,11,13,18,15,22,26]。它们的能力可以通过改变深度和宽度来控制，而且它们还对图像的性质(即统计的平稳性和像素依赖的局部性)做出了强有力且最正确的假设。
因此，与具有相似大小层的标准前馈神经网络相比，CNNs具有更少的连接和参数，因此更容易训练，而其理论上最好的性能可能只会稍微差一些。
Despite the attractive qualities of CNNs, and despite the relative efficiency of their local architecture, they have still been prohibitively expensive to apply in large scale to high-resolution images. Luckily, current GPUs, paired with a highly-optimized implementation of 2D convolution, are powerful enough to facilitate the training of interestingly-large CNNs, and recent datasets such as ImageNet contain enough labeled examples to train such models without severe overfitting.
尽管CNNs的质量很吸引人，尽管它们的本地架构相对高效，但在高分辨率图像上大规模应用仍然非常昂贵。幸运的是，当前的gpu与高度优化的2D卷积实现相结合，已经足够强大，可以方便地训练有趣的大型CNNs，而最近的数据集(如ImageNet)包含了足够多的标记示例，可以在不严重过拟合的情况下训练此类模型。
The specific contributions of this paper are as follows: we trained one of the largest convolutional neural networks to date on the subsets of ImageNet used in the ILSVRC-2010 and ILSVRC-2012 competitions[2] and achieved by far the best results ever reported on these datasets. We wrote a highly-optimized GPU implementation of 2D convolution and all the other operations inherent in training convolutional neural networks, which we make available publicly1. Our network contains a number of new and unusual features which improve its performance and reduce its training time, which are detailed in Section 3. The size of our network made overfitting a significant problem, even with 1.2 million labeled training examples, so we used several effective techniques for preventing overfitting, which are described in Section 4. Our final network contains five convolutional and three fully-connected layers, and this depth seems to be important: we found that removing any convolutional layer (each of which contains no more than 1% of the model’s parameters) resulted in inferior performance.
本文的具体贡献如下：
- 我们在在 ILSVRC-2010 和 ILSVRC-2012 比赛[2]中使用过的 ImageNet 的子集上训练了迄今为止最大的卷积神经网络之一，并取得了这些数据集上迄今为止最好的结果。
- 我们编写了一个关于 2D 卷积和所有其他的训练卷积神经网络时固有的操作的高度优化的 GPU 实现，并将其公开了1。
- 我们的网络包含了许多新的和不寻常的特性，这些特性提高了它的性能并减少了它的训练时间，这些特性在第3节中详细介绍。
- 即使有120万个标记的训练样本，我们的网络规模（过大）使得过度拟合成为一个重要的问题。所以我们使用了一些有效的技术来防止过度拟合，如第4节所述。
- 我们最终的网络包含5个卷积层和3个全连接层，这个深度似乎很重要：我们发现去掉任何卷积层（每个卷积层只包含不到1%的模型参数）都会导致性能下降。
In the end, the network’s size is limited mainly by the amount of memory available on current and by the amount of training that we are willing to tolerate. Our network takes between five and six days to train on two GTX580 3GB GPUs. All of our experiments suggest that our results can be improved simply by waiting for faster GPUs and bigger datasets to become available.
最后，网络的大小主要受到当前gpu上可用内存的大小和我们愿意忍受的训练时间的大小的限制。我们的网络需要5到6天的时间来训练两个GTX 580 3GB GPU。我们所有的实验都表明，只要等待更快的gpu和更大的数据集可用，我们的结果就可以得到改善。
2 The Dataset
ImageNet is a dataset of over 15 million labeled high-resolution images belonging to roughly 22,000 categories. The images were collected from the web and labeled by human labelers using Amazon’s Mechanical Turk crowd-sourcing tool. Starting in 2010, as part of the Pascal Visual Object Challenge, an annual competition called the ImageNet Large-Scale Visual Recognition Challenge (ILSVRC) has been held. ILSVRC uses a subset of ImageNet with roughly 1000 images in each of 1000 categories. In all, there are roughly 1.2 million training images, 50,000 validation images, and 150,000 testing images.
ImageNet是一个包含超过1500万张高分辨率图像的数据集，属于大约22000个类别。这些图片是从网上收集来的，并由人工贴标签者使用亚马逊的土耳其机械众包工具进行标记。从2010年开始，作为Pascal视觉对象挑战赛的一部分，每年都会举办一场名为ImageNet大型视觉识别挑战赛(ILSVRC)的比赛。ILSVRC使用ImageNet的一个子集，每个类别大约有1000张图片。总共大约有120万张训练图像、5万张验证图像和15万张测试图像。
ILSVRC-2010 is the only version of ILSVRC for which the test set labels are available, so this is the version on which we performed most of our experiments. Since we also entered our model in the ILSVRC-2012 competition, in Section 6 we report our results on this version of the dataset as well, for which test set labels are unavailable. On ImageNet, it is customary to report two error rates: top-1 and top-5, where the top-5 error rate is the fraction of test images for which the correct label
is not among the five labels considered most probable by the model.
ILSVRC-2010 是唯一可用测试集标签的 ILSVRC 版本，因此这是我们进行大多数实验的版本。由于我们也在 ILSVRC-2012 竞赛中加入了我们的模型，在第6节中，我们也报告了我们在这个版本的数据集上的结果，对于这个版本的数据集，测试集标签是不可用的。在 ImageNet 上，通常报告两个错误率：top-1 和 top-5，其中 top-5 错误率是测试图像的一部分，其中正确的标签不在模型认为最可能的五个标签中。
ImageNet consists of variable-resolution images, while our system requires a constant input dimensionality. Therefore, we down-sampled the images to a fixed resolution of 256 * 256. Given a rectangular image, we first rescaled the image such that the shorter side was of length 256, and then cropped out the central 256�256 patch from the resulting image. We did not pre-process the images in any other way, except for subtracting the mean activity over the training set from each pixel. So we trained our network on the (centered) raw RGB values of the pixels.
ImageNet由可变分辨率的图像组成，而我们的系统需要一个恒定的输入维数。
因此，我们将图像降采样到256 * 256的固定分辨率。给定一个矩形图像，我们首先重新调整图像的大小，使其短边长度为256，然后从结果图像中裁剪出中心的256%256块。除了从每个像素中减去训练集上的平均活动外，我们没有以任何其他方式对图像进行预处理。因此，我们将网络训练成像素的原始RGB值(居中)。
3 The Architecture
The architecture of our network is summarized in Figure 2. It contains eight learned layers — five convolutional and three fully-connected. Below, we describe some of the novel or unusual features of our network’s architecture. Sections 3.1-3.4 are sorted according to our estimation of their importance, with the most important first.
3.1 ReLU Nonlinearity
The standard way to model a neuron’s output f as a function of its input x is with or . In terms of training time with gradient descent, these saturating nonlinearities are much slower than the non-saturating nonlinearity f(x) = max(0; x). Following Nair and Hinton [20], we refer to neurons with this nonlinearity as Rectified Linear Units (ReLUs). Deep convolutional neural networks with ReLUs train several times faster than their equivalents with tanh units. This is demonstrated in Figure 1, which shows the number of iterations required to reach 25% training error on the CIFAR-10 dataset for a particular four-layer convolutional network. This plot shows that we would not have been able to experiment with such large neural networks for this work if we had used traditional saturating neuron models.
We are not the first to consider alternatives to traditional neuron models in CNNs. For example, Jarrett et al. [11] claim that the nonlinearity works particularly well with their type of contrast normalization followed by local average pooling on the Caltech-101 dataset. However, on this dataset the primary concern is preventing overfitting, so the effect they are observing is different from the accelerated ability to fit the training set which we report when using ReLUs. Faster learning has a great influence on the performance of large models trained on large datasets.
3.2 Training on Multiple GPUs
A single GTX 580 GPU has only 3GB of memory, which limits the maximum size of the networks that can be trained on it. It turns out that 1.2 million training examples are enough to train networks which are too big to fit on one GPU. Therefore we spread the net across two GPUs. Current GPUs are particularly well-suited to cross-GPU parallelization, as they are able to read from and write to one another’s memory directly, without going through host machine memory. The parallelization scheme that we employ essentially puts half of the kernels (or neurons) on each GPU, with one additional trick: the GPUs communicate only in certain layers. This means that, for example, the kernels of layer 3 take input from all kernel maps in layer 2. However, kernels in layer 4 take input only from those kernel maps in layer 3 which reside on the same GPU. Choosing the pattern of connectivity is a problem for cross-validation, but this allows us to precisely tune the amount of communication until it is an acceptable fraction of the amount of computation.
The resultant architecture is somewhat similar to that of the “columnar” CNN employed by Cire¸san et al. [5], except that our columns are not independent (see Figure 2). This scheme reduces our top-1 and top-5 error rates by 1.7% and 1.2%, respectively, as compared with a net with half as many kernels in each convolutional layer trained on one GPU. The two-GPU net takes slightly less time to train than the one-GPU net2.
3.3 Local Response Normalization
ReLUs have the desirable property that they do not require input normalization to prevent them from saturating. If at least some training examples produce a positive input to a ReLU, learning will happen in that neuron. However, we still find that the following local normalization scheme aids generalization. Denoting by ai x;y the activity of a neuron computed by applying kernel i at position (x; y) and then applying the ReLU nonlinearity, the response-normalized activity bi x;y is given by the expression.
where the sum runs over n “adjacent” kernel maps at the same spatial position, and N is the total number of kernels in the layer. The ordering of the kernel maps is of course arbitrary and determined before training begins. This sort of response normalization implements a form of lateral inhibition inspired by the type found in real neurons, creating competition for big activities amongst neuron outputs computed using different kernels. The constants k; n; , and are hyper-parameters whose values are determined using a validation set; we used k = 2, n = 5, = 104, and = 0:75. We applied this normalization after applying the ReLU nonlinearity in certain layers (see Section 3.5).
This scheme bears some resemblance to the local contrast normalization scheme of Jarrett et al. [11], but ours would be more correctly termed “brightness normalization”, since we do not subtract the mean activity. Response normalization reduces our top-1 and top-5 error rates by 1.4% and 1.2%, respectively. We also verified the effectiveness of this scheme on the CIFAR-10 dataset: a four-layer
CNN achieved a 13% test error rate without normalization and 11% with normalization3.
3.4 Overlapping Pooling
Pooling layers in CNNs summarize the outputs of neighboring groups of neurons in the same kernel map. Traditionally, the neighborhoods summarized by adjacent pooling units do not overlap (e.g.,[17, 11, 4]). To be more precise, a pooling layer can be thought of as consisting of a grid of pooling units spaced s pixels apart, each summarizing a neighborhood of size z z centered at the location of the pooling unit. If we set s = z, we obtain traditional local pooling as commonly employed in CNNs. If we set s < z, we obtain overlapping pooling. This is what we use throughout our network, with s = 2 and z = 3. This scheme reduces the top-1 and top-5 error rates by 0.4% and 0.3%, respectively, as compared with the non-overlapping scheme s = 2; z = 2, which produces output of equivalent dimensions. We generally observe during training that models with overlapping pooling find it slightly more difficult to overfit.
3.5 Overall Architecture
Now we are ready to describe the overall architecture of our CNN. As depicted in Figure 2, the net contains eight layers with weights; the first five are convolutional and the remaining three are fully-connected. The output of the last fully-connected layer is fed to a 1000-way softmax which produces a distribution over the 1000 class labels. Our network maximizes the multinomial logistic regression objective, which is equivalent to maximizing the average across training cases of the log-probability of the correct label under the prediction distribution.
The kernels of the second, fourth, and fifth convolutional layers are connected only to those kernel maps in the previous layer which reside on the same GPU (see Figure 2). The kernels of the third convolutional layer are connected to all kernel maps in the second layer. The neurons in the fully-connected layers are connected to all neurons in the previous layer. Response-normalization layers follow the first and second convolutional layers. Max-pooling layers, of the kind described in Section 3.4, follow both response-normalization layers as well as the fifth convolutional layer. The ReLU non-linearity is applied to the output of every convolutional and fully-connected layer.
The first convolutional layer filters the 2242243 input image with 96 kernels of size 11113 with a stride of 4 pixels (this is the distance between the receptive field centers of neighboring neurons in a kernel map). The second convolutional layer takes as input the (response-normalized and pooled) output of the first convolutional layer and filters it with 256 kernels of size 5548. The third, fourth, and fifth convolutional layers are connected to one another without any intervening pooling or normalization layers. The third convolutional layer has 384 kernels of size 33256 connected to the (normalized, pooled) outputs of the second convolutional layer. The fourth convolutional layer has 384 kernels of size 33192 , and the fifth convolutional layer has 256 kernels of size 33192. The fully-connected layers have 4096 neurons each.
4 Reducing Overfitting
Our neural network architecture has 60 million parameters. Although the 1000 classes of ILSVRC make each training example impose 10 bits of constraint on the mapping from image to label, this turns out to be insufficient to learn so many parameters without considerable overfitting. Below, we describe the two primary ways in which we combat overfitting.
4.1 Data Augmentation
The easiest and most common method to reduce overfitting on image data is to artificially enlarge the dataset using label-preserving transformations (e.g., [25, 4, 5]). We employ two distinct forms of data augmentation, both of which allow transformed images to be produced from the original images with very little computation, so the transformed images do not need to be stored on disk. In our implementation, the transformed images are generated in Python code on the CPU while the GPU is training on the previous batch of images. So these data augmentation schemes are, in effect, computationally free.
The first form of data augmentation consists of generating image translations and horizontal reflections. We do this by extracting random 224224 patches (and their horizontal reflections) from the 256256 images and training our network on these extracted patches4. This increases the size of our training set by a factor of 2048, though the resulting training examples are, of course, highly interdependent.
Without this scheme, our network suffers from substantial overfitting, which would have forced us to use much smaller networks. At test time, the network makes a prediction by extracting five 224*224 patches (the four corner patches and the center patch) as well as their horizontal reflections (hence ten patches in all), and averaging the predictions made by the network’s softmax layer on the ten patches.
The second form of data augmentation consists of altering the intensities of the RGB channels in training images. Specifically, we perform PCA on the set of RGB pixel values throughout the ImageNet training set. To each training image, we add multiples of the found principal components, with magnitudes proportional to the corresponding eigenvalues times a random variable drawn from a Gaussian with mean zero and standard deviation 0.1. Therefore to each RGB image pixel Ixy =
[IRxy; IGxy; IBxy]T we add the following quantity:
where and are ith eigenvector and eigenvalue of the 3*3 covariance matrix of RGB pixel values, respectively, and i is the aforementioned random variable. Each i is drawn only once for all the pixels of a particular training image until that image is used for training again, at which point it is re-drawn. This scheme approximately captures an important property of natural images, namely, that object identity is invariant to changes in the intensity and color of the illumination. This scheme reduces the top-1 error rate by over 1%.
4.2 Dropout
Combining the predictions of many different models is a very successful way to reduce test errors[1, 3], but it appears to be too expensive for big neural networks that already take several days to train. There is, however, a very efficient version of model combination that only costs about a factor of two during training. The recently-introduced technique, called “dropout” [10], consists of setting to zero the output of each hidden neuron with probability 0.5. The neurons which are “dropped out” in this way do not contribute to the forward pass and do not participate in backpropagation. So every time an input is presented, the neural network samples a different architecture, but all these architectures share weights. This technique reduces complex co-adaptations of neurons, since a neuron cannot rely on the presence of particular other neurons. It is, therefore, forced to learn more robust features that are useful in conjunction with many different random subsets of the other neurons. At test time, we use all the neurons but multiply their outputs by 0.5, which is a reasonable approximation to taking the geometric mean of the predictive distributions produced by the exponentially-many dropout networks.
结合许多不同模型的预测是减少测试错误的一种非常成功的方法[1,3]，但是对于已经需要几天训练的大型神经网络来说，这似乎太昂贵了。然而，有一个非常有效的模型组合版本，它在训练期间只花费大约2倍的成本。最近介绍的技术称为dropout[10]，它将每个隐藏神经元的输出设置为0，概率为0.5。以这种方式丢弃的神经元不参与正向传递，也不参与反向传播。所以每次输入时，神经网络都会对不同的结构进行采样，但是所有这些结构都共享权重。这种技术减少了神经元之间复杂的相互适应，因为神经元不能依赖于特定的其他神经元的存在。因此，它被迫学习与其他神经元的许多不同随机子集结合使用的更健壮的特征。在测试时，我们使用所有的神经元，但将它们的输出乘以0.5，这是一个合理的近似值，近似于取由指数型多退出网络产生的预测分布的几何平均值。
We use dropout in the first two fully-connected layers of Figure 2. Without dropout, our network exhibits substantial overfitting. Dropout roughly doubles the number of iterations required to converge.
我们在图2的前两个完全连接的层中使用了dropout。没有dropout，我们的网络显示出大量的过拟合。Dropout使收敛所需的迭代次数增加了一倍。
5 Details of learning
We trained our models using stochastic gradient descent with a batch size of 128 examples, momentum of 0.9, and weight decay of 0.0005. We found that this small amount of weight decay was important for the model to learn. In other words, weight decay here is not merely a regularizer: it reduces the model’s training error. The update rule for weight w was
where i is the iteration index, v is the momentum variable, is the learning rate, and <> is the average over the ith batch Di of the derivative of the objective with respect to w, evaluated at wi. We initialized the weights in each layer from a zero-mean Gaussian distribution with standard deviation 0.01. We initialized the neuron biases in the second, fourth, and fifth convolutional layers, as well as in the fully-connected hidden layers, with the constant 1. This initialization accelerates the early stages of learning by providing the ReLUs with positive inputs. We initialized the neuron biases in the remaining layers with the constant 0.
We used an equal learning rate for all layers, which we adjusted manually throughout training. The heuristic which we followed was to divide the learning rate by 10 when the validation error rate stopped improving with the current learning rate. The learning rate was initialized at 0.01 and
7 Discussion
Our results show that a large, deep convolutional neural network is capable of achieving record-breaking results on a highly challenging dataset using purely supervised learning. It is notable that our network’s performance degrades if a single convolutional layer is removed. For example, removing any of the middle layers results in a loss of about 2% for the top-1 performance of the network. So the depth really is important for achieving our results.
To simplify our experiments, we did not use any unsupervised pre-training even though we expect that it will help, especially if we obtain enough computational power to significantly increase the size of the network without obtaining a corresponding increase in the amount of labeled data. Thus far, our results have improved as we have made our network larger and trained it longer but we still have many orders of magnitude to go in order to match the inferotemporal pathway of the human visual system. Ultimately we would like to use very large and deep convolutional nets on video sequences where the temporal structure provides very helpful information that is missing or far less obvious in static images.