In this paper, we study normalization methods for neural networks from the perspective of elimination singularity. Elimination singularities correspond to the points on the training trajectory where neurons become consistently deactivated. They cause degenerate manifolds in the loss landscape which will slow down training and harm model performances. We show that channel-based normalizations (e.g. Layer Normalization and Group Normalization) are unable to guarantee a far distance from elimination singularities, in contrast with Batch Normalization which by design avoids models from getting too close to them. To address this issue, we propose BatchChannel Normalization (BCN), which uses batch knowledge to avoid the elimination singularities in the training of channel-normalized models. Unlike Batch Normalization, BCN is able to run in both large-batch and micro-batch training settings. The effectiveness of BCN is verified on many tasks, including image classification, object detection, instance segmentation, and semantic segmentation. The code is here: https://github.com/joe-siyuan-qiao/Batch-Channel-Normalization.
Despite deep convolutional neural networks' great success in object classification, it suffers from severe generalization performance drop under occlusion due to the inconsistency between training and testing data. Because of the large variance of occluders, our goal is a model trained on occlusion-free data while generalizable to occlusion conditions. In this work, we integrate prototypes, partial matching and top-down attention regulation into deep neural networks to realize robust object classification under occlusion. We first introduce prototype learning as its regularization encourages compact data clusters, which enables better generalization ability under inconsistent conditions. Then, attention map at intermediate layer based on feature dictionary and activation scale is estimated for partial matching, which sifts irrelevant information out when comparing features with prototypes. Further, inspired by neuroscience research that reveals the important role of feedback connection for object recognition under occlusion, a top-down feedback attention regulation is introduced into convolution layers, purposefully reducing the contamination by occlusion during feature extraction stage. Our experiment results on partially occluded MNIST and vehicles from the PASCAL3D+ dataset demonstrate that the proposed network significantly improves the robustness of current deep neural networks under occlusion. Our code will be released.
In this paper, we propose Weight Standardization (WS) to accelerate deep network training. WS is targeted at the micro-batch training setting where each GPU typically has only 1-2 images for training. The micro-batch training setting is hard because small batch sizes are not enough for training networks with Batch Normalization (BN), while other normalization methods that do not rely on batch knowledge still have difficulty matching the performances of BN in large-batch training. Our WS ends this problem because when used with Group Normalization and trained with 1 image/GPU, WS is able to match or outperform the performances of BN trained with large batch sizes with only 2 more lines of code. In micro-batch training, WS significantly outperforms other normalization methods. WS achieves these superior results by standardizing the weights in the convolutional layers, which we show is able to smooth the loss landscape by reducing the Lipschitz constants of the loss and the gradients. The effectiveness of WS is verified on many tasks, including image classification, object detection, instance segmentation, video recognition, semantic segmentation, and point cloud recognition. The code is available here: https://github.com/joe-siyuan-qiao/WeightStandardization.
In this paper, we study the problem of improving computational resource utilization of neural networks. Deep neural networks are usually over-parameterized for their tasks in order to achieve good performances, thus are likely to have underutilized computational resources. This observation motivates a lot of research topics, e.g. network pruning, architecture search, etc. As models with higher computational costs (e.g. more parameters or more computations) usually have better performances, we study the problem of improving the resource utilization of neural networks so that their potentials can be further realized. To this end, we propose a novel optimization method named Neural Rejuvenation. As its name suggests, our method detects dead neurons and computes resource utilization in real time, rejuvenates dead neurons by resource reallocation and reinitialization, and trains them with new training schemes. By simply replacing standard optimizers with Neural Rejuvenation, we are able to improve the performances of neural networks by a very large margin while using similar training efforts and maintaining their original resource usages.
Computer vision is difficult, partly because the mathematical function connecting input and output data is often complex, fuzzy and thus hard to learn. A currently popular solution is to design a deep neural network and optimize it on a large-scale dataset. However, as the number of parameters increases, the generalization ability is often not guaranteed, e.g., the model can over-fit due to the limited amount of training data, or fail to converge because the desired function is too difficult to learn. This paper presents an effective framework named progressive recurrent learning (PRL). The core idea is similar to curriculum learning which gradually increases the difficulty of training data. We generalize it to a wide range of vision problems that were previously considered less proper to apply curriculum learning. PRL starts with inserting a recurrent prediction scheme, based on the motivation of feeding the prediction of a vision model to the same model iteratively, so that the auxiliary cues contained in it can be exploited to improve the quality of itself. In order to better optimize this framework, we start with providing perfect prediction, i.e., ground-truth, to the second stage, but gradually replace it with the prediction of the first stage. In the final status, the ground-truth information is not needed any more, so that the entire model works on the real data distribution as in the testing process. We apply PRL to two challenging visual recognition tasks, namely, object localization and semantic segmentation, and demonstrate consistent accuracy gain compared to the baseline training strategy, especially in the scenarios of more difficult vision tasks.
Recent anchor-based deep face detectors have achieved promising performance, but they are still struggling to detect hard faces, such as small, blurred and partially occluded faces. A reason is that they treat all images and faces equally, without putting more effort on hard ones; however, many training images only contain easy faces, which are less helpful to achieve better performance on hard images. In this paper, we propose that the robustness of a face detector against hard faces can be improved by learning small faces on hard images. Our intuitions are (1) hard images are the images which contain at least one hard face, thus they facilitate training robust face detectors; (2) most hard faces are small faces and other types of hard faces can be easily converted to small faces by shrinking. We build an anchor-based deep face detector, which only output a single feature map with small anchors, to specifically learn small faces and train it by a novel hard image mining strategy. Extensive experiments have been conducted on WIDER FACE, FDDB, Pascal Faces, and AFW datasets to show the effectiveness of our method. Our method achieves APs of 95.7, 94.9 and 89.7 on easy, medium and hard WIDER FACE val dataset respectively, which surpass the previous state-of-the-arts, especially on the hard subset. Code and model are available at https://github.com/bairdzhang/smallhardface.
We focus on the problem of training a deep neural network in generations. The flowchart is that, in order to optimize the target network (student), another network (teacher) with the same architecture is first trained, and used to provide part of supervision signals in the next stage. While this strategy leads to a higher accuracy, many aspects (e.g., why teacher-student optimization helps) still need further explorations. This paper studies this problem from a perspective of controlling the strictness in training the teacher network. Existing approaches mostly used a hard distribution (e.g., one-hot vectors) in training, leading to a strict teacher which itself has a high accuracy, but we argue that the teacher needs to be more tolerant, although this often implies a lower accuracy. The implementation is very easy, with merely an extra loss term added to the teacher network, facilitating a few secondary classes to emerge and complement to the primary class. Consequently, the teacher provides a milder supervision signal (a less peaked distribution), and makes it possible for the student to learn from inter-class similarity and potentially lower the risk of over-fitting. Experiments are performed on standard image classification tasks (CIFAR100 and ILSVRC2012). Although the teacher network behaves less powerful, the students show a persistent ability growth and eventually achieve higher classification accuracies than other competitors. Model ensemble and transfer feature extraction also verify the effectiveness of our approach.
We propose a novel single shot object detection network named Detection with Enriched Semantics (DES). Our motivation is to enrich the semantics of object detection features within a typical deep detector, by a semantic segmentation branch and a global activation module. The segmentation branch is supervised by weak segmentation ground-truth, i.e., no extra annotation is required. In conjunction with that, we employ a global activation module which learns relationship between channels and object classes in a self-supervised manner. Comprehensive experimental results on both PASCAL VOC and MS COCO detection datasets demonstrate the effectiveness of the proposed method. In particular, with a VGG16 based DES, we achieve an mAP of 81.7 on VOC2007 test and an mAP of 32.8 on COCO test-dev with an inference speed of 31.5 milliseconds per image on a Titan Xp GPU. With a lower resolution version, we achieve an mAP of 79.7 on VOC2007 with an inference speed of 13.0 milliseconds per image.
Convolution is spatially-symmetric, i.e., the visual features are independent of its position in the image, which limits its ability to utilize contextual cues for visual recognition. This paper addresses this issue by introducing a recalibration process, which refers to the surrounding region of each neuron, computes an importance value and multiplies it to the original neural response. Our approach is named multi-scale spatially-asymmetric recalibration (MS-SAR), which extracts visual cues from surrounding regions at multiple scales, and designs a weighting scheme which is asymmetric in the spatial domain. MS-SAR is implemented in an efficient way, so that only small fractions of extra parameters and computations are required. We apply MS-SAR to several popular building blocks, including the residual block and the densely-connected block, and demonstrate its superior performance in both CIFAR and ILSVRC2012 classification tasks.
In this paper, we study the problem of semi-supervised image recognition, which is to learn classifiers using both labeled and unlabeled images. We present Deep Co-Training, a deep learning based method inspired by the Co-Training framework. The original Co-Training learns two classifiers on two views which are data from different sources that describe the same instances. To extend this concept to deep learning, Deep Co-Training trains multiple deep neural networks to be the different views and exploits adversarial examples to encourage view difference, in order to prevent the networks from collapsing into each other. As a result, the co-trained networks provide different and complementary information about the data, which is necessary for the Co-Training framework to achieve good results. We test our method on SVHN, CIFAR-10/100 and ImageNet datasets, and our method outperforms the previous state-of-the-art methods by a large margin.