In this paper, we propose a novel training strategy for convolutional neural network(CNN) named Feature Mining, that aims to strengthen the networks learning of the local feature. Through experiments, we find that semantic contained in different part
s of the feature is different, while the network will inevitably lose the local information during feedforward propagation. In order to enhance the learning of local feature, Feature Mining divides the complete feature into two complementary parts and reuse these divided feature to make the network learn more local information, we call the two steps as feature segmentation and feature reusing. Feature Mining is a parameter-free method and has plug-and-play nature, and can be applied to any CNN models. Extensive experiments demonstrate the wide applicability, versatility, and compatibility of our method.
Regularization and data augmentation methods have been widely used and become increasingly indispensable in deep learning training. Researchers who devote themselves to this have considered various possibilities. But so far, there has been little dis
cussion about regularizing outputs of the model. This paper begins with empirical observations that better performances are significantly associated with output distributions, that have smaller average values and variances. By audaciously assuming there is causality involved, we propose a novel regularization term, called Output Decay, that enforces the model to assign smaller and similar output values on each class. Though being counter-intuitive, such a small modification result in a remarkable improvement on performance. Extensive experiments demonstrate the wide applicability, versatility, and compatibility of Output Decay.
Self-supervised learning, which benefits from automatically constructing labels through pre-designed pretext task, has recently been applied for strengthen supervised learning. Since previous self-supervised pretext tasks are based on input, they may
incur huge additional training overhead. In this paper we find that features in CNNs can be also used for self-supervision. Thus we creatively design the emph{feature-based pretext task} which requires only a small amount of additional training overhead. In our task we discard different particular regions of features, and then train the model to distinguish these different features. In order to fully apply our feature-based pretext task in supervised learning, we also propose a novel learning framework containing multi-classifiers for further improvement. Original labels will be expanded to joint labels via self-supervision of feature transformations. With more semantic information provided by our self-supervised tasks, this approach can train CNNs more effectively. Extensive experiments on various supervised learning tasks demonstrate the accuracy improvement and wide applicability of our method.
Traditional self-supervised learning requires CNNs using external pretext tasks (i.e., image- or video-based tasks) to encode high-level semantic visual representations. In this paper, we show that feature transformations within CNNs can also be rega
rded as supervisory signals to construct the self-supervised task, called emph{internal pretext task}. And such a task can be applied for the enhancement of supervised learning. Specifically, we first transform the internal feature maps by discarding different channels, and then define an additional internal pretext task to identify the discarded channels. CNNs are trained to predict the joint labels generated by the combination of self-supervised labels and original labels. By doing so, we let CNNs know which channels are missing while classifying in the hope to mine richer feature information. Extensive experiments show that our approach is effective on various models and datasets. And its worth noting that we only incur negligible computational overhead. Furthermore, our approach can also be compatible with other methods to get better results.
The promising performances of CNNs often overshadow the need to examine whether they are doing in the way we are actually interested. We show through experiments that even over-parameterized models would still solve a dataset by recklessly leveraging
spurious correlations, or so-called shortcuts. To combat with this unintended propensity, we borrow the idea of printer test page and propose a novel approach called White Paper Assistance. Our proposed method involves the white paper to detect the extent to which the model has preference for certain characterized patterns and alleviates it by forcing the model to make a random guess on the white paper. We show the consistent accuracy improvements that are manifest in various architectures, datasets and combinations with other techniques. Experiments have also demonstrated the versatility of our approach on fine-grained recognition, imbalanced classification and robustness to corruptions.
In this paper, we propose Selective Output Smoothing Regularization, a novel regularization method for training the Convolutional Neural Networks (CNNs). Inspired by the diverse effects on training from different samples, Selective Output Smoothing R
egularization improves the performance by encouraging the model to produce equal logits on incorrect classes when dealing with samples that the model classifies correctly and over-confidently. This plug-and-play regularization method can be conveniently incorporated into almost any CNN-based project without extra hassle. Extensive experiments have shown that Selective Output Smoothing Regularization consistently achieves significant improvement in image classification benchmarks, such as CIFAR-100, Tiny ImageNet, ImageNet, and CUB-200-2011. Particularly, our method obtains 77.30$%$ accuracy on ImageNet with ResNet-50, which gains 1.1$%$ than baseline (76.2$%$). We also empirically demonstrate the ability of our method to make further improvements when combining with other widely used regularization techniques. On Pascal detection, using the SOSR-trained ImageNet classifier as the pretrained model leads to better detection performances. Moreover, we demonstrate the effectiveness of our method in small sample size problem and imbalanced dataset problem.
In this paper, we propose a new data augmentation strategy named Thumbnail, which aims to strengthen the networks capture of global features. We get a generated image by reducing an image to a certain size, which is called as the thumbnail, and pasti
ng it in the random position of the original image. The generated image not only retains most of the original image information but also has the global information in the thumbnail. Furthermore, we find that the idea of thumbnail can be perfectly integrated with Mixed Sample Data Augmentation, so we paste the thumbnail in another image where the ground truth labels are also mixed with a certain weight, which makes great achievements on various computer vision tasks. Extensive experiments show that Thumbnail works better than the state-of-the-art augmentation strategies across classification, fine-grained image classification, and object detection. On ImageNet classification, ResNet50 architecture with our method achieves 79.21% accuracy, which is more than 2.89% improvement on the baseline.
The shape dependence for the technologically important nickel oxide (NiO) nanocrystals on (001) strontium titanate substrates is investigated under the generalized Wulff-Kaichew (GWK) theorem framework. It is found that the shape of the NiO nanocryst
als is primarily governed by the existence (or absence) of interfacial strain. Nanocrystals that have a fully pseudomorphic interface with the substrate (i.e. the epitaxial strain is not relaxed) form an embedded smooth ball-crown morphology with {001}, {011}, {111} and high-index {113} exposed facets with a negative Wulff point. On the other hand, when the interfacial strain is relaxed by misfit dislocations, the nanocrystals take on a truncated pyramidal shape, bounded by {111} faces and a {001} flat top, with a positive Wulff point. Our quantitative model is able to predict both experimentally observed shapes and sizes with good accuracy. Given the increasing demand for hetero-epitaxial nanocrystals in various physio-chemical and electro-chemical functional devices, these results lay the important groundwork in exploiting the GWK theorem as a general analytical approach to explain hetero-epitaxial nanocrystal growth on oxide substrates governed by interface strain.
Piezoelectrics interconvert mechanical energy and electric charge and are widely used in actuators and sensors. The best performing materials are ferroelectrics at a morphotropic phase boundary (MPB), where several phases can intimately coexist. Swit
ching between these phases by electric field produces a large electromechanical response. In the ferroelectric BiFeO$_3$, strain can be used to create an MPB-like phase mixture and thus to generate large electric field dependent strains. However, this enhanced response occurs at localized, randomly positioned regions of the film, which potentially complicates nanodevice design. Here, we use epitaxial strain and orientation engineering in tandem - anisotropic epitaxy - to craft a hitherto unavailable low-symmetry phase of BiFeO$_3$ which acts as a structural bridge between the rhombohedral-like and tetragonal-like polymorphs. Interferometric displacement sensor measurements and first-principle calculations reveal that under external electric bias, this phase undergoes a transition to the tetragonal-like polymorph, generating a piezoelectric response enhanced by over 200%, and associated giant field-induced reversible strain. These results offer a new route to engineer giant electromechanical properties in thin films, with broader perspectives for other functional oxide systems.
