1.Unsupervised Learning from Video with Deep Neural Embeddings pdf
Because of the rich dynamical structure of videos and their ubiquity in everyday life, it is a natural idea that video data could serve as a powerful unsupervised learning signal for training visual representations in deep neural networks. However, instantiating this idea, especially at large scale, has remained a significant artificial intelligence challenge. Here we present the Video Instance Embedding (VIE) framework, which extends powerful recent unsupervised loss functions for learning deep nonlinear embeddings to multi-stream temporal processing architectures on large-scale video datasets. We show that VIE-trained networks substantially advance the state of the art in unsupervised learning from video datastreams, both for action recognition in the Kinetics dataset, and object recognition in the ImageNet dataset. We show that a hybrid model with both static and dynamic processing pathways is optimal for both transfer tasks, and provide analyses indicating how the pathways differ. Taken in context, our results suggest that deep neural embeddings are a promising approach to unsupervised visual learning across a wide variety of domains.
2.Cerberus: A Multi-headed Derenderer pdf
To generalize to novel visual scenes with new viewpoints and new object poses, a visual system needs representations of the shapes of the parts of an object that are invariant to changes in viewpoint or pose. 3D graphics representations disentangle visual factors such as viewpoints and lighting from object structure in a natural way. It is possible to learn to invert the process that converts 3D graphics representations into 2D images, provided the 3D graphics representations are available as labels. When only the unlabeled images are available, however, learning to derender is much harder. We consider a simple model which is just a set of free floating parts. Each part has its own relation to the camera and its own triangular mesh which can be deformed to model the shape of the part. At test time, a neural network looks at a single image and extracts the shapes of the parts and their relations to the camera. Each part can be viewed as one head of a multi-headed derenderer. During training, the extracted parts are used as input to a differentiable 3D renderer and the reconstruction error is backpropagated to train the neural net. We make the learning task easier by encouraging the deformations of the part meshes to be invariant to changes in viewpoint and invariant to the changes in the relative positions of the parts that occur when the pose of an articulated body changes. Cerberus, our multi-headed derenderer, outperforms previous methods for extracting 3D parts from single images without part annotations, and it does quite well at extracting natural parts of human figures.
3.FireNet: A Specialized Lightweight Fire & Smoke Detection Model for Real-Time IoT Applications pdf
Fire disasters typically result in lot of loss to life and property. It is therefore imperative that precise, fast, and possibly portable solutions to detect fire be made readily available to the masses at reasonable prices. There have been several research attempts to design effective and appropriately priced fire detection systems with varying degrees of success. However, most of them demonstrate a trade-off between performance and model size (which decides the model's ability to be installed on portable devices). The work presented in this paper is an attempt to deal with both the performance and model size issues in one design. Toward that end, a `designed-from-scratch' neural network, named FireNet, is proposed which is worthy on both the counts: (i) it has better performance than existing counterparts, and (ii) it is lightweight enough to be deploy-able on embedded platforms like Raspberry Pi. Performance evaluations on a standard dataset, as well as our own newly introduced custom-compiled fire dataset, are extremely encouraging.
4.An Analysis of Object Embeddings for Image Retrieval pdf
We present an analysis of embeddings extracted from different pre-trained models for content-based image retrieval. Specifically, we study embeddings from image classification and object detection models. We discover that even with additional human annotations such as bounding boxes and segmentation masks, the discriminative power of the embeddings based on modern object detection models is significantly worse than their classification counterparts for the retrieval task. At the same time, our analysis also unearths that object detection model can help retrieval task by acting as a hard attention module for extracting object embeddings that focus on salient region from the convolutional feature map. In order to efficiently extract object embeddings, we introduce a simple guided student-teacher training paradigm for learning discriminative embeddings within the object detection framework. We support our findings with strong experimental results.
5.Compositional Convolutional Networks For Robust Object Classification under Occlusion pdf
Deep convolutional neural networks (DCNNs) are powerful models that yield impressive results at object classification. However, recent work has shown that they do not generalize well to partially occluded objects and to mask attacks. In contrast to DCNNs, compositional models are robust to partial occlusion, however, they are not as discriminative as deep this http URL this work, we integrate DCNNs and compositional object models to retain the best of both approaches: a discriminative model that is robust to partial occlusion and mask attacks. Our model is learned in two steps. First, a standard DCNN is trained for image classification. Subsequently, we cluster the DCNN features into dictionaries. We show that the dictionary components resemble object part detectors and learn the spatial distribution of parts for each object class. We propose mixtures of compositional models to account for large changes in the spatial activation patterns (e.g. due to changes in the 3D pose of an object). At runtime, an image is first classified by the DCNN in a feedforward manner. The prediction uncertainty is used to detect partially occluded objects, which in turn are classified by the compositional model. Our experimental results demonstrate that such compositional convolutional networks resolve a fundamental problem of current deep learning approaches to computer vision: They recognize occluded objects with exceptional performance, even when they have not been exposed to occluded objects during training, while at the same time maintaining high discriminative performance for non-occluded objects.
6.FaceSwapNet: Landmark Guided Many-to-Many Face Reenactment pdf
Recent face reenactment studies have achieved remarkable success either between two identities or in the many-to-one task. However, existing methods have limited scalability when the target person is not a predefined specific identity. To address this limitation, we present a novel many-to-many face reenactment framework, named FaceSwapNet, which allows transferring facial expressions and movements from one source face to arbitrary targets. Our proposed approach is composed of two main modules: the landmark swapper and the landmark-guided generator. Instead of maintaining independent models for each pair of person, the former module uses two encoders and one decoder to adapt anyone's face landmark to target persons. Using the neutral expression of the target person as a reference image, the latter module leverages geometry information from the swapped landmark to generate photo-realistic and emotion-alike images. In addition, a novel triplet perceptual loss is proposed to force the generator to learn geometry and appearance information simultaneously. We evaluate our model on RaFD dataset and the results demonstrate the superior quality of reenacted images as well as the flexibility of transferring facial movements between identities.
7.Hallucinating Optical Flow Features for Video Classification pdf
Appearance and motion are two key components to depict and characterize the video content. Currently, the two-stream models have achieved state-of-the-art performances on video classification. However, extracting motion information, specifically in the form of optical flow features, is extremely computationally expensive, especially for large-scale video classification. In this paper, we propose a motion hallucination network, namely MoNet, to imagine the optical flow features from the appearance features, with no reliance on the optical flow computation. Specifically, MoNet models the temporal relationships of the appearance features and exploits the contextual relationships of the optical flow features with concurrent connections. Extensive experimental results demonstrate that the proposed MoNet can effectively and efficiently hallucinate the optical flow features, which together with the appearance features consistently improve the video classification performances. Moreover, MoNet can help cutting down almost a half of computational and data-storage burdens for the two-stream video classification. Our code is available at: this https URL.
8.Online Filter Clustering and Pruning for Efficient Convnets pdf
Pruning filters is an effective method for accelerating deep neural networks (DNNs), but most existing approaches prune filters on a pre-trained network directly which limits in acceleration. Although each filter has its own effect in DNNs, but if two filters are the same with each other, we could prune one safely. In this paper, we add an extra cluster loss term in the loss function which can force filters in each cluster to be similar online. After training, we keep one filter in each cluster and prune others and fine-tune the pruned network to compensate for the loss. Particularly, the clusters in every layer can be defined firstly which is effective for pruning DNNs within residual blocks. Extensive experiments on CIFAR10 and CIFAR100 benchmarks demonstrate the competitive performance of our proposed filter pruning method.
9.SizeNet: Weakly Supervised Learning of Visual Size and Fit in Fashion Images pdf
Finding clothes that fit is a hot topic in the e-commerce fashion industry. Most approaches addressing this problem are based on statistical methods relying on historical data of articles purchased and returned to the store. Such approaches suffer from the cold start problem for the thousands of articles appearing on the shopping platforms every day, for which no prior purchase history is available. We propose to employ visual data to infer size and fit characteristics of fashion articles. We introduce SizeNet, a weakly-supervised teacher-student training framework that leverages the power of statistical models combined with the rich visual information from article images to learn visual cues for size and fit characteristics, capable of tackling the challenging cold start problem. Detailed experiments are performed on thousands of textile garments, including dresses, trousers, knitwear, tops, etc. from hundreds of different brands.
10.Progressive Learning of Low-Precision Networks pdf
Recent years have witnessed the great advance of deep learning in a variety of vision tasks. Many state-of-the-art deep neural networks suffer from large size and high complexity, which makes it difficult to deploy in resource-limited platforms such as mobile devices.
To this end, low-precision neural networks are widely studied which quantize weights or activations into the low-bit format.
Though being efficient, low-precision networks are usually hard to train and encounter severe accuracy degradation.
In this paper, we propose a new training strategy through expanding low-precision networks during training and removing the expanded parts for network inference.
First, we equip each low-precision convolutional layer with an ancillary full-precision convolutional layer based on a low-precision network structure, which could guide the network to good local minima.
Second, a decay method is introduced to reduce the output of the added full-precision convolution gradually, which keeps the resulted topology structure the same to the original low-precision one.
Experiments on SVHN, CIFAR and ILSVRC-2012 datasets prove that the proposed method can bring faster convergence and higher accuracy for low-precision neural networks.
11.PHT-bot: Deep-Learning based system for automatic risk stratification of COPD patients based upon signs of Pulmonary Hypertension pdf
Chronic Obstructive Pulmonary Disease (COPD) is a leading cause of morbidity and mortality worldwide. Identifying those at highest risk of deterioration would allow more effective distribution of preventative and surveillance resources. Secondary pulmonary hypertension is a manifestation of advanced COPD, which can be reliably diagnosed by the main Pulmonary Artery (PA) to Ascending Aorta (Ao) ratio. In effect, a PA diameter to Ao diameter ratio of greater than 1 has been demonstrated to be a reliable marker of increased pulmonary arterial pressure. Although clinically valuable and readily visualized, the manual assessment of the PA and the Ao diameters is time consuming and under-reported. The present study describes a non invasive method to measure the diameters of both the Ao and the PA from contrast-enhanced chest Computed Tomography (CT). The solution applies deep learning techniques in order to select the correct axial slice to measure, and to segment both arteries. The system achieves test Pearson correlation coefficient scores of 93% for the Ao and 92% for the PA. To the best of our knowledge, it is the first such fully automated solution.
12.A Cost Efficient Approach to Correct OCR Errors in Large Document Collections pdf
Word error rate of an ocr is often higher than its character error rate. This is especially true when ocrs are designed by recognizing characters. High word accuracies are critical to tasks like the creation of content in digital libraries and text-to-speech applications. In order to detect and correct the misrecognised words, it is common for an ocr module to employ a post-processor to further improve the word accuracy. However, conventional approaches to post-processing like looking up a dictionary or using a statistical language model (slm), are still limited. In many such scenarios, it is often required to remove the outstanding errors manually. We observe that the traditional post-processing schemes look at error words sequentially since ocrs process documents one at a time. We propose a cost-efficient model to address the error words in batches rather than correcting them individually. We exploit the fact that a collection of documents, unlike a single document, has a structure leading to repetition of words. Such words, if efficiently grouped together and corrected as a whole can lead to a significant reduction in the cost. Correction can be fully automatic or with a human in the loop. Towards this, we employ a novel clustering scheme to obtain fairly homogeneous clusters. We compare the performance of our model with various baseline approaches including the case where all the errors are removed by a human. We demonstrate the efficacy of our solution empirically by reporting more than 70% reduction in the human effort with near perfect error correction. We validate our method on Books from multiple languages.
13.Cross-Domain Transferability of Adversarial Perturbations pdf
Adversarial examples reveal the blind spots of deep neural networks (DNNs) and represent a major concern for security-critical applications. The transferability of adversarial examples makes real-world attacks possible in black-box settings, where the attacker is forbidden to access the internal parameters of the model. The underlying assumption in most adversary generation methods, whether learning an instance-specific or an instance-agnostic perturbation, is the direct or indirect reliance on the original domain-specific data distribution. In this work, for the first time, we demonstrate the existence of domain-invariant adversaries, thereby showing common adversarial space among different datasets and models. To this end, we propose a framework capable of launching highly transferable attacks that crafts adversarial patterns to mislead networks trained on wholly different domains. For instance, an adversarial function learned on Paintings, Cartoons or Medical images can successfully perturb ImageNet samples to fool the classifier, with success rates as high as $\sim$99% (
$\ell_{\infty} \le 10$ ). The core of our proposed adversarial function is a generative network that is trained using a relativistic supervisory signal that enables domain-invariant perturbations. Our approach sets the new state-of-the-art for fooling rates, both under the white-box and black-box scenarios. Furthermore, despite being an instance-agnostic perturbation function, our attack outperforms the conventionally much stronger instance-specific attack methods.
14.Integrated Neural Network and Machine Vision Approach For Leather Defect Classification pdf
Leather is a type of natural, durable, flexible, soft, supple and pliable material with smooth texture. It is commonly used as a raw material to manufacture luxury consumer goods for high-end customers. To ensure good quality control on the leather products, one of the critical processes is the visual inspection step to spot the random defects on the leather surfaces and it is usually conducted by experienced experts. This paper presents an automatic mechanism to perform the leather defect classification. In particular, we focus on detecting tick-bite defects on a specific type of calf leather. Both the handcrafted feature extractors (i.e., edge detectors and statistical approach) and data-driven (i.e., artificial neural network) methods are utilized to represent the leather patches. Then, multiple classifiers (i.e., decision trees, Support Vector Machines, nearest neighbour and ensemble classifiers) are exploited to determine whether the test sample patches contain defective segments. Using the proposed method, we managed to get a classification accuracy rate of 84% from a sample of approximately 2500 pieces of 400 * 400 leather patches.
15.Invertible generative models for inverse problems: mitigating representation error and dataset bias pdf
Trained generative models have shown remarkable performance as priors for inverse problems in imaging. For example, Generative Adversarial Network priors permit recovery of test images from 5-10x fewer measurements than sparsity priors. Unfortunately, these models may be unable to represent any particular image because of architectural choices, mode collapse, and bias in the training dataset. In this paper, we demonstrate that invertible neural networks, which have zero representation error by design, can be effective natural signal priors at inverse problems such as denoising, compressive sensing, and inpainting. Given a trained generative model, we study the empirical risk formulation of the desired inverse problem under a regularization that promotes high likelihood images, either directly by penalization or algorithmically by initialization. For compressive sensing, invertible priors can yield higher accuracy than sparsity priors across almost all undersampling ratios. For the same accuracy on test images, they can use 10-20x fewer measurements. We demonstrate that invertible priors can yield better reconstructions than sparsity priors for images that have rare features of variation within the biased training set, including out-of-distribution natural images.
16.OICSR: Out-In-Channel Sparsity Regularization for Compact Deep Neural Networks pdf
Channel pruning can significantly accelerate and compress deep neural networks. Many channel pruning works utilize structured sparsity regularization to zero out all the weights in some channels and automatically obtain structure-sparse network in training stage. However, these methods apply structured sparsity regularization on each layer separately where the correlations between consecutive layers are omitted. In this paper, we first combine one out-channel in current layer and the corresponding in-channel in next layer as a regularization group, namely out-in-channel. Our proposed Out-In-Channel Sparsity Regularization (OICSR) considers correlations between successive layers to further retain predictive power of the compact network. Training with OICSR thoroughly transfers discriminative features into a fraction of out-in-channels. Correspondingly, OICSR measures channel importance based on statistics computed from two consecutive layers, not individual layer. Finally, a global greedy pruning algorithm is designed to remove redundant out-in-channels in an iterative way. Our method is comprehensively evaluated with various CNN architectures including CifarNet, AlexNet, ResNet, DenseNet and PreActSeNet on CIFAR-10, CIFAR-100 and ImageNet-1K datasets. Notably, on ImageNet-1K, we reduce 37.2% FLOPs on ResNet-50 while outperforming the original model by 0.22% top-1 accuracy.
17.The Nipple-Areola Complex for Criminal Identification pdf
In digital and multimedia forensics, identification of child sexual offenders based on digital evidence images is highly challenging due to the fact that the offender's face or other obvious characteristics such as tattoos are occluded, covered, or not visible at all. Nevertheless, other naked body parts, e.g., chest are still visible. Some researchers proposed skin marks, skin texture, vein or androgenic hair patterns for criminal and victim identification. There are no available studies of nipple-areola complex (NAC) for offender identification. In this paper, we present a study of offender identification based on the NAC, and we present NTU-Nipple-v1 dataset, which contains 2732 images of 428 different male nipple-areolae. Popular deep learning and hand-crafted recognition methods are evaluated on the provided dataset. The results indicate that the NAC can be a useful characteristic for offender identification.
18.Image Deformation Meta-Networks for One-Shot Learning pdf
Humans can robustly learn novel visual concepts even when images undergo various deformations and loose certain information. Mimicking the same behavior and synthesizing deformed instances of new concepts may help visual recognition systems perform better one-shot learning, i.e., learning concepts from one or few examples. Our key insight is that, while the deformed images may not be visually realistic, they still maintain critical semantic information and contribute significantly to formulating classifier decision boundaries. Inspired by the recent progress of meta-learning, we combine a meta-learner with an image deformation sub-network that produces additional training examples, and optimize both models in an end-to-end manner. The deformation sub-network learns to deform images by fusing a pair of images -- a probe image that keeps the visual content and a gallery image that diversifies the deformations. We demonstrate results on the widely used one-shot learning benchmarks (miniImageNet and ImageNet 1K Challenge datasets), which significantly outperform state-of-the-art approaches.
19.LatentGNN: Learning Efficient Non-local Relations for Visual Recognition pdf
Capturing long-range dependencies in feature representations is crucial for many visual recognition tasks. Despite recent successes of deep convolutional networks, it remains challenging to model non-local context relations between visual features. A promising strategy is to model the feature context by a fully-connected graph neural network (GNN), which augments traditional convolutional features with an estimated non-local context representation. However, most GNN-based approaches require computing a dense graph affinity matrix and hence have difficulty in scaling up to tackle complex real-world visual problems. In this work, we propose an efficient and yet flexible non-local relation representation based on a novel class of graph neural networks. Our key idea is to introduce a latent space to reduce the complexity of graph, which allows us to use a low-rank representation for the graph affinity matrix and to achieve a linear complexity in computation. Extensive experimental evaluations on three major visual recognition tasks show that our method outperforms the prior works with a large margin while maintaining a low computation cost.
20.Union Visual Translation Embedding for Visual Relationship Detection and Scene Graph Generation pdf
Relations amongst entities play a central role in image understanding. Due to the combinatorial complexity of modeling (subject, predicate, object) relation triplets, it is crucial to develop a method that can not only recognize seen relations, but also generalize well to unseen cases. Inspired by Visual Translation Embedding network (VTransE), we propose the Union Visual Translation Embedding network (UVTransE) to capture both common and rare relations with better accuracy. UVTransE maps the subject, the object, and the union (subject, object) image regions into a low-dimensional relation space where a predicate can be expressed as a vector subtraction, such that predicate
$\approx$ union (subject, object)$-$ subject$-$ object. We present a comprehensive evaluation of our method on multiple challenging benchmarks: the Visual Relationship Detection dataset (VRD); UnRel dataset for rare and unusual relations; two subsets of Visual Genome; and the Open Images Challenge. Our approach decisively outperforms VTransE and comes close to or exceeds the state of the art across a range of settings, from small-scale to large-scale datasets, from common to previously unseen relations. On Visual Genome and Open Images, it also achieves promising results on the recently introduced task of scene graph generation.
21.Local Label Propagation for Large-Scale Semi-Supervised Learning pdf
A significant issue in training deep neural networks to solve supervised learning tasks is the need for large numbers of labelled datapoints. The goal of semi-supervised learning is to leverage ubiquitous unlabelled data, together with small quantities of labelled data, to achieve high task performance. Though substantial recent progress has been made in developing semi-supervised algorithms that are effective for comparatively small datasets, many of these techniques do not scale readily to the large (unlaballed) datasets characteristic of real-world applications. In this paper we introduce a novel approach to scalable semi-supervised learning, called Local Label Propagation (LLP). Extending ideas from recent work on unsupervised embedding learning, LLP first embeds datapoints, labelled and otherwise, in a common latent space using a deep neural network. It then propagates pseudolabels from known to unknown datapoints in a manner that depends on the local geometry of the embedding, taking into account both inter-point distance and local data density as a weighting on propagation likelihood. The parameters of the deep embedding are then trained to simultaneously maximize pseudolabel categorization performance as well as a metric of the clustering of datapoints within each psuedo-label group, iteratively alternating stages of network training and label propagation. We illustrate the utility of the LLP method on the ImageNet dataset, achieving results that outperform previous state-of-the-art scalable semi-supervised learning algorithms by large margins, consistently across a wide variety of training regimes. We also show that the feature representation learned with LLP transfers well to scene recognition in the Places 205 dataset.
22.Improving Action Localization by Progressive Cross-stream Cooperation pdf
Spatio-temporal action localization consists of three levels of tasks: spatial localization, action classification, and temporal segmentation. In this work, we propose a new Progressive Cross-stream Cooperation (PCSC) framework to use both region proposals and features from one stream (i.e. Flow/RGB) to help another stream (i.e. RGB/Flow) to iteratively improve action localization results and generate better bounding boxes in an iterative fashion. Specifically, we first generate a larger set of region proposals by combining the latest region proposals from both streams, from which we can readily obtain a larger set of labelled training samples to help learn better action detection models. Second, we also propose a new message passing approach to pass information from one stream to another stream in order to learn better representations, which also leads to better action detection models. As a result, our iterative framework progressively improves action localization results at the frame level. To improve action localization results at the video level, we additionally propose a new strategy to train class-specific actionness detectors for better temporal segmentation, which can be readily learnt by focusing on "confusing" samples from the same action class. Comprehensive experiments on two benchmark datasets UCF-101-24 and J-HMDB demonstrate the effectiveness of our newly proposed approaches for spatio-temporal action localization in realistic scenarios.
23.JGAN: A Joint Formulation of GAN for Synthesizing Images and Labels pdf
Image generation with explicit condition or label generally works better than unconditional image generation. In modern GAN frameworks, both generator and discriminator are formulated to model the conditional distribution of images given with labels. In this paper, we provide an alternative formulation of GAN which models joint distribution of images and labels. There are two advantages in this joint formulation over conditional approaches. The first advantage is that the joint formulation is more robust to label noises, and the second is we can use any kind of weak labels (or additional information which has dependence on the original image data) to enhance unconditional image generation. We will show the effectiveness of joint formulation in CIFAR-10, CIFAR-100, and STL dataset.
24.Case-Based Histopathological Malignancy Diagnosis using Convolutional Neural Networks pdf
In practice, histopathological diagnosis of tumor malignancy often requires a human expert to scan through histopathological images at multiple magnification levels, after which a final diagnosis can be accurately determined. However, previous research on such classification tasks using convolutional neural networks primarily determine a diagnosis for a single magnification level. In this paper, we propose a case-based approach using deep residual neural networks for histopathological malignancy diagnosis, where a case is defined as a sequence of images from the patient at all available levels of magnification. Effectively, through mimicking what a human expert would actually do, our approach makes a diagnosis decision based on features learned in combination at multiple magnification levels. Our results show that the case-based approach achieves better performance than the state-of-the-art methods when evaluated on BreaKHis, a histopathological image dataset for breast tumors.
25.Road Segmentation with Image-LiDAR Data Fusion pdf
Robust road segmentation is a key challenge in self-driving research. Though many image-based methods have been studied and high performances in dataset evaluations have been reported, developing robust and reliable road segmentation is still a major challenge. Data fusion across different sensors to improve the performance of road segmentation is widely considered an important and irreplaceable solution. In this paper, we propose a novel structure to fuse image and LiDAR point cloud in an end-to-end semantic segmentation network, in which the fusion is performed at decoder stage instead of at, more commonly, encoder stage. During fusion, we improve the multi-scale LiDAR map generation to increase the precision of the multi-scale LiDAR map by introducing pyramid projection method. Additionally, we adapted the multi-path refinement network with our fusion strategy and improve the road prediction compared with transpose convolution with skip layers. Our approach has been tested on KITTI ROAD dataset and has competitive performance.
26.Jointly Learning Structured Analysis Discriminative Dictionary and Analysis Multiclass Classifier pdf
In this paper, we propose an analysis mechanism based structured Analysis Discriminative Dictionary Learning (ADDL) framework. ADDL seamlessly integrates the analysis discriminative dictionary learning, analysis representation and analysis classifier training into a unified model. The applied analysis mechanism can make sure that the learnt dictionaries, representations and linear classifiers over different classes are independent and discriminating as much as possible. The dictionary is obtained by minimizing a reconstruction error and an analytical incoherence promoting term that encourages the sub-dictionaries associated with different classes to be independent. To obtain the representation coefficients, ADDL imposes a sparse l2,1-norm constraint on the coding coefficients instead of using l0 or l1-norm, since the l0 or l1-norm constraint applied in most existing DL criteria makes the training phase time consuming. The codes-extraction projection that bridges data with the sparse codes by extracting special features from the given samples is calculated via minimizing a sparse codes approximation term. Then we compute a linear classifier based on the approximated sparse codes by an analysis mechanism to simultaneously consider the classification and representation powers. Thus, the classification approach of our model is very efficient, because it can avoid the extra time-consuming sparse reconstruction process with trained dictionary for each new test data as most existing DL algorithms. Simulations on real image databases demonstrate that our ADDL model can obtain superior performance over other state-of-the-arts.
27.Semantic Fisher Scores for Task Transfer: Using Objects to Classify Scenes pdf
The transfer of a neural network (CNN) trained to recognize objects to the task of scene classification is considered. A Bag-of-Semantics (BoS) representation is first induced, by feeding scene image patches to the object CNN, and representing the scene image by the ensuing bag of posterior class probability vectors (semantic posteriors). The encoding of the BoS with a Fisher vector(FV) is then studied. A link is established between the FV of any probabilistic model and the Q-function of the expectation-maximization(EM) algorithm used to estimate its parameters by maximum likelihood. A network implementation of the MFA Fisher Score (MFA-FS), denoted as the MFAFSNet, is finally proposed to enable end-to-end training. Experiments with various object CNNs and datasets show that the approach has state-of-the-art transfer performance. Somewhat surprisingly, the scene classification results are superior to those of a CNN explicitly trained for scene classification, using a large scene dataset (Places). This suggests that holistic analysis is insufficient for scene classification. The modeling of local object semantics appears to be at least equally important. The two approaches are also shown to be strongly complementary, leading to very large scene classification gains when combined, and outperforming all previous scene classification approaches by a sizeable margin
28.CGaP: Continuous Growth and Pruning for Efficient Deep Learning pdf
Today a canonical approach to reduce the computation cost of Deep Neural Networks (DNNs) is to pre-define an over-parameterized model before training to guarantee the learning capacity, and then prune unimportant learning units (filters and neurons) during training to improve model compactness. We argue it is unnecessary to introduce redundancy at the beginning of the training but then reduce redundancy for the ultimate inference model. In this paper, we propose a Continuous Growth and Pruning (CGaP) scheme to minimize the redundancy from the beginning. CGaP starts the training from a small network seed, then expands the model continuously by reinforcing important learning units, and finally prunes the network to obtain a compact and accurate model. As the growth phase favors important learning units, CGaP provides a clear learning purpose to the pruning phase. Experimental results on representative datasets and DNN architectures demonstrate that CGaP outperforms previous pruning-only approaches that deal with pre-defined structures. For VGG-19 on CIFAR-100 and SVHN datasets, CGaP reduces the number of parameters by 78.9% and 85.8%, FLOPs by 53.2% and 74.2%, respectively; For ResNet-110 On CIFAR-10, CGaP reduces 64.0% number of parameters and 63.3% FLOPs.
29.Enhancing Salient Object Segmentation Through Attention pdf
Segmenting salient objects in an image is an important vision task with ubiquitous applications. The problem becomes more challenging in the presence of a cluttered and textured background, low resolution and/or low contrast images. Even though existing algorithms perform well in segmenting most of the object(s) of interest, they often end up segmenting false positives due to resembling salient objects in the background. In this work, we tackle this problem by iteratively attending to image patches in a recurrent fashion and subsequently enhancing the predicted segmentation mask. Saliency features are estimated independently for every image patch, which are further combined using an aggregation strategy based on a Convolutional Gated Recurrent Unit (ConvGRU) network. The proposed approach works in an end-to-end manner, removing background noise and false positives incrementally. Through extensive evaluation on various benchmark datasets, we show superior performance to the existing approaches without any post-processing.
30.Shape Evasion: Preventing Body Shape Inference of Multi-Stage Approaches pdf
Modern approaches to pose and body shape estimation have recently achieved strong performance even under challenging real-world conditions. Even from a single image of a clothed person, a realistic looking body shape can be inferred that captures a users' weight group and body shape type well. This opens up a whole spectrum of applications -- in particular in fashion -- where virtual try-on and recommendation systems can make use of these new and automatized cues. However, a realistic depiction of the undressed body is regarded highly private and therefore might not be consented by most people. Hence, we ask if the automatic extraction of such information can be effectively evaded. While adversarial perturbations have been shown to be effective for manipulating the output of machine learning models -- in particular, end-to-end deep learning approaches -- state of the art shape estimation methods are composed of multiple stages. We perform the first investigation of different strategies that can be used to effectively manipulate the automatic shape estimation while preserving the overall appearance of the original image.
31.End-to-End Pore Extraction and Matching in Latent Fingerprints: Going Beyond Minutiae pdf
Latent fingerprint recognition is not a new topic but it has attracted a lot of attention from researchers in both academia and industry over the past 50 years. With the rapid development of pattern recognition techniques, automated fingerprint identification systems (AFIS) have become more and more ubiquitous. However, most AFIS are utilized for live-scan or rolled/slap prints while only a few systems can work on latent fingerprints with reasonable accuracy. The question of whether taking higher resolution scans of latent fingerprints and their rolled/slap mate prints could help improve the identification accuracy still remains an open question in the forensic community. Because pores are one of the most reliable features besides minutiae to identify latent fingerprints, we propose an end-to-end automatic pore extraction and matching system to analyze the utility of pores in latent fingerprint identification. Hence, this paper answers two questions in the latent fingerprint domain: (i) does the incorporation of pores as level-3 features improve the system performance significantly? and (ii) does the 1,000 ppi image resolution improve the recognition results? We believe that our proposed end-to-end pore extraction and matching system will be a concrete baseline for future latent AFIS development.
32.A Symmetric Encoder-Decoder with Residual Block for Infrared and Visible Image Fusion pdf
In computer vision and image processing tasks, image fusion has evolved into an attractive research field. However, recent existing image fusion methods are mostly built on pixel-level operations, which may produce unacceptable artifacts and are time-consuming. In this paper, a symmetric encoder-decoder with a residual block (SEDR) for infrared and visible image fusion is proposed. For the training stage, the SEDR network is trained with a new dataset to obtain a fixed feature extractor. For the fusion stage, first, the trained model is utilized to extract the intermediate features and compensation features of two source images. Then, extracted intermediate features are used to generate two attention maps, which are multiplied to the input features for refinement. In addition, the compensation features generated by the first two convolutional layers are merged and passed to the corresponding deconvolutional layers. At last, the refined features are fused for decoding to reconstruct the final fused image. Experimental results demonstrate that the proposed fusion method (named as SEDRFuse) outperforms the state-of-the-art fusion methods in terms of both subjective and objective evaluations.
33.ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation pdf
Deep neural networks are vulnerable to adversarial attacks. The literature is rich with algorithms that can easily craft successful adversarial examples. In contrast, the performance of defense techniques still lags behind. This paper proposes ME-Net, a defense method that leverages matrix estimation (ME). In ME-Net, images are preprocessed using two steps: first pixels are randomly dropped from the image; then, the image is reconstructed using ME. We show that this process destroys the adversarial structure of the noise, while re-enforcing the global structure in the original image. Since humans typically rely on such global structures in classifying images, the process makes the network mode compatible with human perception. We conduct comprehensive experiments on prevailing benchmarks such as MNIST, CIFAR-10, SVHN, and Tiny-ImageNet. Comparing ME-Net with state-of-the-art defense mechanisms shows that ME-Net consistently outperforms prior techniques, improving robustness against both black-box and white-box attacks.
34.EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks pdf
Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. Based on this observation, we propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient. We demonstrate the effectiveness of this method on scaling up MobileNets and ResNet.
To go even further, we use neural architecture search to design a new baseline network and scale it up to obtain a family of models, called EfficientNets, which achieve much better accuracy and efficiency than previous ConvNets. In particular, our EfficientNet-B7 achieves state-of-the-art 84.4% top-1 / 97.1% top-5 accuracy on ImageNet, while being 8.4x smaller and 6.1x faster on inference than the best existing ConvNet. Our EfficientNets also transfer well and achieve state-of-the-art accuracy on CIFAR-100 (91.7%), Flowers (98.8%), and 3 other transfer learning datasets, with an order of magnitude fewer parameters. Source code is at this https URL.
35.A Compact Representation of Histopathology Images using Digital Stain Separation & Frequency-Based Encoded Local Projections pdf
In recent years, histopathology images have been increasingly used as a diagnostic tool in the medical field. The process of accurately diagnosing a biopsy sample requires significant expertise in the field, and as such can be time-consuming and is prone to uncertainty and error. With the advent of digital pathology, using image recognition systems to highlight problem areas or locate similar images can aid pathologists in making quick and accurate diagnoses. In this paper, we specifically consider the encoded local projections (ELP) algorithm, which has previously shown some success as a tool for classification and recognition of histopathology images. We build on the success of the ELP algorithm as a means for image classification and recognition by proposing a modified algorithm which captures the local frequency information of the image. The proposed algorithm estimates local frequencies by quantifying the changes in multiple projections in local windows of greyscale images. By doing so we remove the need to store the full projections, thus significantly reducing the histogram size, and decreasing computation time for image retrieval and classification tasks. Furthermore, we investigate the effectiveness of applying our method to histopathology images which have been digitally separated into their hematoxylin and eosin stain components. The proposed algorithm is tested on the publicly available invasive ductal carcinoma (IDC) data set. The histograms are used to train an SVM to classify the data. The experiments showed that the proposed method outperforms the original ELP algorithm in image retrieval tasks. On classification tasks, the results are found to be comparable to state-of-the-art deep learning methods and better than many handcrafted features from the literature.
36.Adversarial Domain Adaptation Being Aware of Class Relationships pdf
Adversarial training is a useful approach to promote the learning of transferable representations across the source and target domains, which has been widely applied for domain adaptation (DA) tasks based on deep neural networks. Until very recently, existing adversarial domain adaptation (ADA) methods ignore the useful information from the label space, which is an important factor accountable for the complicated data distributions associated with different semantic classes. Especially, the inter-class semantic relationships have been rarely considered and discussed in the current work of transfer learning. In this paper, we propose a novel relationship-aware adversarial domain adaptation (RADA) algorithm, which first utilizes a single multi-class domain discriminator to enforce the learning of inter-class dependency structure during domain-adversarial training and then aligns this structure with the inter-class dependencies that are characterized from training the label predictor on the source domain. Specifically, we impose a regularization term to penalize the structure discrepancy between the inter-class dependencies respectively estimated from domain discriminator and label predictor. Through this alignment, our proposed method makes the ADA aware of class relationships. Empirical studies show that the incorporation of class relationships significantly improves the performance on benchmark datasets.
37.Network Deconvolution pdf
Convolution is a central operation in Convolutional Neural Networks (CNNs), which applies a kernel or mask to overlapping regions shifted across the image. In this work we show that the underlying kernels are trained with highly correlated data, which leads to co-adaptation of model weights. To address this issue we propose what we call network deconvolution, a procedure that aims to remove pixel-wise and channel-wise correlations before the data is fed into each layer. We show that by removing this correlation we are able to achieve better convergence rates during model training with superior results without the use of batch normalization on the CIFAR-10, CIFAR-100, MNIST, Fashion-MNIST datasets, as well as against reference models from "model zoo" on the ImageNet standard benchmark.
38.BreizhCrops: A Satellite Time Series Dataset for Crop Type Identification pdf
This dataset challenges the time series community with the task of satellite-based vegetation identification on large scale real-world dataset of satellite data acquired during one entire year. It consists of time series data with associated crop types from 580k field parcels in Brittany, France (Breizh in local language). Along with this dataset, we provide results and code of a Long Short-Term Memory network and Transformer network as baselines. We release dataset, along with preprocessing scripts and baseline models in this https URL and encourage methodical researchers to benchmark and develop novel methods applied to satellite-based crop monitoring.
39.Video-based Person Re-identification with Two-stream Convolutional Network and Co-attentive Snippet Embedding pdf
Recently, the applications of person re-identification in visual surveillance and human-computer interaction are sharply increasing, which signifies the critical role of such a problem. In this paper, we propose a two-stream convolutional network (ConvNet) based on the competitive similarity aggregation scheme and co-attentive embedding strategy for video-based person re-identification. By dividing the long video sequence into multiple short video snippets, we manage to utilize every snippet's RGB frames, optical flow maps and pose maps to facilitate residual networks, e.g., ResNet, for feature extraction in the two-stream ConvNet. The extracted features are embedded by the co-attentive embedding method, which allows for the reduction of the effects of noisy frames. Finally, we fuse the outputs of both streams as the embedding of a snippet, and apply competitive snippet-similarity aggregation to measure the similarity between two sequences. Our experiments show that the proposed method significantly outperforms current state-of-the-art approaches on multiple datasets.
40.Snooping Attacks on Deep Reinforcement Learning pdf
Adversarial attacks have exposed a significant security vulnerability in state-of-the-art machine learning models. Among these models include deep reinforcement learning agents. The existing methods for attacking reinforcement learning agents assume the adversary either has access to the target agent's learned parameters or the environment that the agent interacts with. In this work, we propose a new class of threat models, called snooping threat models, that are unique to reinforcement learning. In these snooping threat models, the adversary does not have the ability to personally interact with the environment, and can only eavesdrop on the action and reward signals being exchanged between agent and environment. We show that adversaries operating in these highly constrained threat models can still launch devastating attacks against the target agent by training proxy models on related tasks and leveraging the transferability of adversarial examples.
41.Importance of user inputs while using incremental learning to personalize human activity recognition models pdf
In this study, importance of user inputs is studied in the context of personalizing human activity recognition models using incremental learning. Inertial sensor data from three body positions are used, and the classification is based on Learn++ ensemble method. Three different approaches to update models are compared: non-supervised, semi-supervised and supervised. Non-supervised approach relies fully on predicted labels, supervised fully on user labeled data, and the proposed method for semi-supervised learning, is a combination of these two. In fact, our experiments show that by relying on predicted labels with high confidence, and asking the user to label only uncertain observations (from 12% to 26% of the observations depending on the used base classifier), almost as low error rates can be achieved as by using supervised approach. In fact, the difference was less than 2%-units. Moreover, unlike non-supervised approach, semi-supervised approach does not suffer from drastic concept drift, and thus, the error rate of the non-supervised approach is over 5%-units higher than using semi-supervised approach.
42.Deep Scale-spaces: Equivariance Over Scale pdf
We introduce deep scale-spaces (DSS), a generalization of convolutional neural networks, exploiting the scale symmetry structure of conventional image recognition tasks. Put plainly, the class of an image is invariant to the scale at which it is viewed. We construct scale equivariant cross-correlations based on a principled extension of convolutions, grounded in the theory of scale-spaces and semigroups. As a very basic operation, these cross-correlations can be used in almost any modern deep learning architecture in a plug-and-play manner. We demonstrate our networks on the Patch Camelyon and Cityscapes datasets, to prove their utility and perform introspective studies to further understand their properties.
43.Learning Dynamics of Attention: Human Prior for Interpretable Machine Reasoning pdf
Without relevant human priors, neural networks may learn uninterpretable features. We propose Dynamics of Attention for Focus Transition (DAFT) as a human prior for machine reasoning. DAFT is a novel method that regularizes attention-based reasoning by modelling it as a continuous dynamical system using neural ordinary differential equations. As a proof of concept, we augment a state-of-the-art visual reasoning model with DAFT. Our experiments reveal that applying DAFT yields similar performance to the original model while using fewer reasoning steps, showing that it implicitly learns to skip unnecessary steps. We also propose a new metric, Total Length of Transition (TLT), which represents the effective reasoning step size by quantifying how much a given model's focus drifts while reasoning about a question. We show that adding DAFT results in lower TLT, demonstrating that our method indeed obeys the human prior towards shorter reasoning paths in addition to producing more interpretable attention maps.
44.Discrete Infomax Codes for Meta-Learning pdf
Learning compact discrete representations of data is itself a key task in addition to facilitating subsequent processing. It is also relevant to meta-learning since a latent representation shared across relevant tasks enables a model to adapt to new tasks quickly. In this paper, we present a method for learning a stochastic encoder that yields discrete p-way codes of length d by maximizing the mutual information between representations and labels. We show that previous loss functions for deep metric learning are approximations to this information-theoretic objective function. Our model, Discrete InfoMax Codes (DIMCO), learns to produce a short representation of data that can be used to classify classes with few labeled examples. Our analysis shows that using shorter codes reduces overfitting in the context of few-shot classification. Experiments show that DIMCO requires less memory (i.e., code length) for performance similar to previous methods and that our method is particularly effective when the training dataset is small.
45.Adaptive Lighting for Data-Driven Non-Line-of-Sight 3D Localization and Object Identification pdf
Non-line-of-sight (NLOS) imaging of objects not visible to either the camera or illumination source is a challenging task with vital applications including surveillance and robotics. Recent NLOS reconstruction advances have been achieved using time-resolved measurements which requires expensive and specialized detectors and laser sources. In contrast, we propose a data-driven approach for NLOS 3D localization requiring only a conventional camera and projector. We achieve an average identification of 79% object identification for three classes of objects, and localization of the NLOS object's centroid for a mean-squared error (MSE) of 2.89cm in the occluded region for real data taken from a hardware prototype. To generalize to line-of-sight (LOS) scenes with non-planar surfaces, we introduce an adaptive lighting algorithm. This algorithm, based on radiosity, identifies and illuminates scene patches in the LOS which most contribute to the NLOS light paths, and can factor in system power constraints. We further improve our average NLOS object identification to 87.8% accuracy and localization to 1.94cm MSE on a complex LOS scene using adaptive lighting for real data, demonstrating the advantage of combining the physics of light transport with active illumination for data-driven NLOS imaging.
46.Label Universal Targeted Attack pdf
We introduce Label Universal Targeted Attack (LUTA) that makes a deep model predict a label of attacker's choice for `any' sample of a given source class with high probability. Our attack stochastically maximizes the log-probability of the target label for the source class with first order gradient optimization, while accounting for the gradient moments. It also suppresses the leakage of attack information to the non-source classes for avoiding the attack suspicions. The perturbations resulting from our attack achieve high fooling ratios on the large-scale ImageNet and VGGFace models, and transfer well to the Physical World. Given full control over the perturbation scope in LUTA, we also demonstrate it as a tool for deep model autopsy. The proposed attack reveals interesting perturbation patterns and observations regarding the deep models.
47.Improved Training Speed, Accuracy, and Data Utilization Through Loss Function Optimization pdf
As the complexity of neural network models has grown, it has become increasingly important to optimize their design automatically through metalearning. Methods for discovering hyperparameters, topologies, and learning rate schedules have lead to significant increases in performance. This paper shows that loss functions can be optimized with metalearning as well, and result in similar improvements. The method, Genetic Loss-function Optimization (GLO), discovers loss functions de novo, and optimizes them for a target task. Leveraging techniques from genetic programming, GLO builds loss functions hierarchically from a set of operators and leaf nodes. These functions are repeatedly recombined and mutated to find an optimal structure, and then a covariance-matrix adaptation evolutionary strategy (CMA-ES) is used to find optimal coefficients. Networks trained with GLO loss functions are found to outperform the standard cross-entropy loss on standard image classification tasks. Training with these new loss functions requires fewer steps, results in lower test error, and allows for smaller datasets to be used. Loss-function optimization thus provides a new dimension of metalearning, and constitutes an important step towards AutoML.
48.FAN: Focused Attention Networks pdf
Attention networks show promise for both vision and language tasks, by emphasizing relationships between constituent elements through appropriate weighting functions. Such elements could be regions in an image output by a region proposal network, or words in a sentence, represented by word embedding. Thus far, however, the learning of attention weights has been driven solely by the minimization of task specific loss functions. We here introduce a method of learning attention weights to better emphasize informative pair-wise relations between entities. The key idea is to use a novel center-mass cross entropy loss, which can be applied in conjunction with the task specific ones. We then introduce a focused attention backbone to learn these attention weights for general tasks. We demonstrate that the focused attention module leads to a new state-of-the-art for the recovery of relations in a relationship proposal task. Our experiments show that it also boosts performance for diverse vision and language tasks, including object detection, scene categorization and document classification.
49.Scaleable input gradient regularization for adversarial robustness pdf
Input gradient regularization is not thought to be an effective means for promoting adversarial robustness. In this work we revisit this regularization scheme with some new ingredients. First, we derive new per-image theoretical robustness bounds based on local gradient information, and curvature information when available. These bounds strongly motivate input gradient regularization. Second, we implement a scaleable version of input gradient regularization which avoids double backpropagation: adversarially robust ImageNet models are trained in 33 hours on four consumer grade GPUs. Finally, we show experimentally that input gradient regularization is competitive with adversarial training.
50.Capsule Routing via Variational Bayes pdf
Capsule Networks are a recently proposed alternative for constructing Neural Networks, and early indications suggest that they can provide greater generalisation capacity using fewer parameters. In capsule networks scalar neurons are replaced with capsule vectors or matrices, whose entries represent different properties of objects. The relationships between objects and its parts are learned via trainable viewpoint-invariant transformation matrices, and the presence of a given object is decided by the level of agreement among votes from its parts. This interaction occurs between capsule layers and is a process called routing-by-agreement. Although promising, capsule networks remain underexplored by the community, and in this paper we present a new capsule routing algorithm based of Variational Bayes for a mixture of transforming gaussians. Our Bayesian approach addresses some of the inherent weaknesses of EM routing such as the 'variance collapse' by modelling uncertainty over the capsule parameters in addition to the routing assignment posterior probabilities. We test our method on public domain datasets and outperform the state-of-the-art performance on smallNORB using 50% less capsules.
51.Differentiable Quantization of Deep Neural Networks pdf
We propose differentiable quantization (DQ) for efficient deep neural network (DNN) inference where gradient descent is used to learn the quantizer's step size, dynamic range and bitwidth. Training with differentiable quantizers brings two main benefits: first, DQ does not introduce hyperparameters; second, we can learn for each layer a different step size, dynamic range and bitwidth. Our experiments show that DNNs with heterogeneous and learned bitwidth yield better performance than DNNs with a homogeneous one. Further, we show that there is one natural DQ parametrization especially well suited for training. We confirm our findings with experiments on CIFAR-10 and ImageNet and we obtain quantized DNNs with learned quantization parameters achieving state-of-the-art performance.
52.Equivalent and Approximate Transformations of Deep Neural Networks pdf
Two networks are equivalent if they produce the same output for any given input. In this paper, we study the possibility of transforming a deep neural network to another network with a different number of units or layers, which can be either equivalent, a local exact approximation, or a global linear approximation of the original network. On the practical side, we show that certain rectified linear units (ReLUs) can be safely removed from a network if they are always active or inactive for any valid input. If we only need an equivalent network for a smaller domain, then more units can be removed and some layers collapsed. On the theoretical side, we constructively show that for any feed-forward ReLU network, there exists a global linear approximation to a 2-hidden-layer shallow network with a fixed number of units. This result is a balance between the increasing number of units for arbitrary approximation with a single layer and the known upper bound of
$\lceil log(n_0+1)\rceil +1$ layers for exact representation, where$n_0$ is the input dimension. While the transformed network may require an exponential number of units to capture the activation patterns of the original network, we show that it can be made substantially smaller by only accounting for the patterns that define linear regions. Based on experiments with ReLU networks on the MNIST dataset, we found that$l_1$ -regularization and adversarial training reduces the number of linear regions significantly as the number of stable units increases due to weight sparsity. Therefore, we can also intentionally train ReLU networks to allow for effective loss-less compression and approximation.
53.Automatic Delineation of Kidney Region in DCE-MRI pdf
Delineation of the kidney region in dynamic contrast-enhanced magnetic resonance Imaging (DCE-MRI) is required during post-acquisition analysis in order to quantify various aspects of renal function, such as filtration and perfusion or blood flow. However, this can be obfuscated by the Partial Volume Effect (PVE), caused due to the mixing of any single voxel with two or more signal intensities from adjacent regions such as liver region and other tissues. To avoid this problem, firstly, a kidney region of interest (ROI) needs to be defined for the analysis. A clinician may choose to select a region avoiding edges where PV mixing is likely to be significant. However, this approach is time-consuming and labour intensive. To address this issue, we present Dynamic Mode Decomposition (DMD) coupled with thresholding and blob analysis as a framework for automatic delineation of the kidney region. This method is first validated on synthetically generated data with ground-truth available and then applied to ten healthy volunteers' kidney DCE-MRI datasets. We found that the result obtained from our proposed framework is comparable to that of a human expert. For example, while our result gives an average Root Mean Square Error (RMSE) of 0.0097, the baseline achieves an average RMSE of 0.1196 across the 10 datasets. As a result, we conclude automatic modelling via DMD framework is a promising approach.
54.Trust but Verify: An Information-Theoretic Explanation for the Adversarial Fragility of Machine Learning Systems, and a General Defense against Adversarial Attacks pdf
Deep-learning based classification algorithms have been shown to be susceptible to adversarial attacks: minor changes to the input of classifiers can dramatically change their outputs, while being imperceptible to humans. In this paper, we present a simple hypothesis about a feature compression property of artificial intelligence (AI) classifiers and present theoretical arguments to show that this hypothesis successfully accounts for the observed fragility of AI classifiers to small adversarial perturbations. Drawing on ideas from information and coding theory, we propose a general class of defenses for detecting classifier errors caused by abnormally small input perturbations. We further show theoretical guarantees for the performance of this detection method. We present experimental results with (a) a voice recognition system, and (b) a digit recognition system using the MNIST database, to demonstrate the effectiveness of the proposed defense methods. The ideas in this paper are motivated by a simple analogy between AI classifiers and the standard Shannon model of a communication system.
55.Additive Noise Annealing and Approximation Properties of Quantized Neural Networks pdf
We present a theoretical and experimental investigation of the quantization problem for artificial neural networks. We provide a mathematical definition of quantized neural networks and analyze their approximation capabilities, showing in particular that any Lipschitz-continuous map defined on a hypercube can be uniformly approximated by a quantized neural network. We then focus on the regularization effect of additive noise on the arguments of multi-step functions inherent to the quantization of continuous variables. In particular, when the expectation operator is applied to a non-differentiable multi-step random function, and if the underlying probability density is differentiable (in either classical or weak sense), then a differentiable function is retrieved, with explicit bounds on its Lipschitz constant. Based on these results, we propose a novel gradient-based training algorithm for quantized neural networks that generalizes the straight-through estimator, acting on noise applied to the network's parameters. We evaluate our algorithm on the CIFAR-10 and ImageNet image classification benchmarks, showing state-of-the-art performance on AlexNet and MobileNetV2 for ternary networks.