Link prediction and node classification tasks in Social Network Analysis (SNA) remain open research problems with respect to Artificial Intelligence (AI). Thus, the inherent representations about social network structures can be effectively harnessed for training AI models in a bid to predict ties as well as detect clusters via classification of actors with regard to a given social network structure. In this paper, we have proposed a special hybrid model comprising dual layers of Feature Learning (FL): Representation Learning via Knowledge- Graph Embeddings and Convolution Operations (RLVECO). The architecture of RLVECO is tailored towards analyzing and extracting meaningful representations from social network structures so as to aid in link prediction, node classification, and community detection tasks. RLVECO utilizes an edge sampling approach for exploiting features of the social graph via learning the context of each actor with respect to its neighboring actors.

Clustering is an unsupervised machine learning method giving insights on data without early knowledge. Classes of data are return by assembling similar elements together. Giving the increasing of the available data, this method is now applied in a lot of fields with various data types. Here, we propose to explore the case of time series clustering. Indeed, time series are one of the most classic data type, and are present in various fields such as medical or finance. This kind of data can be pre-processed by of dimension reduction methods, such as the recent UMAP algorithm. In this paper, a benchmark of time series clustering is created, comparing the results with and without UMAP as a pre-processing step. UMAP is used to enhance clustering results. For completeness, three different clustering algorithms and two different geometric representation for the time series (Classic Euclidean geometry, and Riemannian geometry on the Stiefel Manifold) are applied. The results are compared with and without UMAP as a pre-processing step on the databases available at UCR Time Series Classification Archive www.cs.ucr.edu/~eamonn/time_series_data/.

In recent years, deep learning has been popular in combining with cytopathology diagnosis. Using the whole slide images (WSI) scanned by electronic scanners at clinics, researchers have developed many algorithms to classify the slide (benign or malignant). However, the key area that support the diagnosis result can be relatively small in a thyroid WSI, and only the global label can be acquired, which make the direct use of the strongly supervised learning framework infeasible. What’s more, because the clinical diagnosis of the thyroid cells requires the use of visual features in different scales, a generic feature extraction way may not achieve good performance. In this paper, we propose a weakly supervised multi-instance learning framework based on attention mechanism with multi-scale feature fusion (MSF) using convolutional neural network (CNN) for thyroid cytopathological diagnosis. We take each WSI as a bag, each bag contains multiple instances which are the different regions of the WSI, our framework is trained to learn the key area automatically and make the classification. We also propose a feature fusion structure, merge the low-level features into the final feature map and add an instance-level attention module in it, which improves the classification accuracy. Our model is trained and tested on the collected clinical data, reaches the accuracy of 93.2%, which outperforms the other existing methods. We also tested our model on a public histopathology dataset and achieves better result than the state-of-the-art deep multi-instance method.

Affinity measure in object tracking outputs a similarity or distance score for given detections. As an affinity measure is typically imperfect, it generally has an uncertain region in which regarding two groups of detections as the same object or different objects based on the score can be wrong. How to reduce the uncertain region is a major challenge for most similarity-based tracking methods. Early mistakes often result in distribution drifts for tracked objects and this is another major issue for object tracking. In this paper, we propose a new offline tracking method called agglomerative hierarchical clustering with ensemble of tracking experts (AHC_ETE), to tackle the uncertain region and early mistake issues. We conduct tracking from certain to uncertain to reduce early mistakes. Meanwhile, we ensemble multiple tracking experts to reduce the uncertain region as the final one is the union of that of each tracking expert. Experiments on MOT16 datasets demonstrated the effectiveness of our method.

Objects and activities (temporal objects) are composed of a recursive hierarchy of perceptual wholes and parts, whose properties, such as shape, reflectance, and color, constitute a hierarchy of intrinsic causal factors of object appearance.However, object appearance is the compositional consequence of both an object’s intrinsic and extrinsic causal factors, where the extrinsic causal factors are related to illumination, and imaging conditions.We propose a unified tensor model of wholes and parts that statistically models the mechanism of data formation. We derive a compositional hierarchical block tensor factorization that computes a disentangled representation of the causal factors of object image formation by optimizing simultaneously across wholes and parts. Given computational efficiency considerations,we introduce an incremental bottom-up computational alternative that employs the lower level part representations to represent the higher level of abstractions, the parent wholes. This incremental computational approach may also be employed to update the causal model representation when data that becomes available incrementally. The resulting object representation is an interpretable combinatorial choice of intrinsic causal factor representations related to an object’s recursive hierarchy of wholes and parts that renders object recognition robust to occlusion and reduces training data requirements.

Recent research on super-resolution (SR) has wit- nessed major developments with the advancements of deep convolutional neural networks. There is a need for information extraction from scenic text images or even document images on device, most of which are low-resolution (LR) images. Therefore, SR becomes an essential pre-processing step as Bicubic Upsampling, which is conventionally present in smartphones, performs poorly on LR images. To give the user more control over his privacy, and to reduce the carbon footprint by reducing the overhead of cloud computing and hours of GPU usage, executing SR models on the edge is a necessity in the recent times. There are various challenges in running and optimizing a model on resource-constrained platforms like smartphones. In this paper, we present a novel deep neural network that reconstructs sharper character edges and thus boosts OCR confidence. The proposed architecture not only achieves significant improvement in PSNR over bicubic upsampling on various benchmark datasets but also runs with an average inference time of 11.7 ms per image. We have outperformed state-of-the-art on the Text330 dataset. We also achieve an OCR accuracy of 75.89% on the ICDAR 2015 TextSR dataset, where ground truth has an accuracy of 78.10%.

Traditionally, a Neural Network classification training loss function follows the same principle: minimizing the distance between samples that belong to the same class, while maximizing the distance to the other classes. There are no restrictions on the spatial placement of deep features (last layer input). This paper addresses this issue when dealing with Neural Networks, providing a set of loss functions that are able to train a classifier by forcing the deep features to be projected over a predefined orthogonal basis. Experimental results shows that these `data placement' functions can overcome the training accuracy provided by the classic cross-entropy loss function.

Visual Question Answering (VQA) models employ attention mechanisms to discover image locations that are most relevant for answering a specific question. For this purpose, several multimodal fusion strategies have been proposed, ranging from relatively simple operations (e.g., linear sum) to more complex ones (e.g., Block). The resulting multimodal representations define an intermediate feature space for capturing the interplay between visual and semantic features, that is helpful in selectively focusing on image content. In this paper, we propose a question-agnostic attention mechanism that is complementary to the existing question-dependent attention mechanisms. Our proposed model parses object instances to obtain an `object map' and applies this map on the visual features to generate Question-Agnostic Attention (QAA) features. In contrast to question-dependent attention approaches that are learned end-to-end, the proposed QAA does not involve question-specific training, and can be easily included in almost any existing VQA model as a generic light-weight pre-processing step, thereby adding minimal computation overhead for training. Further, when used in complement with the question-dependent attention, the QAA allows the model to focus on the regions containing objects that might have been overlooked by the learned attention representation. Through extensive evaluation on VQAv1, VQAv2 and TDIUC datasets, we show that incorporating complementary QAA allows state-of-the-art VQA models to perform better, and provides significant boost to simplistic VQA models, enabling them to performance on par with highly sophisticated fusion strategies.

Learning how to adapt to complex and dynamic environments is one of the most important factors that contribute to our intelligence. Endowing artificial agents with this ability is not a simple task, particularly in competitive scenarios. In this paper, we present a broad study on how popular reinforcement learning algorithms can be adapted and implemented to learn and to play a real-world implementation of a competitive multiplayer card game. We propose specific training and validation routines for the learning agents, in order to evaluate how the agents learn to be competitive and explain how they adapt to each others' playing style. Finally, we pinpoint how the behavior of each agent derives from their learning style and create a baseline for future research on this scenario.

Unsupervised Domain Adaptation (UDA) aims at improving the generalization capability of a model trained on a source domain to perform well on a target domain for which no labeled data is available. In this paper, we consider the semantic segmentation of urban scenes and we propose an approach to adapt a deep neural network trained on synthetic data to real scenes addressing the domain shift between the two different data distributions. We introduce a novel UDA framework where a standard supervised loss on labeled synthetic data is supported by an adversarial module and a self-training strategy aiming at aligning the two domain distributions. The adversarial module is driven by a couple of fully convolutional discriminators dealing with different domains: the first discriminates between ground truth and generated maps, while the second between segmentation maps coming from synthetic or real world data. The self-training module exploits the confidence estimated by the discriminators on unlabeled data to select the regions used to reinforce the learning process. Furthermore, the confidence is thresholded with an adaptive mechanism based on the per-class overall confidence. Experimental results prove the effectiveness of the proposed strategy in adapting a segmentation network trained on synthetic datasets like GTA5 and SYNTHIA, to real world datasets like Cityscapes and Mapillary.

This paper describes the analysis of defocus information when a finger touching a virtual keyboard by using DCT (Discrete Cosine Transform) coefficient without detecting 3D position (especially exact depth value) of thr finger. We use 2 cameras and a half mirror to get the images of the finger with same optical axis. The focal length of the two cameras are slightly different. We can know the finger touching the virtual keyboard when the finger is in the middle of the two focal length from the defocus information, because the position of the virtual keyboard is in the middle of the two focal position where the defocus information of the two cameras are same. Previous work of a virtual keyboard is realized by detecting the feature point of the finger at first. After that the finger position and the depth are checked if they are same to the position and the depth of the keyboard. A problem of the previous virtual keyboard is 3D point of the finger (feature point or edge, small region) should be detected. This is very time consuming. We overcome this problem by comparing the defocus information of the finger in the key area of virtual keyboard on the 2 images captured by the two cameras. In this paper we describe the optical system of the virtual keyboard and the resolution of the depth of the finger. Experimental result make it clear that middle range of the DCT coefficients is effective for detecting the finger. In our system, the finger touch is detected by comparing the DCT coefficients of the 2 images. The minimum distance of the detectable finger is depended on the diameter and the focal length of the lens, the resolution of the image sensor. We formulate this theoretical minimum distance and verify it by experiment. Besides finger, our system can also be generalized to detect other object touching the keyboard or not.

Time series clustering is a challenging task due to the specificities of this type of data. Temporal correlation and invariance to transformations such as shifting, warping or noise prevent the use of standard data mining methods. Time series clustering has been mostly studied under the angle of finding efficient algorithms and distance metrics adapted to the specific nature of time series data. Much less attention has been devoted to the general problem of model selection. Clustering stability has emerged as a universal and model-agnostic principle for clustering model selection. This principle can be stated as follows: an algorithm should find a structure in the data that is resilient to perturbation by sampling or noise. We propose to apply stability analysis to time series by leveraging prior knowledge on the nature and invariances of the data. These invariances determine the perturbation process used to assess stability. Based on a recently introduced criterion combining between-cluster and within-cluster stability, we propose an invariance-guided method for model selection, applicable to a wide range of clustering algorithms. Experiments conducted on artificial and benchmark data sets demonstrate the ability of our criterion to discover structure and select the correct number of clusters, whenever data invariances are known beforehand.

Malware evolves rapidly over time, which makes existing solutions being ineffective in detecting newly released malware. Machine learning models that can learn to capture malicious patterns directly from the data play an increasingly important role in malware analysis. However, traditional machine learning models heavily depend on feature engineering. The extracted static features are vulnerable as hackers could create new malware with different feature values to deceive the machine learning models. In this paper, we propose an end-to-end malware detection framework consisting of convolutional neural network, autoencoder and neural decision trees. It learns the features from multiple domains for malware detection without feature engineering. In addition, since anti-virus products should have a very low false alarm rate to avoid annoying users, we propose a special loss function, which optimizes the recall for a fixed low false positive rate (e.g., less than 0.1%). Experiments show that the proposed framework has achieved a better recall than the baseline models, and the derived loss function also makes a difference.

The recent growth in the number of satellite images fosters the development of effective deep-learning techniques for Remote Sensing (RS). However, their full potential is untapped due to the lack of large annotated datasets. Such a problem is usually countered by fine-tuning a feature extractor that is previously trained on the ImageNet dataset. Unfortunately, the domain of natural images differs from the RS one, which hinders the final performance. In this work, we propose to learn meaningful representations from satellite imagery, leveraging its high-dimensionality spectral bands to reconstruct the visible colors. We conduct experiments on land cover classification (BigEarthNet) and West Nile Virus detection, showing that colorization is a solid pretext task for training a feature extractor. Furthermore, we qualitatively observe that guesses based on natural images and colorization rely on different parts of the input. This paves the way to an ensemble model that eventually outperforms both the above-mentioned techniques.

In this work, we address the problem of audio-based near-duplicate video retrieval. We propose the Audio Similarity Learning (AuSiL) approach that effectively captures temporal patterns of audio similarity between video pairs. For the robust similarity calculation between two videos, we first extract representative audio-based video descriptors by leveraging transfer learning based on a Convolutional Neural Network (CNN) trained on a large scale dataset of audio events, and then we calculate the similarity matrix derived from the pairwise similarity of these descriptors. The similarity matrix is subsequently fed to a CNN network that captures the temporal structures existing within its content. We train our network following a triplet generation process and optimizing the triplet loss function. To evaluate the effectiveness of the proposed approach, we have manually annotated two publicly available video datasets based on the audio duplicity between their videos. The proposed approach achieves very competitive results compared to three state-of-the-art methods. Also, unlike the competing methods, it is very robust for the retrieval of audio duplicates generated with speed transformations.

We propose a new task named Audio-driven Performance Video Generation (APVG), which aims to synthesize the video of a person playing a certain instrument guided by a given music audio clip. It is a challenging task to generate the high-dimensional temporal consistent videos from low-dimensional audio modality. In this paper, we propose a multi-staged framework to achieve this new task to generate realistic and synchronized performance video from given music. Firstly, we provide both global appearance and local spatial information by generating the coarse videos and keypoints of body and hands from a given music respectively. Then, we propose to transform the generated keypoints to heatmap via a differentiable space transformer, since the heatmap offers more spatial information but is harder to generate directly from audio. Finally, we propose a Structured Temporal UNet (STU) to extract both intra-frame structured information and inter-frame temporal consistency. They are obtained via graph-based structure module, and CNN-GRU based high-level temporal module respectively for final video generation. Comprehensive experiments validate the effectiveness of our proposed framework.

In recent times, graph-based semi-supervised learning proved to be a powerful paradigm for processing and mining large datasets. The main advantage relies on the fact that these methods can be useful in propagating a small set of known labels to a large set of unlabeled data. The scarcity of labeled data may affect the performance of the semi-learning. This paper introduces a new semi-supervised framework for simultaneous linear feature extraction and label propagation. The proposed method simultaneously estimates a discriminant transformation and the unknown label by exploiting both labeled and unlabeled data. In addition, the unknowns of the learning model are estimated by integrating two types of graph-based smoothness constraints. The resulting semi-supervised model is expected to learn more discriminative information. Experiments are conducted on six public image datasets. These experimental results show that the performance of the proposed method can be better than that of many state-of-the-art graph-based semi-supervised algorithms.

This paper describes a classifier pool generation method guided by the diversity estimated on the data complexity and classifier decisions. First, the behavior of complexity measures is assessed by considering several subsamples of the dataset. The complexity measures with high variability across the subsamples are selected for posterior pool adaptation, where an evolutionary algorithm optimizes diversity in both complexity and decision spaces. A robust experimental protocol with 28 datasets and 20 replications is used to evaluate the proposed method. Results show significant accuracy improvements in 69.4\% of the experiments when Dynamic Classifier Selection and Dynamic Ensemble Selection methods are applied.

Recently, deep learning-based image super-resolution (SR) has made a remarkable progress. However, previous SR methods rarely focus on the correlation between adjacent layers, which leads to underutilization of the information extracted by each convolutional layer. To address these problem, we design a simple and efficient cross-layer information refining network (CIRN) for single image super-resolution. Concretely, we propose the interlaced spatial attention block (ISAB) to measure the correlation between the adjacent layers feature maps and adaptively rescale spatial-wise features for refining the information. Owing to the two stage information propagation strategy, the CIRN can distill the primary information of adjacent layers without introducing too many parameters. Extensive experiments on benchmark datasets illustrate that our method achieves better accuracy than state-of-the-art methods even in 16× scale, spcifically it has a better banlance between performance and parameters.

Recent developments in the field of Visual Question Answering (VQA) have witnessed promising improvements in performance through contributions in attention based networks. Most such approaches have focused on unidirectional attention that leverage over attention from textual domain (question) on visual space. These approaches mostly focused on learning high-quality attention in the visual space. In contrast, this work proposes an alternating bi-directional attention framework. First, a question to image attention helps to learn the robust visual space embedding, and second, an image to question attention helps to improve the question embedding. This attention mechanism is realized in an alternating fashion i.e. question-to-image followed by image-to-question and is repeated for maximizing performance. We believe that this process of alternating attention generation helps both the modalities and leads to better representations for the VQA task. This proposal is benchmark on TDIUC dataset and against state-of-art approaches. Our ablation analysis shows that alternate attention is the key to achieve high performance in VQA.

Along with the performance improvements of neural network models, developing methods that enable the explanation of their behavior is a significant research topic. For convolutional neural networks, the explainability is usually achieved with attribution (heatmap) that visualizes pixel-level importance or contribution of input to its corresponding result. This attribution should reflect the relation (dependency) between inputs and outputs, which has been studied with a variety of methods, e.g., derivative of an output with respect to an input pixel value, a weighted sum of gradients, amount of output changes to input perturbations, and so on. In this paper, we present a new method that improves the measure of attribution, and incorporates it into the integrated gradients method. To be precise, rather than using the conventional chain-rule, we propose a method called guided non-linearity that propagates gradients more effectively through non-linear units (e.g., ReLU and max-pool) so that only positive gradients backpropagate through non-linear units. Our method is inspired by the mechanism of action potential generation in postsynaptic neurons, where the firing of action potentials depends on the sum of excitatory (EPSP) and inhibitory postsynaptic potentials (IPSP). We believe that paths consisting of EPSP-giving-neurons faithfully reflect the contribution of inputs to the output, and we make gradients flow only along those paths (i.e., paths of positive chain reactions). Experiments with 5 deep neural networks have shown that the proposed method outperforms others in terms of the deletion metrics, and yields fine-grained and more human-interpretable attribution.