It has been shown that financial news leads to the fluctuation of stock prices. However, previous work on news-driven financial market prediction focused only on predicting stock price movement without providing an explanation. In this paper, we propose a dual-layer attention-based neural network to address this issue. In the initial stage, we introduce a knowledge-based method to adaptively extract relevant financial news. Then, we use input attention to pay more attention to the more influential news and concatenate the day embeddings with the output of the news representation. Finally, we use an output attention mechanism to allocate different weights to different days in terms of their contribution to stock price movement. Thorough empirical studies based upon historical prices of several individual stocks demonstrate the superiority of our proposed method in stock price prediction compared to state-of-the-art methods.
http://arxiv.org/abs/1902.04994
Deep learning has raised hopes and expectations as a general solution for many applications; indeed it has proven effective, but it also showed a strong dependence on large quantities of data. Luckily, it has been shown that, even when data is scarce, a successful model can be trained by reusing prior knowledge. Thus, developing techniques for transfer learning, in its broadest definition, is a crucial element towards the deployment of effective and accurate intelligent systems. This thesis will focus on a family of transfer learning methods applied to the task of visual object recognition, specifically image classification. Transfer learning is a general term, and specific settings have been given specific names: when the learner has only access to unlabeled data from the a target domain and labeled data from a different domain (the source), the problem is known as that of “unsupervised domain adaptation” (DA). The first part of this work will focus on three methods for this setting: one of these methods deals with features, one with images while the third one uses both. The second part will focus on the real life issues of robotic perception, specifically RGB-D recognition. Robotic platforms are usually not limited to color perception; very often they also carry a Depth camera. Unfortunately, the depth modality is rarely used for visual recognition due to the lack of pretrained models from which to transfer and little data to train one on from scratch. Two methods for dealing with this scenario will be presented: one using synthetic data and the other exploiting cross-modality transfer learning.
http://arxiv.org/abs/1902.04992
This paper considers the task of locating articulated poses of multiple robots in images. Our approach simultaneously infers the number of robots in a scene, identifies joint locations and estimates sparse depth maps around joint locations. The proposed method applies staged convolutional feature detectors to 2D image inputs and computes robot instance masks using a recurrent network architecture. In addition, regression maps of most likely joint locations in pixel coordinates together with depth information are computed. Compositing 3D robot joint kinematics is accomplished by applying masks to joint readout maps. Our end-to-end formulation is in contrast to previous work in which the composition of robot joints into kinematics is performed in a separate post-processing step. Despite the fact that our models are trained on artificial data, we demonstrate generalizability to real world images.
http://arxiv.org/abs/1902.04987
The CFR framework has been a powerful tool for solving large-scale extensive-form games in practice. However, the theoretical rate at which past CFR-based algorithms converge to the Nash equilibrium is on the order of $O(T^{-1/2})$, where $T$ is the number of iterations. In contrast, first-order methods can be used to achieve a $O(T^{-1})$ dependence on iterations, yet these methods have been less successful in practice. In this work we present the first CFR variant that breaks the square-root dependence on iterations. By combining and extending recent advances on predictive and stable regret minimizers for the matrix-game setting we show that it is possible to leverage “optimistic” regret minimizers to achieve a $O(T^{-3/4})$ convergence rate within CFR. This is achieved by introducing a new notion of stable-predictivity, and by setting the stability of each counterfactual regret minimizer relative to its location in the decision tree. Experiments show that this method is faster than the original CFR algorithm, although not as fast as newer variants, in spite of their worst-case $O(T^{-1/2})$ dependence on iterations.
http://arxiv.org/abs/1902.04982
In this paper, we adapt Recurrent Neural Networks with Stochastic Layers, which are the state-of-the-art for generating text, music and speech, to the problem of acoustic novelty detection. By integrating uncertainty into the hidden states, this type of network is able to learn the distribution of complex sequences. Because the learned distribution can be calculated explicitly in terms of probability, we can evaluate how likely an observation is then detect low-probability events as novel. The model is robust, highly unsupervised, end-to-end and requires minimum preprocessing, feature engineering or hyperparameter tuning. An experiment on a benchmark dataset shows that our model outperforms the state-of-the-art acoustic novelty detectors.
http://arxiv.org/abs/1902.04980
Learning to solve diagrammatic reasoning (DR) can be a challenging but interesting problem to the computer vision research community. It is believed that next generation pattern recognition applications should be able to simulate human brain to understand and analyze reasoning of images. However, due to the lack of benchmarks of diagrammatic reasoning, the present research primarily focuses on visual reasoning that can be applied to real-world objects. In this paper, we present a diagrammatic reasoning dataset that provides a large variety of DR problems. In addition, we also propose a Knowledge-based Long Short Term Memory (KLSTM) to solve diagrammatic reasoning problems. Our proposed analysis is arguably the first work in this research area. Several state-of-the-art learning frameworks have been used to compare with the proposed KLSTM framework in the present context. Preliminary results indicate that the domain is highly related to computer vision and pattern recognition research with several challenging avenues.
http://arxiv.org/abs/1902.04955
In this paper Gabor scattering, a feature extractor based on Gabor frames and Mallat’s scattering transform, is introduced. By using a simple signal model for audio signals, i.e. a class of tones consisting of fundamental frequency and its multiples and an according envelope, we analyse specific properties of Gabor scattering. We show that for each separate layer, different invariances to certain signal characteristics occur. Furthermore, deformation stability of the coefficient vector generated by the feature extractor is derived by using a decoupling technique which exploits the contractivity of general scattering networks. Here, we are interested in robustness with respect to changes in spectral shape and frequency modulation. Our findings are illustrated by numerical examples and experiments. We specifically give numerical evidence that the invariances encoded by the Gabor scattering transform lead to improved generalization properties in comparison with the standard Mel-spectrogram coefficients, in particular in the case of the availability of a restricted amount of data.
http://arxiv.org/abs/1706.08818
Traditional 3D models learn a latent representation of faces using linear subspaces from no more than 300 training scans of a single database. The main roadblock of building a large-scale face model from diverse 3D databases lies in the lack of dense correspondence among raw scans. To address these problems, this paper proposes an innovative framework to jointly learn a nonlinear face model from a diverse set of raw 3D scan databases and establish dense point-to-point correspondence among their scans. Specifically, by treating input raw scans as unorganized point clouds, we explore the use of PointNet architectures for converting point clouds to identity and expression feature representations, from which the decoder networks recover their 3D face shapes. Further, we propose a weakly supervised learning approach that does not require correspondence label for the scans. We demonstrate the superior dense correspondence and representation power of our proposed method in shape and expression, and its contribution to single-image 3D face reconstruction.
http://arxiv.org/abs/1902.04943
Generating informative responses in end-to-end neural dialogue systems attracts a lot of attention in recent years. Various previous work leverages external knowledge and the dialogue contexts to generate such responses. Nevertheless, few has demonstrated their capability on incorporating the appropriate knowledge in response generation. Motivated by this, we propose a novel open-domain conversation generation model in this paper, which employs the posterior knowledge distribution to guide knowledge selection, therefore generating more appropriate and informative responses in conversations. To the best of our knowledge, we are the first one who utilize the posterior knowledge distribution to facilitate conversation generation. Our experiments on both automatic and human evaluation clearly verify the superior performance of our model over the state-of-the-art baselines.
http://arxiv.org/abs/1902.04911
Inspired by the behavior of humans talking in noisy environments, we propose an embodied embedded cognition approach to improve automatic speech recognition (ASR) systems for robots in challenging environments, such as with ego noise, using binaural sound source localization (SSL). The approach is verified by measuring the impact of SSL with a humanoid robot head on the performance of an ASR system. More specifically, a robot orients itself toward the angle where the signal-to-noise ratio (SNR) of speech is maximized for one microphone before doing an ASR task. First, a spiking neural network inspired by the midbrain auditory system based on our previous work is applied to calculate the sound signal angle. Then, a feedforward neural network is used to handle high levels of ego noise and reverberation in the signal. Finally, the sound signal is fed into an ASR system. For ASR, we use a system developed by our group and compare its performance with and without the support from SSL. We test our SSL and ASR systems on two humanoid platforms with different structural and material properties. With our approach we halve the sentence error rate with respect to the common downmixing of both channels. Surprisingly, the ASR performance is more than two times better when the angle between the humanoid head and the sound source allows sound waves to be reflected most intensely from the pinna to the ear microphone, rather than when sound waves arrive perpendicularly to the membrane.
http://arxiv.org/abs/1902.05446
The current state of the art of Simultaneous Localisation and Mapping, or SLAM, on low power embedded systems is about sparse localisation and mapping with low resolution results in the name of efficiency. Meanwhile, research in this field has provided many advances for information rich processing and semantic understanding, combined with high computational requirements for real-time processing. This work provides a solution to bridging this gap, in the form of a scalable SLAM-specific architecture for depth estimation for direct semi-dense SLAM. Targeting an off-the-shelf FPGA-SoC this accelerator architecture achieves a rate of more than 60 mapped frames/sec at a resolution of 640x480 achieving performance on par to a highly-optimised parallel implementation on a high-end desktop CPU with an order of magnitude improved power consumption. Furthermore, the developed architecture is combined with our previous work for the task of tracking, to form the first complete accelerator for semi-dense SLAM on FPGAs, establishing the state of the art in the area of embedded low-power systems.
http://arxiv.org/abs/1902.04907
Deep learning revolutionized data science, and recently its popularity has grown exponentially, as did the amount of papers employing deep networks. Vision tasks, such as human pose estimation, did not escape from this trend. There is a large number of deep models, where small changes in the network architecture, or in the data pre-processing, together with the stochastic nature of the optimization procedures, produce notably different results, making extremely difficult to sift methods that significantly outperform others. This situation motivates the current study, in which we perform a systematic evaluation and statistical analysis of vanilla deep regression, i.e. convolutional neural networks with a linear regression top layer. This is the first comprehensive analysis of deep regression techniques. We perform experiments on four vision problems, and report confidence intervals for the median performance as well as the statistical significance of the results, if any. Surprisingly, the variability due to different data pre-processing procedures generally eclipses the variability due to modifications in the network architecture. Our results reinforce the hypothesis according to which, in general, a general-purpose network (e.g. VGG-16 or ResNet-50) adequately tuned can yield results close to the state-of-the-art without having to resort to more complex and ad-hoc regression models.
http://arxiv.org/abs/1803.08450
Recently, the Magnetic Resonance Imaging (MRI) images have limited and unsatisfactory resolutions due to various constraints such as physical, technological and economic considerations. Super-resolution techniques can obtain high-resolution MRI images. The traditional methods obtained the resolution enhancement of brain MRI by interpolations, affecting the accuracy of the following diagnose process. The requirement for brain image quality is fast increasing. In this paper, we propose an image super-resolution (SR) method based on overcomplete dictionaries and inherent similarity of an image to recover the high-resolution (HR) image from a single low-resolution (LR) image. We explore the nonlocal similarity of the image to tentatively search for similar blocks in the whole image and present a joint reconstruction method based on compressive sensing (CS) and similarity constraints. The sparsity and self-similarity of the image blocks are taken as the constraints. The proposed method is summarized in the following steps. First, a dictionary classification method based on the measurement domain is presented. The image blocks are classified into smooth, texture and edge parts by analyzing their features in the measurement domain. Then, the corresponding dictionaries are trained using the classified image blocks. Equally important, in the reconstruction part, we use the CS reconstruction method to recover the HR brain MRI image, considering both nonlocal similarity and the sparsity of an image as the constraints. This method performs better both visually and quantitatively than some existing methods.
http://arxiv.org/abs/1902.04902
Soft robotic grippers are shown to be high effective for grasping unstructured objects with simple sensing and control strategies. However, they are still limited by their speed, sensing capabilities and actuation mechanism. Hence, their usage have been restricted in highly dynamic grasping tasks. This paper presents a soft robotic gripper with tunable bistable properties for sensor-less dynamic grasping. The bistable mechanism allows us to store arbitrarily large strain energy in the soft system which is then released upon contact. The mechanism also provides flexibility on the type of actuation mechanism as the grasping and sensing phase is completely passive. Theoretical background behind the mechanism is presented with finite element analysis to provide insights into design parameters. Finally, we experimentally demonstrate sensor-less dynamic grasping of an unknown object within 0.02 seconds, including the time to sense and actuate.
http://arxiv.org/abs/1902.04896
As an ubiquitous method in natural language processing, word embeddings are extensively employed to map semantic properties of words into a dense vector representation. They capture semantic and syntactic relations among words but the vector corresponding to the words are only meaningful relative to each other. Neither the vector nor its dimensions have any absolute, interpretable meaning. We introduce an additive modification to the objective function of the embedding learning algorithm that encourages the embedding vectors of words that are semantically related to a predefined concept to take larger values along a specified dimension, while leaving the original semantic learning mechanism mostly unaffected. In other words, we align words that are already determined to be related, along predefined concepts. Therefore, we impart interpretability to the word embedding by assigning meaning to its vector dimensions. The predefined concepts are derived from an external lexical resource, which in this paper is chosen as Roget’s Thesaurus. We observe that alignment along the chosen concepts is not limited to words in the Thesaurus and extends to other related words as well. We quantify the extent of interpretability and assignment of meaning from our experimental results. We also demonstrate the preservation of semantic coherence of the resulting vector space by using word-analogy and word-similarity tests. These tests show that the interpretability-imparted word embeddings that are obtained by the proposed framework do not sacrifice performances in common benchmark tests.
http://arxiv.org/abs/1807.07279
Saliency Map, the gradient of the score function with respect to the input, is the most basic technique for interpreting deep neural network decisions. However, saliency maps are often visually noisy. Although several hypotheses were proposed to account for this phenomenon, there are few works that provide rigorous analyses of noisy saliency maps. In this paper, we identify that noise occurs in saliency maps when irrelevant features pass through ReLU activation functions. Then we propose Rectified Gradient, a method that solves this problem through layer-wise thresholding during backpropagation. Experiments with neural networks trained on CIFAR-10 and ImageNet showed effectiveness of our method and its superiority to other attribution methods.
http://arxiv.org/abs/1902.04893
People re-identification task has seen enormous improvements in the latest years, mainly due to the development of better image features extraction from deep Convolutional Neural Networks (CNN) and the availability of large datasets. However, little research has been conducted on animal identification and re-identification, even if this knowledge may be useful in a rich variety of different scenarios. Here, we tackle cattle re-identification exploiting deep CNN and show how this task is poorly related with the human one, presenting unique challenges that makes it far from being solved. We present various baselines, both based on deep architectures or on standard machine learning algorithms, and compared them with our solution. Finally, a rich ablation study has been conducted to further investigate the unique peculiarities of this task.
http://arxiv.org/abs/1902.04886
Today’s AI still faces two major challenges. One is that in most industries, data exists in the form of isolated islands. The other is the strengthening of data privacy and security. We propose a possible solution to these challenges: secure federated learning. Beyond the federated learning framework first proposed by Google in 2016, we introduce a comprehensive secure federated learning framework, which includes horizontal federated learning, vertical federated learning and federated transfer learning. We provide definitions, architectures and applications for the federated learning framework, and provide a comprehensive survey of existing works on this subject. In addition, we propose building data networks among organizations based on federated mechanisms as an effective solution to allow knowledge to be shared without compromising user privacy.
http://arxiv.org/abs/1902.04885
In a resource-constrained, contested environment, computing resources need to be aware of possible size, weight, and power (SWaP) restrictions. SWaP-aware computational efficiency depends upon optimization of computational resources and intelligent time versus efficiency tradeoffs in decision making. In this paper we address the complexity of various optimization strategies related to low SWaP computing. Due to these restrictions, only a small subset of less complicated and fast computable algorithms can be used for tactical, adaptive computing.
http://arxiv.org/abs/1902.05070
Typical person re-identification frameworks search for k best matches in a gallery of images that are often collected in varying conditions. The gallery may contain image sequences when re-identification is done on videos. However, such a process is time consuming as re-identification has to be carried out multiple times. In this paper, we extract spatio-temporal sequences of frames (referred to as tubes) of moving persons and apply a multi-stage processing to match a given query tube with a gallery of stored tubes recorded through other cameras. Initially, we apply a binary classifier to remove noisy images from the input query tube. In the next step, we use a key-pose detection-based query minimization. This reduces the length of the query tube by removing redundant frames. Finally, a 3-stage hierarchical re-identification framework is used to rank the output tubes as per the matching scores. Experiments with publicly available video re-identification datasets reveal that our framework is better than state-of-the-art methods. It ranks the tubes with an increased CMC accuracy of 6-8% across multiple datasets. Also, our method significantly reduces the number of false positives. A new video re-identification dataset, named Tube-based Reidentification Video Dataset (TRiViD), has been prepared with an aim to help the re-identification research community
http://arxiv.org/abs/1902.04856
In this paper, we propose a novel method for highly efficient follicular segmentation of thyroid cytopathological WSIs. Firstly, we propose a hybrid segmentation architecture, which integrates a classifier into Deeplab V3 by adding a branch. A large amount of the WSI segmentation time is saved by skipping the irrelevant areas using the classification branch. Secondly, we merge the low scale fine features into the original atrous spatial pyramid pooling (ASPP) in Deeplab V3 to accurately represent the details in cytopathological images. Thirdly, our hybrid model is trained by a criterion-oriented adaptive loss function, which leads the model converging much faster. Experimental results on a collection of thyroid patches demonstrate that the proposed model reaches 80.9% on the segmentation accuracy. Besides, 93% time is reduced for the WSI segmentation by using our proposed method, and the WSI-level accuracy achieves 53.4%.
http://arxiv.org/abs/1902.05431
Rational decision making in its linguistic description means making logical decisions. In essence, a rational agent optimally processes all relevant information to achieve its goal. Rationality has two elements and these are the use of relevant information and the efficient processing of such information. In reality, relevant information is incomplete, imperfect and the processing engine, which is a brain for humans, is suboptimal. Humans are risk averse rather than utility maximizers. In the real world, problems are predominantly non-convex and this makes the idea of rational decision-making fundamentally unachievable and Herbert Simon called this bounded rationality. There is a trade-off between the amount of information used for decision-making and the complexity of the decision model used. This explores whether machine rationality is subjective and concludes that indeed it is.
http://arxiv.org/abs/1902.04832
Surgical tool segmentation in endoscopic images is the first step towards pose estimation and (sub-)task automation in challenging minimally invasive surgical operations. While many approaches in the literature have shown great results using modern machine learning methods such as convolutional neural networks, the main bottleneck lies in the acquisition of a large number of manually-annotated images for efficient learning. This is especially true in surgical context, where patient-to-patient differences impede the overall generalizability. In order to cope with this lack of annotated data, we propose a self-supervised approach in a robot-assisted context. To our knowledge, the proposed approach is the first to make use of the kinematic model of the robot in order to generate training labels. The core contribution of the paper is to propose an optimization method to obtain good labels for training despite an unknown hand-eye calibration and an imprecise kinematic model. The labels can subsequently be used for fine-tuning a fully-convolutional neural network for pixel-wise classification. As a result, the tool can be segmented in the endoscopic images without needing a single manually-annotated image. Experimental results on phantom and in vivo datasets obtained using a flexible robotized endoscopy system are very promising.
http://arxiv.org/abs/1902.04810
Deep Models, typically Deep neural networks, have millions of parameters, analyze medical data accurately, yet in a time-consuming method. However, energy cost effectiveness and computational efficiency are important for prerequisites developing and deploying mobile-enabled devices, the mainstream trend in connected healthcare.
http://arxiv.org/abs/1902.05429
When searching for information, a human reader first glances over a document, spots relevant sections and then focuses on a few sentences for resolving her intention. However, the high variance of document structure complicates to identify the salient topic of a given section at a glance. To tackle this challenge, we present SECTOR, a model to support machine reading systems by segmenting documents into coherent sections and assigning topic labels to each section. Our deep neural network architecture learns a latent topic embedding over the course of a document. This can be leveraged to classify local topics from plain text and segment a document at topic shifts. In addition, we contribute WikiSection, a publicly available dataset with 242k labeled sections in English and German from two distinct domains: diseases and cities. From our extensive evaluation of 20 architectures, we report a highest score of 71.6% F1 for the segmentation and classification of 30 topics from the English city domain, scored by our SECTOR LSTM model with bloom filter embeddings and bidirectional segmentation. This is a significant improvement of 29.5 points F1 compared to state-of-the-art CNN classifiers with baseline segmentation.
http://arxiv.org/abs/1902.04793
Classification of audio samples is an important part of many auditory systems. Deep learning models based on the Convolutional and the Recurrent layers are state-of-the-art in many such tasks. In this paper, we approach audio classification tasks using capsule networks trained by recently proposed dynamic routing-by-agreement mechanism. We propose an architecture for capsule networks fit for audio classification tasks and study the impact of various parameters on classification accuracy. Further, we suggest modifications for regularization and multi-label classification. We also develop insights into the data using capsule outputs and show the utility of the learned network for transfer learning. We perform experiments on 7 datasets of different domains and sizes and show significant improvements in performance compared to strong baseline models. To the best of our knowledge, this is the first detailed study about the application of capsule networks in the audio domain.
http://arxiv.org/abs/1902.05069
In Chinese societies, superstition is of paramount importance, and vehicle license plates with desirable numbers can fetch very high prices in auctions. Unlike other valuable items, license plates are not allocated an estimated price before auction. I propose that the task of predicting plate prices can be viewed as a natural language processing (NLP) task, as the value depends on the meaning of each individual character on the plate and its semantics. I construct a deep recurrent neural network (RNN) to predict the prices of vehicle license plates in Hong Kong, based on the characters on a plate. I demonstrate the importance of having a deep network and of retraining. Evaluated on 13 years of historical auction prices, the deep RNN outperforms previous models by a significant margin.
http://arxiv.org/abs/1701.08711
Consider a Markov decision process (MDP) that admits a set of state-action features, which can linearly express the process’s probabilistic transition model. We propose a parametric Q-learning algorithm that finds an approximate-optimal policy using a sample size proportional to the feature dimension $K$ and invariant with respect to the size of the state space. To further improve its sample efficiency, we exploit the monotonicity property and intrinsic noise structure of the Bellman operator, provided the existence of anchor state-actions that imply implicit non-negativity in the feature space. We augment the algorithm using techniques of variance reduction, monotonicity preservation, and confidence bounds. It is proved to find a policy which is $\epsilon$-optimal from any initial state with high probability using $\widetilde{O}(K/\epsilon^2(1-\gamma)^3)$ sample transitions for arbitrarily large-scale MDP with a discount factor $\gamma\in(0,1)$. A matching information-theoretical lower bound is proved, confirming the sample optimality of the proposed method with respect to all parameters (up to polylog factors).
http://arxiv.org/abs/1902.04779
Variational inference (VI) is a widely used framework in Bayesian estimation. For most of the non-Gaussian statistical models, it is infeasible to find an analytically tractable solution to estimate the posterior distributions of the parameters. Recently, an improved framework, namely the extended variational inference (EVI), has been introduced and applied to derive analytically tractable solution by employing lower-bound approximation to the variational objective function. Two conditions required for EVI implementation, namely the weak condition and the strong condition, are discussed and compared in this paper. In practical implementation, the convergence of the EVI depends on the selection of the lower-bound approximation, no matter with the weak condition or the strong condition. In general, two approximation strategies, the single lower-bound (SLB) approximation and the multiple lower-bounds (MLB) approximation, can be applied to carry out the lower-bound approximation. To clarify the differences between the SLB and the MLB, we will also discuss the convergence properties of the aforementioned two approximations. Extensive comparisons are made based on some existing EVI-based non-Gaussian statistical models. Theoretical analysis are conducted to demonstrate the differences between the weak and the strong conditions. Qualitative and quantitative experimental results are presented to show the advantages of the SLB approximation.
http://arxiv.org/abs/1902.05068
In this paper, we study the challenging unconstrained set-based face recognition problem where each subject face is instantiated by a set of media (images and videos) instead of a single image. Naively aggregating information from all the media within a set would suffer from the large intra-set variance caused by heterogeneous factors (e.g., varying media modalities, poses and illuminations) and fail to learn discriminative face representations. A novel Multi-Prototype Network (MPNet) model is thus proposed to learn multiple prototype face representations adaptively from the media sets. Each learned prototype is representative for the subject face under certain condition in terms of pose, illumination and media modality. Instead of handcrafting the set partition for prototype learning, MPNet introduces a Dense SubGraph (DSG) learning sub-net that implicitly untangles inconsistent media and learns a number of representative prototypes. Qualitative and quantitative experiments clearly demonstrate superiority of the proposed model over state-of-the-arts.
http://arxiv.org/abs/1902.04755
In robotic surgery, the surgeon controls robotic instruments using dedicated interfaces. One critical limitation of current interfaces is that they are designed to be operated by only the hands. This means that the surgeon can only control at most two robotic instruments at one time while many interventions require three instruments. This paper introduces a novel four-degree-of-freedom foot-machine interface which allows the surgeon to control a third robotic instrument using the foot, giving the surgeon a “third hand”. This interface is essentially a parallel-serial hybrid mechanism with springs and force sensors. Unlike existing switch-based interfaces that can only un-intuitively generate motion in discrete directions, this interface allows intuitive control of a slave robotic arm in continuous directions and speeds, naturally matching the foot movements with dynamic force & position feedbacks. An experiment with ten naive subjects was conducted to test the system. In view of the significant variance of motion patterns between subjects, a subject-specific mapping from foot movements to command outputs was developed using Independent Component Analysis (ICA). Results showed that the ICA method could accurately identify subjects’ foot motion patterns and significantly improve the prediction accuracy of motion directions from 68% to 88% as compared with the forward kinematics-based approach. This foot-machine interface can be applied for the teleoperation of industrial/surgical robots independently or in coordination with hands in the future.
http://arxiv.org/abs/1902.04752
Sentence compression is an important problem in natural language processing. In this paper, we firstly establish a new sentence compression model based on the probability model and the parse tree model. Our sentence compression model is equivalent to an integer linear program (ILP) which can both guarantee the syntax correctness of the compression and save the main meaning. We propose using a DC (Difference of convex) programming approach (DCA) for finding local optimal solution of our model. Combing DCA with a parallel-branch-and-bound framework, we can find global optimal solution. Numerical results demonstrate the good quality of our sentence compression model and the excellent performance of our proposed solution algorithm.
http://arxiv.org/abs/1902.07248
The task of multi-image cued story generation, such as visual storytelling dataset (VIST) challenge, is to compose multiple coherent sentences from a given sequence of images. The main difficulty is how to generate image-specific sentences within the context of overall images. Here we propose a deep learning network model, GLAC Net, that generates visual stories by combining global-local (glocal) attention and context cascading mechanisms. The model incorporates two levels of attention, i.e., overall encoding level and image feature level, to construct image-dependent sentences. While standard attention configuration needs a large number of parameters, the GLAC Net implements them in a very simple way via hard connections from the outputs of encoders or image features onto the sentence generators. The coherency of the generated story is further improved by conveying (cascading) the information of the previous sentence to the next sentence serially. We evaluate the performance of the GLAC Net on the visual storytelling dataset (VIST) and achieve very competitive results compared to the state-of-the-art techniques. Our code and pre-trained models are available here.
http://arxiv.org/abs/1805.10973
Multi-instance learning (MIL) deals with tasks where each example is represented by a bag of instances. Unlike traditional supervised learning, only the bag labels are observed whereas the label for each instance in the bags is not available. Previous MIL studies typically assume that training and the test data follow the same distribution, which is often violated in real-world applications. Existing methods address distribution changes by reweighting the training bags with the density ratio between the test and the training data. However, models are frequently trained without prior knowledge of the testing distribution which renders existing methods ineffective. In this paper, we propose a novel multi-instance learning algorithm which links MIL with causal inference to achieve stable prediction without knowing the distribution of the test dataset. Experimental results show that the performance of our approach is stable to the distribution changes.
http://arxiv.org/abs/1902.05066
We consider the problem of accurately identifying cell boundaries and labeling individual cells in confocal microscopy images, specifically, 3D image stacks of cells with tagged cell membranes. Precise identification of cell boundaries, their shapes, and quantifying inter-cellular space leads to a better understanding of cell morphogenesis. Towards this, we outline a cell segmentation method that uses a deep neural network architecture to extract a confidence map of cell boundaries, followed by a 3D watershed algorithm and a final refinement using a conditional random field. In addition to improving the accuracy of segmentation compared to other state-of-the-art methods, the proposed approach also generalizes well to different datasets without the need to retrain the network for each dataset. Detailed experimental results are provided, and the source code is available on GitHub.
http://arxiv.org/abs/1902.04729
In this paper, we propose a novel conditional-generative-adversarial-nets-based image captioning framework as an extension of traditional reinforcement-learning (RL)-based encoder-decoder architecture. To deal with the inconsistent evaluation problem among different objective language metrics, we are motivated to design some “discriminator” networks to automatically and progressively determine whether generated caption is human described or machine generated. Two kinds of discriminator architectures (CNN and RNN-based structures) are introduced since each has its own advantages. The proposed algorithm is generic so that it can enhance any existing RL-based image captioning framework and we show that the conventional RL training method is just a special case of our approach. Empirically, we show consistent improvements over all language evaluation metrics for different state-of-the-art image captioning models. In addition, the well-trained discriminators can also be viewed as objective image captioning evaluators
https://arxiv.org/abs/1805.07112
Automated segmentation of the optic cup and disk on retinal fundus images is fundamental for the automated detection / analysis of glaucoma. Traditional segmentation approaches depend heavily upon hand-crafted features and a priori knowledge of the user. As such, these methods are difficult to be adapt to the clinical environment. Recently, deep learning methods based on fully convolutional networks (FCNs) have been successful in resolving segmentation problems. However, the reliance on large annotated training data is problematic when dealing with medical images. If a sufficient amount of annotated training data to cover all possible variations is not available, FCNs do not provide accurate segmentation. In addition, FCNs have a large receptive field in the convolutional layers, and hence produce coarse outputs of boundaries. Hence, we propose a new fully automated method that we refer to as a dual-stage fully convolutional networks (DSFCN). Our approach leverages deep residual architectures and FCNs and learns and infers the location of the optic cup and disk in a step-wise manner with fine-grained details. During training, our approach learns from the training data and the estimated results derived from the previous iteration. The ability to learn from the previous iteration optimizes the learning of the optic cup and the disk boundaries. During testing (prediction), DSFCN uses test (input) images and the estimated probability map derived from previous iterations to gradually improve the segmentation accuracy. Our method achieved an average Dice co-efficient of 0.8488 and 0.9441 for optic cup and disk segmentation and an area under curve (AUC) of 0.9513 for glaucoma detection.
http://arxiv.org/abs/1902.04713
Obtaining reliable data describing local Food Security Metrics (FSM) at a granularity that is informative to policy-makers requires expensive and logistically difficult surveys, particularly in the developing world. We train a CNN on publicly available satellite data describing land cover classification and use both transfer learning and direct training to build a model for FSM prediction purely from satellite imagery data. We then propose efficient tasking algorithms for high resolution satellite assets via transfer learning, Markovian search algorithms, and Bayesian networks.
http://arxiv.org/abs/1902.05433
We present a method for fast training of vision based control policies on real robots. The key idea behind our method is to perform multi-task Reinforcement Learning with auxiliary tasks that differ not only in the reward to be optimized but also in the state-space in which they operate. In particular, we allow auxiliary task policies to utilize task features that are available only at training-time. This allows for fast learning of auxiliary policies, which subsequently generate good data for training the main, vision-based control policies. This method can be seen as an extension of the Scheduled Auxiliary Control (SAC-X) framework. We demonstrate the efficacy of our method by using both a simulated and real-world Ball-in-a-Cup game controlled by a robot arm. In simulation, our approach leads to significant learning speed-ups when compared to standard SAC-X. On the real robot we show that the task can be learned from-scratch, i.e., with no transfer from simulation and no imitation learning. Videos of our learned policies running on the real robot can be found at https://sites.google.com/view/rss-2019-sawyer-bic/.
http://arxiv.org/abs/1902.04706
Illuminant estimation plays a key role in digital camera pipeline system, it aims at reducing color casting effect due to the influence of non-white illuminant. Recent researches handle this task by using Convolution Neural Network (CNN) as a mapping function from input image to a single illumination vector. However, global mapping approaches are difficult to deal with scenes under multi-light-sources. In this paper, we proposed a self-adaptive single and multi-illuminant estimation framework, which includes the following novelties: (1) Learning local self-adaptive kernels from the entire image for illuminant estimation with encoder-decoder CNN structure; (2) Providing confidence measurement for the prediction; (3) Clustering-based iterative fitting for computing single and multi-illumination vectors. The proposed global-to-local aggregation is able to predict multi-illuminant regionally by utilizing global information instead of training in patches, as well as brings significant improvement for single illuminant estimation. We outperform the state-of-the-art methods on standard benchmarks with the largest relative improvement of 16%. In addition, we collect a dataset contains over 13k images for illuminant estimation and evaluation. The code and dataset is available on https://github.com/LiamLYJ/KPF_WB
http://arxiv.org/abs/1902.04705
We study the interplay between memorization and generalization of overparametrized networks in the extreme case of a single training example. The learning task is to predict an output which is as similar as possible to the input. We examine both fully-connected and convolutional networks that are initialized randomly and then trained to minimize the reconstruction error. The trained networks take one of the two forms: the constant function (“memorization”) and the identity function (“generalization”). We show that different architectures exhibit vastly different inductive bias towards memorization and generalization. An important consequence of our study is that even in extreme cases of overparameterization, deep learning can result in proper generalization.
http://arxiv.org/abs/1902.04698
Trajectory following is one of the complicated control problems when its dynamics are nonlinear, stochastic and include a large number of parameters. The problem has significant difficulties including a large number of trials required for data collection and a massive volume of computations required to find a closed-loop controller for high dimensional and stochastic domains. For solving this type of problem, if we have an appropriate reward function and dynamics model; finding an optimal control policy is possible by using model-based reinforcement learning and optimal control algorithms. However, defining an accurate dynamics model is not possible for complicated problems. Pieter Abbeel and Andrew Ng recently presented an algorithm that requires only an approximate model and only a small number of real-life trials. This algorithm has broad applicability; however, there are some problems regarding the convergence of the algorithm. In this research, required modifications are presented that provide more powerful assurance for converging to optimal control policy. Also updated algorithm is implemented to evaluate the efficiency of the new algorithm by comparing the acquired results with human expert performance. We are using differential dynamic programming (DDP) as the locally trajectory optimizer, and a 2D dynamics and kinematics simulator is used to evaluate the accuracy of the presented algorithm.
http://arxiv.org/abs/1902.04696
After a disaster, teams of structural engineers collect vast amounts of images from damaged buildings to obtain lessons and gain knowledge from the event. Images of damaged buildings and components provide valuable evidence to understand the consequences on our structures. However, in many cases, images of damaged buildings are often captured without sufficient spatial context. Also, they may be hard to recognize in cases with severe damage. Incorporating past images showing a pre-disaster condition of such buildings is helpful to accurately evaluate possible circumstances related to a building’s failure. One of the best resources to observe the pre-disaster condition of the buildings is Google Street View. A sequence of 360 panorama images which are captured along streets enables all-around views at each location on the street. Once a user knows the GPS information near the building, all external views of the building can be made available. In this study, we develop an automated technique to extract past building images from 360 panorama images serviced by Google Street View. Users only need to provide a geo-tagged image, collected near the target building, and the rest of the process is fully automated. High-quality and undistorted building images are extracted from past panoramas. Since the panoramas are collected from various locations near the building along the street, the user can identify its pre-disaster conditions from the full set of external views.
http://arxiv.org/abs/1902.10816
In this paper we introduce a new digital image forensics approach called forensic similarity, which determines whether two image patches contain the same forensic trace or different forensic traces. One benefit of this approach is that prior knowledge, e.g. training samples, of a forensic trace are not required to make a forensic similarity decision on it in the future. To do this, we propose a two part deep-learning system composed of a CNN-based feature extractor and a three-layer neural network, called the similarity network. This system maps pairs of image patches to a score indicating whether they contain the same or different forensic traces. We evaluated system accuracy of determining whether two image patches were 1) captured by the same or different camera model, 2) manipulated by the same or different editing operation, and 3) manipulated by the same or different manipulation parameter, given a particular editing operation. Experiments demonstrate applicability to a variety of forensic traces, and importantly show efficacy on “unknown” forensic traces that were not used to train the system. Experiments also show that the proposed system significantly improves upon prior art, reducing error rates by more than half. Furthermore, we demonstrated the utility of the forensic similarity approach in two practical applications: forgery detection and localization, and database consistency verification.
http://arxiv.org/abs/1902.04684
This paper investigates the hybrid precoding design for millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems with finite-alphabet inputs. The mmWave MIMO system employs partially-connected hybrid precoding architecture with dynamic subarrays, where each radio frequency (RF) chain is connected to a dynamic subset of antennas. We consider the design of analog and digital precoders utilizing statistical and/or mixed channel state information (CSI), which involve solving an extremely difficult problem in theory: First, designing the optimal partition of antennas over RF chains is a combinatorial optimization problem, whose optimal solution requires an exhaustive search over all antenna partitioning solutions; Second, the average mutual information under mmWave MIMO channels lacks closed-form expression and involves prohibitive computational burden; Third, the hybrid precoding problem with given partition of antennas is nonconvex with respect to the analog and digital precoders. To address these issues, this study first presents a simple criterion and the corresponding low complexity algorithm to design the optimal partition of antennas using statistical CSI. Then it derives the lower bound and its approximation for the average mutual information, in which the computational complexity is greatly reduced compared to calculating the average mutual information directly. In addition, it also shows that the lower bound with a constant shift offers a very accurate approximation to the average mutual information. This paper further proposes utilizing the lower bound approximation as a low-complexity and accurate alternative for developing a manifold-based gradient ascent algorithm to find near optimal analog and digital precoders. Several numerical results are provided to show that our proposed algorithm outperforms existing hybrid precoding algorithms.
https://arxiv.org/abs/1902.04677
Artificial Intelligence (AI) is necessary to address the large and growing deficit in retina and healthcare access globally. And mobile AI diagnostic platforms running in the Cloud may effectively and efficiently distribute such AI capability. Here we sought to evaluate the feasibility of Cloud-based mobile artificial intelligence for detection of retinal disease. And to evaluate the accuracy of a particular such system for detection of subretinal fluid (SRF) and macula edema (ME) on OCT scans. A multicenter retrospective image analysis was conducted in which board-certified ophthalmologists with fellowship training in retina evaluated OCT images of the macula. They noted the presence or absence of ME or SRF, then compared their assessment to that obtained from Fluid Intelligence, a mobile AI app that detects SRF and ME on OCT scans. Investigators consecutively selected retinal OCTs, while making effort to balance the number of scans with retinal fluid and scans without. Exclusion criteria included poor scan quality, ambiguous features, macula holes, retinoschisis, and dense epiretinal membranes. Accuracy in the form of sensitivity and specificity of the AI mobile App was determined by comparing its assessments to those of the retina specialists. At the time of this submission, five centers have completed their initial studies. This consists of a total of 283 OCT scans of which 155 had either ME or SRF (“wet”) and 128 did not (“dry”). The sensitivity ranged from 82.5% to 97% with a weighted average of 89.3%. The specificity ranged from 52% to 100% with a weighted average of 81.23%. CONCLUSION: Cloud-based Mobile AI technology is feasible for the detection retinal disease. In particular, Fluid Intelligence (alpha version), is sufficiently accurate as a screening tool for SRF and ME, especially in underserved areas. Further studies and technology development is needed.
http://arxiv.org/abs/1902.02905
A free-standing bulk gallium nitride layer with a thickness of 365 $\mu$m and a diameter of 50 mm was obtained by hydride vapor phase epitaxy on a sapphire substrate with a carbon buffer layer. The carbon buffer layer was deposited by thermal decomposition of methane $\textit{in situ}$ in the same process with the growth of a bulk GaN layer. The bulk GaN layer grown on the carbon buffer layer self-separated from the sapphire substrate during the cooling after the growth. The dislocation density was $8\cdot10^{6}$ cm$^{-2}$. The (0002) X-Ray rocking curve full width at half maximum was 164 arcsec.
https://arxiv.org/abs/1902.04672
Recently, Generative Adversarial Networks (GANs) have emerged as a popular alternative for modeling complex high dimensional distributions. Most of the existing works implicitly assume that the clean samples from the target distribution are easily available. However, in many applications, this assumption is violated. In this paper, we consider the observation setting when the samples from target distribution are given by the superposition of two structured components and leverage GANs for learning the structure of the components. We propose two novel frameworks: denoising-GAN and demixing-GAN. The denoising-GAN assumes access to clean samples from the second component and try to learn the other distribution, whereas demixing-GAN learns the distribution of the components at the same time. Through extensive numerical experiments, we demonstrate that proposed frameworks can generate clean samples from unknown distributions, and provide competitive performance in tasks such as denoising, demixing, and compressive sensing.
https://arxiv.org/abs/1902.04664
Cardiovascular disease (CVD) is the global leading cause of death. A strong risk factor for CVD events is the amount of coronary artery calcium (CAC). To meet demands of the increasing interest in quantification of CAC, i.e. coronary calcium scoring, especially as an unrequested finding for screening and research, automatic methods have been proposed. Current automatic calcium scoring methods are relatively computationally expensive and only provide scores for one type of CT. To address this, we propose a computationally efficient method that employs two ConvNets: the first performs registration to align the fields of view of input CTs and the second performs direct regression of the calcium score, thereby circumventing time-consuming intermediate CAC segmentation. Optional decision feedback provides insight in the regions that contributed to the calcium score. Experiments were performed using 903 cardiac CT and 1,687 chest CT scans. The method predicted calcium scores in less than 0.3 s. Intra-class correlation coefficient between predicted and manual calcium scores was 0.98 for both cardiac and chest CT. The method showed almost perfect agreement between automatic and manual CVD risk categorization in both datasets, with a linearly weighted Cohen’s kappa of 0.95 in cardiac CT and 0.93 in chest CT. Performance is similar to that of state-of-the-art methods, but the proposed method is hundreds of times faster. By providing visual feedback, insight is given in the decision process, making it readily implementable in clinical and research settings.
http://arxiv.org/abs/1902.05408
Testing cyber-physical systems involves the execution of test cases on target-machines equipped with the latest release of a software control system. When testing industrial robots, it is common that the target machines need to share some common resources, e.g., costly hardware devices, and so there is a need to schedule test case execution on the target machines, accounting for these shared resources. With a large number of such tests executed on a regular basis, this scheduling becomes difficult to manage manually. In fact, with manual test execution planning and scheduling, some robots may remain unoccupied for long periods of time and some test cases may not be executed. This paper introduces TC-Sched, a time-aware method for automated test case execution scheduling. TC-Sched uses Constraint Programming to schedule tests to run on multiple machines constrained by the tests’ access to shared resources, such as measurement or networking devices. The CP model is written in SICStus Prolog and uses the Cumulatives global constraint. Given a set of test cases, a set of machines, and a set of shared resources, TC-Sched produces an execution schedule where each test is executed once with minimal time between when a source code change is committed and the test results are reported to the developer. Experiments reveal that TC-Sched can schedule 500 test cases over 100 machines in less than 4 minutes for 99.5% of the instances. In addition, TC-Sched largely outperforms simpler methods based on a greedy algorithm and is suitable for deployment on industrial robot testing.
http://arxiv.org/abs/1902.04627