Multi-output learning aims to simultaneously predict multiple outputs given an input. It is an important learning problem due to the pressing need for sophisticated decision making in real-world applications. Inspired by big data, the 4Vs characteristics of multi-output imposes a set of challenges to multi-output learning, in terms of the volume, velocity, variety and veracity of the outputs. Increasing number of works in the literature have been devoted to the study of multi-output learning and the development of novel approaches for addressing the challenges encountered. However, it lacks a comprehensive overview on different types of challenges of multi-output learning brought by the characteristics of the multiple outputs and the techniques proposed to overcome the challenges. This paper thus attempts to fill in this gap to provide a comprehensive review on this area. We first introduce different stages of the life cycle of the output labels. Then we present the paradigm on multi-output learning, including its myriads of output structures, definitions of its different sub-problems, model evaluation metrics and popular data repositories used in the study. Subsequently, we review a number of state-of-the-art multi-output learning methods, which are categorized based on the challenges.
https://arxiv.org/abs/1901.00248
Machine learning models have become more and more complex in order to better approximate complex functions. Although fruitful in many domains, the added complexity has come at the cost of model interpretability. The once popular k-nearest neighbors (kNN) approach, which finds and uses the most similar data for reasoning, has received much less attention in recent decades due to numerous problems when compared to other techniques. We show that many of these historical problems with kNN can be overcome, and our contribution has applications not only in machine learning but also in online learning, data synthesis, anomaly detection, model compression, and reinforcement learning, without sacrificing interpretability. We introduce a synthesis between kNN and information theory that we hope will provide a clear path towards models that are innately interpretable and auditable. Through this work we hope to gather interest in combining kNN with information theory as a promising path to fully auditable machine learning and artificial intelligence.
https://arxiv.org/abs/1901.00246
In many real-world learning scenarios, features are only acquirable at a cost constrained under a budget. In this paper, we propose a novel approach for cost-sensitive feature acquisition at the prediction-time. The suggested method acquires features incrementally based on a context-aware feature-value function. We formulate the problem in the reinforcement learning paradigm, and introduce a reward function based on the utility of each feature. Specifically, MC dropout sampling is used to measure expected variations of the model uncertainty which is used as a feature-value function. Furthermore, we suggest sharing representations between the class predictor and value function estimator networks. The suggested approach is completely online and is readily applicable to stream learning setups. The solution is evaluated on three different datasets including the well-known MNIST dataset as a benchmark as well as two cost-sensitive datasets: Yahoo Learning to Rank and a dataset in the medical domain for diabetes classification. According to the results, the proposed method is able to efficiently acquire features and make accurate predictions.
https://arxiv.org/abs/1901.00243
We present Variational Aspect-based Latent Topic Allocation (VALTA), a family of autoencoding topic models that learn aspect-based representations of reviews. VALTA defines a user-item encoder that maps bag-of-words vectors for combined reviews associated with each paired user and item onto structured embeddings, which in turn define per-aspect topic weights. We model individual reviews in a structured manner by inferring an aspect assignment for each sentence in a given review, where the per-aspect topic weights obtained by the user-item encoder serve to define a mixture over topics, conditioned on the aspect. The result is an autoencoding neural topic model for reviews, which can be trained in a fully unsupervised manner to learn topics that are structured into aspects. Experimental evaluation on large number of datasets demonstrates that aspects are interpretable, yield higher coherence scores than non-structured autoencoding topic model variants, and can be utilized to perform aspect-based comparison and genre discovery.
http://arxiv.org/abs/1812.05035
Collecting a large-scale and well-annotated dataset for image processing has become a common practice in computer vision. However, in the ancient painting area, this task is not practical as the number of paintings is limited and their style is greatly diverse. We, therefore, propose a novel solution for the problems that come with ancient painting processing. This is to use domain transfer to convert ancient paintings to photo-realistic natural images. By doing so, the ancient painting processing problems become natural image processing problems and models trained on natural images can be directly applied to the transferred paintings. Specifically, we focus on Chinese ancient flower, bird and landscape paintings in this work. A novel Domain Style Transfer Network (DSTN) is proposed to transfer ancient paintings to natural images which employ a compound loss to ensure that the transferred paintings still maintain the color composition and content of the input paintings. The experiment results show that the transferred paintings generated by the DSTN have a better performance in both the human perceptual test and other image processing tasks than other state-of-art methods, indicating the authenticity of the transferred paintings and the superiority of the proposed method.
https://arxiv.org/abs/1901.00224
Over the last few years, deep learning techniques have yielded significant improvements in image inpainting. However, many of these techniques fail to reconstruct reasonable structures as they are commonly over-smoothed and/or blurry. This paper develops a new approach for image inpainting that does a better job of reproducing filled regions exhibiting fine details. We propose a two-stage adversarial model EdgeConnect that comprises of an edge generator followed by an image completion network. The edge generator hallucinates edges of the missing region (both regular and irregular) of the image, and the image completion network fills in the missing regions using hallucinated edges as a priori. We evaluate our model end-to-end over the publicly available datasets CelebA, Places2, and Paris StreetView, and show that it outperforms current state-of-the-art techniques quantitatively and qualitatively.
https://arxiv.org/abs/1901.00212
The goal of this paper is to implement a system, titled as Drone Map Creator (DMC) using Computer Vision techniques. DMC can process visual information from an HD camera in a drone and automatically create a map by stitching together visual information captured by a drone. The proposed approach employs the Speeded up robust features (SURF) method to detect the key points for each image frame; then the corresponding points between the frames are identified by maximizing the determinant of a Hessian matrix. Finally, two images are stitched together by using the identified points. Our results show that despite some limitations from the external environment, we could have successfully stitched images together along video sequences.
https://arxiv.org/abs/1901.00211
Strong worst-case performance bounds for episodic reinforcement learning exist but fortunately in practice RL algorithms perform much better than such bounds would predict. Algorithms and theory that provide strong problem-dependent bounds could help illuminate the key features of what makes a RL problem hard and reduce the barrier to using RL algorithms in practice. As a step towards this we derive an algorithm for finite horizon discrete MDPs and associated analysis that both yields state-of-the art worst-case regret bounds in the dominant terms and yields substantially tighter bounds if the RL environment has small environmental norm, which is a function of the variance of the next-state value functions. An important benefit of our algorithmic is that it does not require apriori knowledge of a bound on the environmental norm. As a result of our analysis, we also help address an open learning theory question~\cite{jiang2018open} about episodic MDPs with a constant upper-bound on the sum of rewards, providing a regret bound with no $H$-dependence in the leading term that scales a polynomial function of the number of episodes.
https://arxiv.org/abs/1901.00210
Multi-label classification aims to classify instances with discrete non-exclusive labels. Most approaches on multi-label classification focus on effective adaptation or transformation of existing binary and multi-class learning approaches but fail in modelling the joint probability of labels or do not preserve generalization abilities for unseen label combinations. To address these issues we propose a new multi-label classification scheme, LNEMLC - Label Network Embedding for Multi-Label Classification, that embeds the label network and uses it to extend input space in learning and inference of any base multi-label classifier. The approach allows capturing of labels’ joint probability at low computational complexity providing results comparable to the best methods reported in the literature. We demonstrate how the method reveals statistically significant improvements over the simple kNN baseline classifier. We also provide hints for selecting the robust configuration that works satisfactorily across data domains.
http://arxiv.org/abs/1812.02956
When designing a neural caption generator, a convolutional neural network can be used to extract image features. Is it possible to also use a neural language model to extract sentence prefix features? We answer this question by trying different ways to transfer the recurrent neural network and embedding layer from a neural language model to an image caption generator. We find that image caption generators with transferred parameters perform better than those trained from scratch, even when simply pre-training them on the text of the same captions dataset it will later be trained on. We also find that the best language models (in terms of perplexity) do not result in the best caption generators after transfer learning.
https://arxiv.org/abs/1901.01216
Convolutional neural networks (CNNs) are a widely used form of deep neural networks, introducing state-of-the-art results for different problems such as image classification, computer vision tasks, and speech recognition. However, CNNs are compute intensive, requiring billions of multiply-accumulate (MAC) operations per input. To reduce the number of MACs in CNNs, we propose a value prediction method that exploits the spatial correlation of zero-valued activations within the CNN output feature maps, thereby saving convolution operations. Our method reduces the number of MAC operations by 30.4%, averaged on three modern CNNs for ImageNet, with top-1 accuracy degradation of 1.7%, and top-5 accuracy degradation of 1.1%.
http://arxiv.org/abs/1807.10598
The potential of the nasal region for expression robust 3D face recognition is thoroughly investigated by a novel five-step algorithm. First, the nose tip location is coarsely detected and the face is segmented, aligned and the nasal region cropped. Then, a very accurate and consistent nasal landmarking algorithm detects seven keypoints on the nasal region. In the third step, a feature extraction algorithm based on the surface normals of Gabor-wavelet filtered depth maps is utilised and, then, a set of spherical patches and curves are localised over the nasal region to provide the feature descriptors. The last step applies a genetic algorithm-based feature selector to detect the most stable patches and curves over different facial expressions. The algorithm provides the highest reported nasal region-based recognition ranks on the FRGC, Bosphorus and BU-3DFE datasets. The results are comparable with, and in many cases better than, many state-of-the-art 3D face recognition algorithms, which use the whole facial domain. The proposed method does not rely on sophisticated alignment or denoising steps, is very robust when only one sample per subject is used in the gallery, and does not require a training step for the landmarking algorithm. this https URL
https://arxiv.org/abs/1901.00206
One of the most important tasks in network management is identifying different types of traffic flows. As a result, a type of management service, called Network Traffic Classifier (NTC), has been introduced. One type of NTCs that has gained huge attention in recent years applies deep learning on packets in order to classify flows. Internet is an imbalanced environment i.e., some classes of applications are a lot more populated than others e.g., HTTP. Additionally, one of the challenges in deep learning methods is that they do not perform well in imbalanced environments in terms of evaluation metrics such as precision, recall, and $\mathrm{F_1}$ measure. In order to solve this problem, we recommend the use of augmentation methods to balance the dataset. In this paper, we propose a novel data augmentation approach based on the use of Long Short Term Memory (LSTM) networks for generating traffic flow patterns and Kernel Density Estimation (KDE) for replicating the numerical features of each class. First, we use the LSTM network in order to learn and generate the sequence of packets in a flow for classes with less population. Then, we complete the features of the sequence with generating random values based on the distribution of a certain feature, which will be estimated using KDE. Finally, we compare the training of a Convolutional Recurrent Neural Network (CRNN) in large-scale imbalanced, sampled, and augmented datasets. The contribution of our augmentation scheme is then evaluated on all of the datasets through measurements of precision, recall, and F1 measure for every class of application. The results demonstrate that our scheme is well suited for network traffic flow datasets and improves the performance of deep learning algorithms when it comes to above-mentioned metrics.
https://arxiv.org/abs/1901.00204
AI Safety researchers attempting to align values of highly capable intelligent systems with those of humanity face a number of challenges including personal value extraction, multi-agent value merger and finally in-silico encoding. State-of-the-art research in value alignment shows difficulties in every stage in this process, but merger of incompatible preferences is a particularly difficult challenge to overcome. In this paper we assume that the value extraction problem will be solved and propose a possible way to implement an AI solution which optimally aligns with individual preferences of each user. We conclude by analyzing benefits and limitations of the proposed approach.
http://arxiv.org/abs/1901.01851
A basic task in the design of a robotic production cell is the relative placement of robot and workpiece. The fundamental requirement is that the robot can reach all process positions; only then one can think of further optimization. Therefore an algorithm that automatically places an object into the workspace is very desirable. However many iterative optimzation algorithms cannot guarantee that all intermediate steps are reachable, resulting in complicated procedures. We present a novel approach which extends a robot by a virtual prismatic joint - which measures the distance to the workspace - such that any TCP frames are reachable. This allows higher order nonlinear programming algorithms to be used for placement of an object alone as well as the optimal placement under some differentiable criterion.
http://arxiv.org/abs/1901.00168
Convolutional neural networks(CNNs) has become one of the primary algorithms for various computer vision tasks. Handwritten character recognition is a typical example of such task that has also attracted attention. CNN architectures such as LeNet and AlexNet have become very prominent over the last two decades however the spatial invariance of the different kernels has been a prominent issue till now. With the introduction of capsule networks, kernels can work together in consensus with one another with the help of dynamic routing, that combines individual opinions of multiple groups of kernels called capsules to employ equivariance among kernels. In the current work, we have implemented capsule network on handwritten Indic digits and character datasets to show its superiority over networks like LeNet. Furthermore, it has also been shown that they can boost the performance of other networks like LeNet and AlexNet.
https://arxiv.org/abs/1901.00166
Recent years have seen remarkable progress of text generation in different contexts, such as the most common setting of generating text from scratch, and the emerging paradigm of retrieval-and-rewriting. Text infilling, which fills missing text portions of a sentence or paragraph, is also of numerous use in real life, yet is under-explored. Previous work has focused on restricted settings by either assuming single word per missing portion or limiting to a single missing portion to the end of the text. This paper studies the general task of text infilling, where the input text can have an arbitrary number of portions to be filled, each of which may require an arbitrary unknown number of tokens. We study various approaches for the task, including a self-attention model with segment-aware position encoding and bidirectional context modeling. We create extensive supervised data by masking out text with varying strategies. Experiments show the self-attention model greatly outperforms others, creating a strong baseline for future research.
https://arxiv.org/abs/1901.00158
Existing pose estimation approaches can be categorized into single-stage and multi-stage methods. While a multi-stage architecture is seemingly more suitable for the task, the performance of current multi-stage methods is not as competitive as single-stage ones. This work studies this issue. We argue that the current unsatisfactory performance comes from various insufficient design in current methods. We propose several improvements on the architecture design, feature flow, and loss function. The resulting multi-stage network outperforms all previous works and obtains the best performance on COCO keypoint challenge 2018. The source code will be released.
https://arxiv.org/abs/1901.00148
Although end-to-end (E2E) learning has led to impressive progress on a variety of visual understanding tasks, it is often impeded by hardware constraints (e.g., GPU memory) and is prone to overfitting. When it comes to video captioning, one of the most challenging benchmark tasks in computer vision, those limitations of E2E learning are especially amplified by the fact that both the input videos and output captions are lengthy sequences. Indeed, state-of-the-art methods for video captioning process video frames by convolutional neural networks and generate captions by unrolling recurrent neural networks. If we connect them in an E2E manner, the resulting model is both memory-consuming and data-hungry, making it extremely hard to train. In this paper, we propose a multitask reinforcement learning approach to training an E2E video captioning model. The main idea is to mine and construct as many effective tasks (e.g., attributes, rewards, and the captions) as possible from the human captioned videos such that they can jointly regulate the search space of the E2E neural network, from which an E2E video captioning model can be found and generalized to the testing phase. To the best of our knowledge, this is the first video captioning model that is trained end-to-end from the raw video input to the caption output. Experimental results show that such a model outperforms existing ones to a large margin on two benchmark video captioning datasets.
https://arxiv.org/abs/1803.07950
We propose a new long video dataset (called Track Long and Prosper - TLP) and benchmark for single object tracking. The dataset consists of 50 HD videos from real world scenarios, encompassing a duration of over 400 minutes (676K frames), making it more than 20 folds larger in average duration per sequence and more than 8 folds larger in terms of total covered duration, as compared to existing generic datasets for visual tracking. The proposed dataset paves a way to suitably assess long term tracking performance and train better deep learning architectures (avoiding/reducing augmentation, which may not reflect real world behaviour). We benchmark the dataset on 17 state of the art trackers and rank them according to tracking accuracy and run time speeds. We further present thorough qualitative and quantitative evaluation highlighting the importance of long term aspect of tracking. Our most interesting observations are (a) existing short sequence benchmarks fail to bring out the inherent differences in tracking algorithms which widen up while tracking on long sequences and (b) the accuracy of trackers abruptly drops on challenging long sequences, suggesting the potential need of research efforts in the direction of long-term tracking.
http://arxiv.org/abs/1712.01358
Due to recent advances in digital technologies, and availability of credible data, an area of artificial intelligence, deep learning, has emerged, and has demonstrated its ability and effectiveness in solving complex learning problems not possible before. In particular, convolution neural networks (CNNs) have demonstrated their effectiveness in image detection and recognition applications. However, they require intensive CPU operations and memory bandwidth that make general CPUs fail to achieve desired performance levels. Consequently, hardware accelerators that use application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), and graphic processing units (GPUs) have been employed to improve the throughput of CNNs. More precisely, FPGAs have been recently adopted for accelerating the implementation of deep learning networks due to their ability to maximize parallelism as well as due to their energy efficiency. In this paper, we review recent existing techniques for accelerating deep learning networks on FPGAs. We highlight the key features employed by the various techniques for improving the acceleration performance. In addition, we provide recommendations for enhancing the utilization of FPGAs for CNNs acceleration. The techniques investigated in this paper represent the recent trends in FPGA-based accelerators of deep learning networks. Thus, this review is expected to direct the future advances on efficient hardware accelerators and to be useful for deep learning researchers.
https://arxiv.org/abs/1901.00121
Different types of Convolutional Neural Networks (CNNs) have been applied to detect cancerous lung nodules from computed tomography (CT) scans. However, the size of a nodule is very diverse and can range anywhere between 3 and 30 millimeters. The high variation of nodule sizes makes classifying them a difficult and challenging task. In this study, we propose a novel CNN architecture called Gated-Dilated (GD) Networks to classify nodules as malignant or benign. Unlike previous studies, the GD network uses multiple dilated convolutions instead of max-poolings to capture the scale variations. Moreover, the GD network has a Context-Aware sub-network that analyzes the input features and guides the features to a suitable dilated convolution. We evaluated the proposed network on more than 1,000 CT scans from the LIDC-LDRI dataset. Our proposed network outperforms baseline models including conventional CNNs, Resnet, and Densenet, with an AUC of >0.95. Compared to the baseline models, the GD network improves the classification accuracies of mid-range sized nodules. Furthermore, we observe a relationship between the size of the nodule and the attention signal generated by the Context-Aware sub-network, which validates our new network architecture.
https://arxiv.org/abs/1901.00120
Robust Policy Search is the problem of learning policies that do not degrade in performance when subject to unseen environment model parameters. It is particularly relevant for transferring policies learned in a simulation environment to the real world. Several existing approaches involve sampling large batches of trajectories which reflect the differences in various possible environments, and then selecting some subset of these to learn robust policies, such as the ones that result in the worst performance. We propose an active learning based framework, EffAcTS, to selectively choose model parameters for this purpose so as to collect only as much data as necessary to select such a subset. We apply this framework to an existing method, namely EPOpt, and experimentally validate the gains in sample efficiency and the performance of our approach on standard continuous control tasks. We also present a Multi-Task Learning perspective to the problem of Robust Policy Search, and draw connections from our proposed framework to existing work on Multi-Task Learning.
http://arxiv.org/abs/1901.00117
Designing a driving policy for autonomous vehicles is a difficult task. Recent studies suggested an end-toend (E2E) training of a policy to predict car actuators directly from raw sensory inputs. It is appealing due to the ease of labeled data collection and since handcrafted features are avoided. Explicit drawbacks such as interpretability, safety enforcement and learning efficiency limit the practical application of the approach. In this paper, we amend the basic E2E architecture to address these shortcomings, while retaining the power of end-to-end learning. A key element in our proposed architecture is formulation of the learning problem as learning of trajectory. We also apply a Gaussian mixture model loss to contend with multi-modal data, and adopt a finance risk measure, conditional value at risk, to emphasize rare events. We analyze the effect of each concept and present driving performance in a highway scenario in the TORCS simulator. Video is available in this link: https://www.youtube.com/watch?v=1JYNBZNOe_4
http://arxiv.org/abs/1901.00114
Artificial neural networks are built on the basic operation of linear combination and non-linear activation function. Theoretically this structure can approximate any continuous function with three layer architecture. But in practice learning the parameters of such network can be hard. Also the choice of activation function can greatly impact the performance of the network. In this paper we are proposing to replace the basic linear combination operation with non-linear operations that do away with the need of additional non-linear activation function. To this end we are proposing the use of elementary morphological operations (dilation and erosion) as the basic operation in neurons. We show that these networks (Denoted as DenMo-Net) with morphological operations can approximate any smooth function requiring less number of parameters than what is necessary for normal neural networks. The results show that our network perform favorably when compared with similar structured network.
https://arxiv.org/abs/1901.00109
Deep neural networks (DNNs) have been widely used in the fields such as natural language processing, computer vision and image recognition. But several studies have been shown that deep neural networks can be easily fooled by artificial examples with some perturbations, which are widely known as adversarial examples. Adversarial examples can be used to attack deep neural networks or to improve the robustness of deep neural networks. A common way of generating adversarial examples is to first generate some noises and then add them into original examples. In practice, different examples have different noise-sensitive. To generate an effective adversarial example, it may be necessary to add a lot of noise to low noise-sensitive example, which may make the adversarial example meaningless. In this paper, we propose a noise-sensitivity-analysis-based test prioritization technique to pick out examples by their noise sensitivity. We construct an experiment to validate our approach on four image sets and two DNN models, which shows that examples are sensitive to noise and our method can effectively pick out examples by their noise sensitivity.
https://arxiv.org/abs/1901.00054
In this paper, we introduce a new probabilistically safe local steering primitive for sampling-based motion planning in complex high-dimensional configuration spaces. Our local steering procedure is based on a new notion of a convex probabilistically safe corridor that is constructed around a configuration using tangent hyperplanes of confidence ellipsoids of Gaussian mixture models learned from prior collision history. Accordingly, we propose to expand a random motion planning graph towards a sample goal using its projection onto probabilistically safe corridors, which efficiently exploits the local geometry of configuration spaces for selecting proper steering direction and adapting steering stepsize. We observe that the proposed local steering procedure generates effective steering motion around difficult regions of configuration spaces, such as narrow passages, while minimizing collision likelihood. We evaluate the proposed steering method with randomized motion planners in a number of planning scenarios, both in simulation and on a physical 7DoF robot arm, demonstrating the effectiveness of our safety guided local planner over the standard straight-line planner.
http://arxiv.org/abs/1901.00101
High-performance visual recognition systems generally require a large collection of labeled images to train. The expensive data curation can be an obstacle for improving recognition performance. Sharing more data allows training for better models. But personal and private information in the data prevent such sharing. To promote sharing visual data for learning a recognition model, we propose to obfuscate the images so that humans are not able to recognize their detailed contents, while machines can still utilize them to train new models. We validate our approach by comprehensive experiments on three challenging visual recognition tasks; image classification, attribute classification, and facial landmark detection on several datasets including SVHN, CIFAR10, Pascal VOC 2012, CelebA, and MTFL. Our method successfully obfuscates the images from humans recognition, but a machine model trained with them performs within about 1% margin (up to 0.48%) of the performance of a model trained with the original, non-obfuscated data.
https://arxiv.org/abs/1901.00098
An ideal description for a given video should fix its gaze on salient and representative content, which is capable of distinguishing this video from others. However, the distribution of different words is unbalanced in video captioning datasets, where distinctive words for describing video-specific salient objects are far less than common words such as ‘a’ ‘the’ and ‘person’. The dataset bias often results in recognition error or detail deficiency of salient but unusual objects. To address this issue, we propose a novel learning strategy called Information Loss, which focuses on the relationship between the video-specific visual content and corresponding representative words. Moreover, a framework with hierarchical visual representations and an optimized hierarchical attention mechanism is established to capture the most salient spatial-temporal visual information, which fully exploits the potential strength of the proposed learning strategy. Extensive experiments demonstrate that the ingenious guidance strategy together with the optimized architecture outperforms state-of-the-art video captioning methods on MSVD with CIDEr score 87.5, and achieves superior CIDEr score 47.7 on MSR-VTT. We also show that our Information Loss is generic which improves various models by significant margins.
https://arxiv.org/abs/1901.00097
The emergence of biometric tools and its increased usage in day to day devices has brought simplicity in the authentication process for the users as compared to the passwords and pattern locks being used. The ease of use of biometric reduces the manual work and helps in faster and automatic authentication. Among different biometric traits, the face is one which can be captured without much cooperation of users. Moreover, face recognition is one of the most widely publicized feature in the devices today and hence represents an important problem that should be studied with the utmost priority. As per the recent trends, the Convolutional Neural Network (CNN) based approaches are highly successful in many tasks of Computer Vision including face recognition. The loss function is used on the top of CNN to judge the goodness of any network. The loss functions play an important role in CNN training. Basically, it generates a huge loss, if the network does not perform well using the current parameter setting. In this paper, we present a performance comparison of different loss functions such as Cross-Entropy, Angular Softmax, Additive-Margin Softmax, ArcFace and Marginal Loss for face recognition. The experiments are conducted with two CNN architectures namely, ResNet and MobileNet. Two widely used face datasets namely, CASIA-Webface and MS-Celeb-1M are used for the training and benchmark Labeled Faces in the Wild (LFW) face dataset is used for the testing. The training and test results are analyzed in this paper.
https://arxiv.org/abs/1901.05903
The activation function in neural network is one of the important aspects which facilitates the deep training by introducing the non-linearity into the learning process. However, because of zero-hard rectification, some the of existing activations function such as ReLU and Swish miss to utilize the negative input values and may suffer from the dying gradient problem. Thus, it is important to look for a better activation function which is free from such problems. As a remedy, this paper proposes a new non-parametric function, called Linearly Scaled Hyperbolic Tangent (LiSHT) for Neural Networks (NNs). The proposed LiSHT activation function is an attempt to scale the non-linear Hyperbolic Tangent (Tanh) function by a linear function and tackle the dying gradient problem. The training and classification experiments are performed over benchmark Car Evaluation, Iris, MNIST, CIFAR10, CIFAR100 and twitter140 datasets to show that the proposed activation achieves faster convergence and higher performance. A very promising performance improvement is observed on three different type of neural networks including Multi-layer Perceptron (MLP), Convolutional Neural Network (CNN) and Recurrent neural network like Long-short term memory (LSTM). The advantages of proposed activation function are also visualized in terms of the feature activation maps, weight distribution and loss landscape.
https://arxiv.org/abs/1901.05894
The workloads running in the modern data centers of large scale Internet service providers (such as Amazon, Baidu, Facebook, Google, and Microsoft) support billions of users and span globally distributed infrastructure. Yet, the devices used in modern data centers fail due to a variety of causes, from faulty components to bugs to misconfiguration. Faulty devices make operating large scale data centers challenging because the workloads running in modern data centers consist of interdependent programs distributed across many servers, so failures that are isolated to a single device can still have a widespread effect on a workload. In this dissertation, we measure and model the device failures in a large scale Internet service company, Facebook. We focus on three device types that form the foundation of Internet service data center infrastructure: DRAM for main memory, SSDs for persistent storage, and switches and backbone links for network connectivity. For each of these device types, we analyze long term device failure data broken down by important device attributes and operating conditions, such as age, vendor, and workload. We also build and release statistical models to examine the failure trends for the devices we analyze. Our key conclusion in this dissertation is that we can gain a deep understanding of why devices fail—and how to predict their failure—using measurement and modeling. We hope that the analysis, techniques, and models we present in this dissertation will enable the community to better measure, understand, and prepare for the hardware reliability challenges we face in the future.
https://arxiv.org/abs/1901.03401
The CREST 2018 workshop is the third in a series of workshops addressing formal approaches to reasoning about causation in systems engineering. The topic of formally identifying the cause(s) of specific events - usually some form of failures -, and explaining why they occurred, are increasingly in the focus of several, disjoint communities. The main objective of CREST is to bring together researchers and practitioners from industry and academia in order to enable discussions how explicit and implicit reasoning about causation is performed. A further objective is to link to the foundations of causal reasoning in the philosophy of sciences and to causal reasoning performed in other areas of computer science, engineering, and beyond.
https://arxiv.org/abs/1901.00073
Triangular, overlapping Mel-scaled filters (“f-banks”) are the current standard input for acoustic models that exploit their input’s time-frequency geometry, because they provide a psycho-acoustically motivated time-frequency geometry for a speech signal. F-bank coefficients are provably robust to small deformations in the scale. In this paper, we explore two ways in which filter banks can be adjusted for the purposes of speech recognition. First, triangular filters can be replaced with Gabor filters, a compactly supported filter that better localizes events in time, or Gammatone filters, a psychoacoustically-motivated filter. Second, by rearranging the order of operations in computing filter bank features, features can be integrated over smaller time scales while simultaneously providing better frequency resolution. We make all feature implementations available online through open-source repositories. Initial experimentation with a modern end-to-end CNN phone recognizer yielded no significant improvements to phone error rate due to either modification. The result, and its ramifications with respect to learned filter banks, is discussed.
https://arxiv.org/abs/1901.00072
In this paper, we describe how to efficiently implement an acoustic room simulator to generate large-scale simulated data for training deep neural networks. Even though Google Room Simulator in [1] was shown to be quite effective in reducing the Word Error Rates (WERs) for far-field applications by generating simulated far-field training sets, it requires a very large number of Fast Fourier Transforms (FFTs) of large size. Room Simulator in [1] used approximately 80 percent of Central Processing Unit (CPU) usage in our CPU + Graphics Processing Unit (GPU) training architecture [2]. In this work, we implement an efficient OverLap Addition (OLA) based filtering using the open-source FFTW3 library. Further, we investigate the effects of the Room Impulse Response (RIR) lengths. Experimentally, we conclude that we can cut the tail portions of RIRs whose power is less than 20 dB below the maximum power without sacrificing the speech recognition accuracy. However, we observe that cutting RIR tail more than this threshold harms the speech recognition accuracy for rerecorded test sets. Using these approaches, we were able to reduce CPU usage for the room simulator portion down to 9.69 percent in CPU/GPU training architecture. Profiling result shows that we obtain 22.4 times speed-up on a single machine and 37.3 times speed up on Google’s distributed training infrastructure.
http://arxiv.org/abs/1712.03439
In Natural Language Processing (NLP), we often need to extract information from tree topology. Sentence structure can be represented via a dependency tree or a constituency tree structure. For this reason, a variant of LSTMs, named Tree-LSTM, was proposed to work on tree topology. In this paper, we design a generalized attention framework for both dependency and constituency trees by encoding variants of decomposable attention inside a Tree-LSTM cell. We evaluated our models on a semantic relatedness task and achieved notable results compared to Tree-LSTM based methods with no attention as well as other neural and non-neural methods and good results compared to Tree-LSTM based methods with attention.
https://arxiv.org/abs/1901.00066
Throughout this chapter, we provide theoretical insights into why and how deep learning can generalize well, despite its large capacity, complexity, possible algorithmic instability, nonrobustness, and sharp minima, responding to an open question in the literature. We also propose new open problems and discuss the limitations of our results.
http://arxiv.org/abs/1710.05468
Computer vision relies on labeled datasets for training and evaluation in detecting and recognizing objects. The popular computer vision program, YOLO (“You Only Look Once”), has been shown to accurately detect objects in many major image datasets. However, the images found in those datasets, are independent of one another and cannot be used to test YOLO’s consistency at detecting the same object as its environment (e.g. ambient lighting) changes. This paper describes a novel effort to evaluate YOLO’s consistency for large-scale applications. It does so by working (a) at large scale and (b) by using consecutive images from a curated network of public video cameras deployed in a variety of real-world situations, including traffic intersections, national parks, shopping malls, university campuses, etc. We specifically examine YOLO’s ability to detect objects in different scenarios (e.g., daytime vs. night), leveraging the cameras’ ability to rapidly retrieve many successive images for evaluating detection consistency. Using our camera network and advanced computing resources (supercomputers), we analyzed more than 5 million images captured by 140 network cameras in 24 hours. Compared with labels marked by humans (considered as “ground truth”), YOLO struggles to consistently detect the same humans and cars as their positions change from one frame to the next; it also struggles to detect objects at night time. Our findings suggest that state-of-the art vision solutions should be trained by data from network camera with contextual information before they can be deployed in applications that demand high consistency on object detection.
https://arxiv.org/abs/1812.11901
In the last few years, generative adversarial networks (GAN) have shown tremendous potential for a number of applications in computer vision and related fields. With the current pace of progress, it is a sure bet they will soon be able to generate high-quality images and videos, virtually indistinguishable from real ones. Unfortunately, realistic GAN-generated images pose serious threats to security, to begin with a possible flood of fake multimedia, and multimedia forensic countermeasures are in urgent need. In this work, we show that each GAN leaves its specific fingerprint in the images it generates, just like real-world cameras mark acquired images with traces of their photo-response non-uniformity pattern. Source identification experiments with several popular GANs show such fingerprints to represent a precious asset for forensic analyses.
https://arxiv.org/abs/1812.11842
Siamese network based trackers formulate tracking as convolutional feature cross-correlation between target template and searching region. However, Siamese trackers still have accuracy gap compared with state-of-the-art algorithms and they cannot take advantage of feature from deep networks, such as ResNet-50 or deeper. In this work we prove the core reason comes from the lack of strict translation invariance. By comprehensive theoretical analysis and experimental validations, we break this restriction through a simple yet effective spatial aware sampling strategy and successfully train a ResNet-driven Siamese tracker with significant performance gain. Moreover, we propose a new model architecture to perform depth-wise and layer-wise aggregations, which not only further improves the accuracy but also reduces the model size. We conduct extensive ablation studies to demonstrate the effectiveness of the proposed tracker, which obtains currently the best results on four large tracking benchmarks, including OTB2015, VOT2018, UAV123, and LaSOT. Our model will be released to facilitate further studies based on this problem.
https://arxiv.org/abs/1812.11703
Memory-Augmented Neural Networks (MANNs) are a class of neural networks equipped with an external memory, and are reported to be effective for tasks requiring a large long-term memory and its selective use. The core module of a MANN is called a controller, which is usually implemented as a recurrent neural network (RNN) (e.g., LSTM) to enable the use of contextual information in controlling the other modules. However, such an RNN-based controller often allows a MANN to directly solve the given task by using the (small) internal memory of the controller, and prevents the MANN from making the best use of the external memory, thereby resulting in a suboptimally trained model. To address this problem, we present a novel type of RNN-based controller that is partially non-recurrent and avoids the direct use of its internal memory for solving the task, while keeping the ability of using contextual information in controlling the other modules. Our empirical experiments using Neural Turing Machines and Differentiable Neural Computers on the Toy and bAbI tasks demonstrate that the proposed controllers give substantially better results than standard RNN-based controllers.
https://arxiv.org/abs/1812.11485
Adversarial learning methods have been proposed for a wide range of applications, but the training of adversarial models can be notoriously unstable. Effectively balancing the performance of the generator and discriminator is critical, since a discriminator that achieves very high accuracy will produce relatively uninformative gradients. In this work, we propose a simple and general technique to constrain information flow in the discriminator by means of an information bottleneck. By enforcing a constraint on the mutual information between the observations and the discriminator’s internal representation, we can effectively modulate the discriminator’s accuracy and maintain useful and informative gradients. We demonstrate that our proposed variational discriminator bottleneck (VDB) leads to significant improvements across three distinct application areas for adversarial learning algorithms. Our primary evaluation studies the applicability of the VDB to imitation learning of dynamic continuous control skills, such as running. We show that our method can learn such skills directly from \emph{raw} video demonstrations, substantially outperforming prior adversarial imitation learning methods. The VDB can also be combined with adversarial inverse reinforcement learning to learn parsimonious reward functions that can be transferred and re-optimized in new settings. Finally, we demonstrate that VDB can train GANs more effectively for image generation, improving upon a number of prior stabilization methods.
https://arxiv.org/abs/1810.00821
This letter describes the characterization of a low-offset Hall-effect plate using the AlGaN/GaN two-dimensional electron gas(2DEG). Four-phase current spinning was used to reduce sensor offset voltage to values in the range of 20 nV, which corresponds to a low residual offset of 2.6 micro-Tesla when supplied with low voltages (0.04 to 0.5V). These offsets are 50x smaller than the values previously reported for GaN Hall-effect plates, and it is on par with state-of-the-art silicon Hall-effect plates. In addition, the offset does not exceed 10 micro-Tesla even at higher supply voltage of 2.34V. The sensor also shows stable current-scaled sensitivity over a wide temperature range of -100C to 200C, with temperature drift of -125 ppm/C. This value is 3x better than state-of-the-art Silicon Hall-effect plates. Additionally, the sensor’s voltage sensitivity (57 mV/V/T) is also similar. Because of their low offset values, AlGaN/GaN Hall-effect plates are viable candidates for low-field and high temperature magnetic sensing in monolithic GaN systems used in extreme temperature environments such as power inverter, down-well, combustion, and space applications.
https://arxiv.org/abs/1812.00363
Search strategies for generating a response from a neural dialogue model have received relatively little attention compared to improving network architectures and learning algorithms in recent years. In this paper, we consider a standard neural dialogue model based on recurrent networks with an attention mechanism, and focus on evaluating the impact of the search strategy. We compare four search strategies: greedy search, beam search, iterative beam search and iterative beam search followed by selection scoring. We evaluate these strategies using human evaluation of full conversations and compare them using automatic metrics including log-probabilities, scores and diversity metrics. We observe a significant gap between greedy search and the proposed iterative beam search augmented with selection scoring, demonstrating the importance of the search algorithm in neural dialogue generation.
https://arxiv.org/abs/1811.00907
High quality upsampling of sparse 3D point clouds is critically useful for a wide range of geometric operations such as reconstruction, rendering, meshing, and analysis. In this paper, we propose a data-driven algorithm that enables an upsampling of 3D point clouds without the need for hard-coded rules. Our approach uses a deep network with Chamfer distance as the loss function, capable of learning the latent features in point clouds belonging to different object categories. We evaluate our algorithm across different amplification factors, with upsampling learned and performed on objects belonging to the same category as well as different categories. We also explore the desirable characteristics of input point clouds as a function of the distribution of the point samples. Finally, we demonstrate the performance of our algorithm in single-category training versus multi-category training scenarios. The final proposed model is compared against a baseline, optimization-based upsampling method. Results indicate that our algorithm is capable of generating more uniform and accurate upsamplings.
http://arxiv.org/abs/1807.02740
Quantum algorithm design lies in the hallmark of applications of quantum computation and quantum simulation. Here we put forward a deep reinforcement learning (RL) architecture for automated algorithm design in the framework of quantum adiabatic algorithm, where the optimal Hamiltonian path to reach a quantum ground state that encodes a compution problem is obtained by RL techniques. We benchmark our approach in Grover search and 3-SAT problems, and find that the adiabatic algorithm obtained by our RL approach leads to significant improvement in the success probability and computing speedups for both moderate and large number of qubits compared to conventional algorithms. The RL-designed algorithm is found to be qualitatively distinct from the linear algorithm in the resultant distribution of success probability. Considering the established complexity-equivalence of circuit and adiabatic quantum algorithms, we expect the RL-designed adiabatic algorithm to inspire novel circuit algorithms as well. Our approach offers a recipe to design quantum algorithms for generic problems through a machinery RL process, which paves a novel way to automated quantum algorithm design using artificial intelligence, potentially applicable to different quantum simulation and computation platforms from trapped ions and optical lattices to superconducting-qubit devices.
https://arxiv.org/abs/1812.10797
Tokenization or segmentation is a wide concept that covers simple processes such as separating punctuation from words, or more sophisticated processes such as applying morphological knowledge. Neural Machine Translation (NMT) requires a limited-size vocabulary for computational cost and enough examples to estimate word embeddings. Separating punctuation and splitting tokens into words or subwords has proven to be helpful to reduce vocabulary and increase the number of examples of each word, improving the translation quality. Tokenization is more challenging when dealing with languages with no separator between words. In order to assess the impact of the tokenization in the quality of the final translation on NMT, we experimented on five tokenizers over ten language pairs. We reached the conclusion that the tokenization significantly affects the final translation quality and that the best tokenizer differs for different language pairs.
https://arxiv.org/abs/1812.08621
Neural architecture search (NAS) enabled the discovery of state-of-the-art architectures in many domains. However, the success of NAS depends on the definition of the search space, i.e. the set of the possible to generate neural architectures. State-of-the-art search spaces are defined as a static sequence of decisions and a set of available actions for each decision, where each possible sequence of actions defines an architecture. We propose a more expressive formulation of NAS, using a graph search space. Our search space is defined as a graph where each decision is a vertex and each action is an edge. Thus the sequence of decisions defining an architecture is not fixed but is determined dynamically by the actions selected. The proposed approach allows to model iterative and branching aspects of the architecture design process. In this form, stronger priors about the search can be induced. We demonstrate in simulation basic iterative and branching search structures and show that using the graph representation improves sample efficiency.
https://arxiv.org/abs/1812.10666
Identifying applicable anode materials is a significant task for Li- and Na-ion battery (LIB and NIB) technologies. We propose the GaN monolayer (2D GaN) can be a good anode candidate. The GaN monolayer manifests stable Li/Na adsorption and inherently low theoretical voltages. Most excitingly, both high storage capacity and extremely fast Li/Na diffusion can be simultaneously realized in the GaN monolayer. For Li, the storage capacity and diffusion barrier is 938 mA h g-1 and 80 meV , respectively. And the values for Na are 625 mA h g-1 and 22 meV. Comparing with known 2D anode materials with similar scale of ion diffusion barriers, the GaN monolayer almost possesses the highest Li/Na storage capacity discovered to date. Our work suggests that the 2D GaN is a prospective anode material offering fast ion diffusion and high storage capacity.
https://arxiv.org/abs/1812.10610
We introduce an architecture to learn joint multilingual sentence representations for 93 languages, belonging to more than 30 different language families and written in 28 different scripts. Our system uses a single BiLSTM encoder with a shared BPE vocabulary for all languages, which is coupled with an auxiliary decoder and trained on publicly available parallel corpora. This enables us to learn a classifier on top of the resulting sentence embeddings using English annotated data only, and transfer it to any of the 93 languages without any modification. Our approach sets a new state-of-the-art on zero-shot cross-lingual natural language inference for all the 14 languages in the XNLI dataset but one. We also achieve very competitive results in cross-lingual document classification (MLDoc dataset). Our sentence embeddings are also strong at parallel corpus mining, establishing a new state-of-the-art in the BUCC shared task for 3 of its 4 language pairs. Finally, we introduce a new test set of aligned sentences in 122 languages based on the Tatoeba corpus, and show that our sentence embeddings obtain strong results in multilingual similarity search even for low-resource languages. Our PyTorch implementation, pre-trained encoder and the multilingual test set will be freely available.
https://arxiv.org/abs/1812.10464