October 20, 2019

3189 words 15 mins read

Paper Group AWR 343

Learning Contextual Bandits in a Non-stationary Environment. DELIMIT PyTorch - An extension for Deep Learning in Diffusion Imaging. The Tsetlin Machine - A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic. DeepChannel: Salience Estimation by Contrastive Learning for Extractive Document Summarization. Gra …

Learning Contextual Bandits in a Non-stationary Environment


Title	Learning Contextual Bandits in a Non-stationary Environment
Authors	Qingyun Wu, Naveen Iyer, Hongning Wang
Abstract	Multi-armed bandit algorithms have become a reference solution for handling the explore/exploit dilemma in recommender systems, and many other important real-world problems, such as display advertisement. However, such algorithms usually assume a stationary reward distribution, which hardly holds in practice as users’ preferences are dynamic. This inevitably costs a recommender system consistent suboptimal performance. In this paper, we consider the situation where the underlying distribution of reward remains unchanged over (possibly short) epochs and shifts at unknown time instants. In accordance, we propose a contextual bandit algorithm that detects possible changes of environment based on its reward estimation confidence and updates its arm selection strategy respectively. Rigorous upper regret bound analysis of the proposed algorithm demonstrates its learning effectiveness in such a non-trivial environment. Extensive empirical evaluations on both synthetic and real-world datasets for recommendation confirm its practical utility in a changing environment.
Tasks	Multi-Armed Bandits, Recommendation Systems
Published	2018-05-23
URL	http://arxiv.org/abs/1805.09365v1
PDF	http://arxiv.org/pdf/1805.09365v1.pdf
PWC	https://paperswithcode.com/paper/learning-contextual-bandits-in-a-non
Repo	https://github.com/YRussac/WeightedLinearBandits
Framework	none

DELIMIT PyTorch - An extension for Deep Learning in Diffusion Imaging


Title	DELIMIT PyTorch - An extension for Deep Learning in Diffusion Imaging
Authors	Simon Koppers, Dorit Merhof
Abstract	DELIMIT is a framework extension for deep learning in diffusion imaging, which extends the basic framework PyTorch towards spherical signals. Based on several novel layers, deep learning can be applied to spherical diffusion imaging data in a very convenient way. First, two spherical harmonic interpolation layers are added to the extension, which allow to transform the signal from spherical surface space into the spherical harmonic space, and vice versa. In addition, a local spherical convolution layer is introduced that adds the possibility to include gradient neighborhood information within the network. Furthermore, these extensions can also be utilized for the preprocessing of diffusion signals.
Tasks
Published	2018-08-04
URL	http://arxiv.org/abs/1808.01517v1
PDF	http://arxiv.org/pdf/1808.01517v1.pdf
PWC	https://paperswithcode.com/paper/delimit-pytorch-an-extension-for-deep
Repo	https://github.com/SimonKoppers/DELIMIT
Framework	pytorch

The Tsetlin Machine - A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic


Title	The Tsetlin Machine - A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic
Authors	Ole-Christoffer Granmo
Abstract	Although simple individually, artificial neurons provide state-of-the-art performance when interconnected in deep networks. Unknown to many, there exists an arguably even simpler and more versatile learning mechanism, namely, the Tsetlin Automaton. Merely by means of a single integer as memory, it learns the optimal action in stochastic environments through increment and decrement operations. In this paper, we introduce the Tsetlin Machine, which solves complex pattern recognition problems with easy-to-interpret propositional formulas, composed by a collective of Tsetlin Automata. To eliminate the longstanding problem of vanishing signal-to-noise ratio, the Tsetlin Machine orchestrates the automata using a novel game. Our theoretical analysis establishes that the Nash equilibria of the game align with the propositional formulas that provide optimal pattern recognition accuracy. This translates to learning without local optima, only global ones. We argue that the Tsetlin Machine finds the propositional formula that provides optimal accuracy, with probability arbitrarily close to unity. In five benchmarks, the Tsetlin Machine provides competitive accuracy compared with SVMs, Decision Trees, Random Forests, Naive Bayes Classifier, Logistic Regression, and Neural Networks. The Tsetlin Machine further has an inherent computational advantage since both inputs, patterns, and outputs are expressed as bits, while recognition and learning rely on bit manipulation. The combination of accuracy, interpretability, and computational simplicity makes the Tsetlin Machine a promising tool for a wide range of domains. Being the first of its kind, we believe the Tsetlin Machine will kick-start new paths of research, with a potentially significant impact on the AI field and the applications of AI.
Tasks
Published	2018-04-04
URL	http://arxiv.org/abs/1804.01508v10
PDF	http://arxiv.org/pdf/1804.01508v10.pdf
PWC	https://paperswithcode.com/paper/the-tsetlin-machine-a-game-theoretic-bandit
Repo	https://github.com/cair/regression-tsetlin-machine
Framework	none

DeepChannel: Salience Estimation by Contrastive Learning for Extractive Document Summarization


Title	DeepChannel: Salience Estimation by Contrastive Learning for Extractive Document Summarization
Authors	Jiaxin Shi, Chen Liang, Lei Hou, Juanzi Li, Zhiyuan Liu, Hanwang Zhang
Abstract	We propose DeepChannel, a robust, data-efficient, and interpretable neural model for extractive document summarization. Given any document-summary pair, we estimate a salience score, which is modeled using an attention-based deep neural network, to represent the salience degree of the summary for yielding the document. We devise a contrastive training strategy to learn the salience estimation network, and then use the learned salience score as a guide and iteratively extract the most salient sentences from the document as our generated summary. In experiments, our model not only achieves state-of-the-art ROUGE scores on CNN/Daily Mail dataset, but also shows strong robustness in the out-of-domain test on DUC2007 test set. Moreover, our model reaches a ROUGE-1 F-1 score of 39.41 on CNN/Daily Mail test set with merely $1 / 100$ training set, demonstrating a tremendous data efficiency.
Tasks	Document Summarization, Extractive Document Summarization
Published	2018-11-06
URL	http://arxiv.org/abs/1811.02394v2
PDF	http://arxiv.org/pdf/1811.02394v2.pdf
PWC	https://paperswithcode.com/paper/deepchannel-salience-estimation-by
Repo	https://github.com/lliangchenc/DeepChannel
Framework	pytorch

Graph Laplacian for Image Anomaly Detection


Title	Graph Laplacian for Image Anomaly Detection
Authors	Francesco Verdoja, Marco Grangetto
Abstract	Reed-Xiaoli detector (RXD) is recognized as the benchmark algorithm for image anomaly detection; however, it presents known limitations, namely the dependence over the image following a multivariate Gaussian model, the estimation and inversion of a high-dimensional covariance matrix, and the inability to effectively include spatial awareness in its evaluation. In this work, a novel graph-based solution to the image anomaly detection problem is proposed; leveraging the graph Fourier transform, we are able to overcome some of RXD’s limitations while reducing computational cost at the same time. Tests over both hyperspectral and medical images, using both synthetic and real anomalies, prove the proposed technique is able to obtain significant gains over performance by other algorithms in the state of the art.
Tasks	Anomaly Detection
Published	2018-02-27
URL	https://arxiv.org/abs/1802.09843v6
PDF	https://arxiv.org/pdf/1802.09843v6.pdf
PWC	https://paperswithcode.com/paper/graph-laplacian-for-image-anomaly-detection
Repo	https://github.com/fverdoja/LAD-Laplacian-Anomaly-Detector
Framework	none

Adversarial Attacks Against Medical Deep Learning Systems


Title	Adversarial Attacks Against Medical Deep Learning Systems
Authors	Samuel G. Finlayson, Hyung Won Chung, Isaac S. Kohane, Andrew L. Beam
Abstract	The discovery of adversarial examples has raised concerns about the practical deployment of deep learning systems. In this paper, we demonstrate that adversarial examples are capable of manipulating deep learning systems across three clinical domains. For each of our representative medical deep learning classifiers, both white and black box attacks were highly successful. Our models are representative of the current state of the art in medical computer vision and, in some cases, directly reflect architectures already seeing deployment in real world clinical settings. In addition to the technical contribution of our paper, we synthesize a large body of knowledge about the healthcare system to argue that medicine may be uniquely susceptible to adversarial attacks, both in terms of monetary incentives and technical vulnerability. To this end, we outline the healthcare economy and the incentives it creates for fraud and provide concrete examples of how and why such attacks could be realistically carried out. We urge practitioners to be aware of current vulnerabilities when deploying deep learning systems in clinical settings, and encourage the machine learning community to further investigate the domain-specific characteristics of medical learning systems.
Tasks
Published	2018-04-15
URL	http://arxiv.org/abs/1804.05296v3
PDF	http://arxiv.org/pdf/1804.05296v3.pdf
PWC	https://paperswithcode.com/paper/adversarial-attacks-against-medical-deep
Repo	https://github.com/sgfin/adversarial-medicine
Framework	tf

auditor: an R Package for Model-Agnostic Visual Validation and Diagnostic


Title	auditor: an R Package for Model-Agnostic Visual Validation and Diagnostic
Authors	Alicja Gosiewska, Przemyslaw Biecek
Abstract	Machine learning has spread to almost every area of life. It is successfully applied in biology, medicine, finance, physics, and other fields. The problem arises if models fail when confronted with the real-world data. Therefore, there is a need for validation methods. This paper describes methodology and tools for model-agnostic audit. Introduced techniques facilitate assessing and comparing the goodness of fit and performance of models. In addition, they may be used for analysis of the similarity of residuals and for the identification of outliers and influential observations. The examination is carried out by diagnostic scores and visual verification. Presented methods are implemented in the auditor package for R. Due to the flexible and consistent grammar, it is simple to validate models of any classes.
Tasks
Published	2018-09-19
URL	http://arxiv.org/abs/1809.07763v3
PDF	http://arxiv.org/pdf/1809.07763v3.pdf
PWC	https://paperswithcode.com/paper/auditor-an-r-package-for-model-agnostic
Repo	https://github.com/MI2DataLab/auditor
Framework	none

Left ventricle quantification through spatio-temporal CNNs


Title	Left ventricle quantification through spatio-temporal CNNs
Authors	Alejandro Debus, Enzo Ferrante
Abstract	Cardiovascular diseases are among the leading causes of death globally. Cardiac left ventricle (LV) quantification is known to be one of the most important tasks for the identification and diagnosis of such pathologies. In this paper, we propose a deep learning method that incorporates 3D spatio-temporal convolutions to perform direct left ventricle quantification from cardiac MR sequences. Instead of analysing slices independently, we process stacks of temporally adjacent slices by means of 3D convolutional kernels which fuse the spatio-temporal information, incorporating the temporal dynamics of the heart to the learned model. We show that incorporating such information by means of spatio-temporal convolutions into standard LV quantification architectures improves the accuracy of the predictions when compared with single-slice models, achieving competitive results for all cardiac indices and significantly breaking the state of the art (Xue et al., 2018, MedIA) for cardiac phase estimation.
Tasks
Published	2018-08-23
URL	http://arxiv.org/abs/1808.07967v1
PDF	http://arxiv.org/pdf/1808.07967v1.pdf
PWC	https://paperswithcode.com/paper/left-ventricle-quantification-through-spatio
Repo	https://github.com/alejandrodebus/IndicesNet
Framework	pytorch

The Alignment of the Spheres: Globally-Optimal Spherical Mixture Alignment for Camera Pose Estimation


Title	The Alignment of the Spheres: Globally-Optimal Spherical Mixture Alignment for Camera Pose Estimation
Authors	Dylan Campbell, Lars Petersson, Laurent Kneip, Hongdong Li, Stephen Gould
Abstract	Determining the position and orientation of a calibrated camera from a single image with respect to a 3D model is an essential task for many applications. When 2D-3D correspondences can be obtained reliably, perspective-n-point solvers can be used to recover the camera pose. However, without the pose it is non-trivial to find cross-modality correspondences between 2D images and 3D models, particularly when the latter only contains geometric information. Consequently, the problem becomes one of estimating pose and correspondences jointly. Since outliers and local optima are so prevalent, robust objective functions and global search strategies are desirable. Hence, we cast the problem as a 2D-3D mixture model alignment task and propose the first globally-optimal solution to this formulation under the robust $L_2$ distance between mixture distributions. We search the 6D camera pose space using branch-and-bound, which requires novel bounds, to obviate the need for a pose estimate and guarantee global optimality. To accelerate convergence, we integrate local optimization, implement GPU bound computations, and provide an intuitive way to incorporate side information such as semantic labels. The algorithm is evaluated on challenging synthetic and real datasets, outperforming existing approaches and reliably converging to the global optimum.
Tasks	Pose Estimation
Published	2018-12-04
URL	https://arxiv.org/abs/1812.01232v2
PDF	https://arxiv.org/pdf/1812.01232v2.pdf
PWC	https://paperswithcode.com/paper/the-alignment-of-the-spheres-globally-optimal
Repo	https://github.com/Awesome-Image-Registration-Organization/2D-3D-matching
Framework	none

Gated Context Aggregation Network for Image Dehazing and Deraining


Title	Gated Context Aggregation Network for Image Dehazing and Deraining
Authors	Dongdong Chen, Mingming He, Qingnan Fan, Jing Liao, Liheng Zhang, Dongdong Hou, Lu Yuan, Gang Hua
Abstract	Image dehazing aims to recover the uncorrupted content from a hazy image. Instead of leveraging traditional low-level or handcrafted image priors as the restoration constraints, e.g., dark channels and increased contrast, we propose an end-to-end gated context aggregation network to directly restore the final haze-free image. In this network, we adopt the latest smoothed dilation technique to help remove the gridding artifacts caused by the widely-used dilated convolution with negligible extra parameters, and leverage a gated sub-network to fuse the features from different levels. Extensive experiments demonstrate that our method can surpass previous state-of-the-art methods by a large margin both quantitatively and qualitatively. In addition, to demonstrate the generality of the proposed method, we further apply it to the image deraining task, which also achieves the state-of-the-art performance. Code has been made available at https://github.com/cddlyf/GCANet.
Tasks	Image Dehazing, Rain Removal
Published	2018-11-21
URL	http://arxiv.org/abs/1811.08747v2
PDF	http://arxiv.org/pdf/1811.08747v2.pdf
PWC	https://paperswithcode.com/paper/gated-context-aggregation-network-for-image
Repo	https://github.com/cddlyf/GCANet
Framework	pytorch

Deep Bayesian Active Semi-Supervised Learning


Title	Deep Bayesian Active Semi-Supervised Learning
Authors	Matthias Rottmann, Karsten Kahl, Hanno Gottschalk
Abstract	In many applications the process of generating label information is expensive and time consuming. We present a new method that combines active and semi-supervised deep learning to achieve high generalization performance from a deep convolutional neural network with as few known labels as possible. In a setting where a small amount of labeled data as well as a large amount of unlabeled data is available, our method first learns the labeled data set. This initialization is followed by an expectation maximization algorithm, where further training reduces classification entropy on the unlabeled data by targeting a low entropy fit which is consistent with the labeled data. In addition the algorithm asks at a specified frequency an oracle for labels of data with entropy above a certain entropy quantile. Using this active learning component we obtain an agile labeling process that achieves high accuracy, but requires only a small amount of known labels. For the MNIST dataset we report an error rate of 2.06% using only 300 labels and 1.06% for 1000 labels. These results are obtained without employing any special network architecture or data augmentation.
Tasks	Active Learning, Data Augmentation
Published	2018-03-03
URL	http://arxiv.org/abs/1803.01216v1
PDF	http://arxiv.org/pdf/1803.01216v1.pdf
PWC	https://paperswithcode.com/paper/deep-bayesian-active-semi-supervised-learning
Repo	https://github.com/mrottmann/DeepBASS
Framework	tf

Progressive Feature Fusion Network for Realistic Image Dehazing


Title	Progressive Feature Fusion Network for Realistic Image Dehazing
Authors	Kangfu Mei, Aiwen Jiang, Juncheng Li, Mingwen Wang
Abstract	Single image dehazing is a challenging ill-posed restoration problem. Various prior-based and learning-based methods have been proposed. Most of them follow a classic atmospheric scattering model which is an elegant simplified physical model based on the assumption of single-scattering and homogeneous atmospheric medium. The formulation of haze in realistic environment is more complicated. In this paper, we propose to take its essential mechanism as “black box”, and focus on learning an input-adaptive trainable end-to-end dehazing model. An U-Net like encoder-decoder deep network via progressive feature fusions has been proposed to directly learn highly nonlinear transformation function from observed hazy image to haze-free ground-truth. The proposed network is evaluated on two public image dehazing benchmarks. The experiments demonstrate that it can achieve superior performance when compared with popular state-of-the-art methods. With efficient GPU memory usage, it can satisfactorily recover ultra high definition hazed image up to 4K resolution, which is unaffordable by many deep learning based dehazing algorithms.
Tasks	Image Dehazing, Single Image Dehazing
Published	2018-10-04
URL	http://arxiv.org/abs/1810.02283v1
PDF	http://arxiv.org/pdf/1810.02283v1.pdf
PWC	https://paperswithcode.com/paper/progressive-feature-fusion-network-for
Repo	https://github.com/MKFMIKU/PFFNet
Framework	pytorch

Learning to Attend On Essential Terms: An Enhanced Retriever-Reader Model for Open-domain Question Answering


Title	Learning to Attend On Essential Terms: An Enhanced Retriever-Reader Model for Open-domain Question Answering
Authors	Jianmo Ni, Chenguang Zhu, Weizhu Chen, Julian McAuley
Abstract	Open-domain question answering remains a challenging task as it requires models that are capable of understanding questions and answers, collecting useful information, and reasoning over evidence. Previous work typically formulates this task as a reading comprehension or entailment problem given evidence retrieved from search engines. However, existing techniques struggle to retrieve indirectly related evidence when no directly related evidence is provided, especially for complex questions where it is hard to parse precisely what the question asks. In this paper we propose a retriever-reader model that learns to attend on essential terms during the question answering process. We build (1) an essential term selector which first identifies the most important words in a question, then reformulates the query and searches for related evidence; and (2) an enhanced reader that distinguishes between essential terms and distracting words to predict the answer. We evaluate our model on multiple open-domain multiple-choice QA datasets, notably performing at the level of the state-of-the-art on the AI2 Reasoning Challenge (ARC) dataset.
Tasks	Open-Domain Question Answering, Question Answering, Reading Comprehension
Published	2018-08-28
URL	https://arxiv.org/abs/1808.09492v5
PDF	https://arxiv.org/pdf/1808.09492v5.pdf
PWC	https://paperswithcode.com/paper/learning-to-attend-on-essential-terms-an
Repo	https://github.com/ZHO9504/Select-Key-Terms-in-a-Question
Framework	pytorch

Critical initialisation for deep signal propagation in noisy rectifier neural networks


Title	Critical initialisation for deep signal propagation in noisy rectifier neural networks
Authors	Arnu Pretorius, Elan Van Biljon, Steve Kroon, Herman Kamper
Abstract	Stochastic regularisation is an important weapon in the arsenal of a deep learning practitioner. However, despite recent theoretical advances, our understanding of how noise influences signal propagation in deep neural networks remains limited. By extending recent work based on mean field theory, we develop a new framework for signal propagation in stochastic regularised neural networks. Our noisy signal propagation theory can incorporate several common noise distributions, including additive and multiplicative Gaussian noise as well as dropout. We use this framework to investigate initialisation strategies for noisy ReLU networks. We show that no critical initialisation strategy exists using additive noise, with signal propagation exploding regardless of the selected noise distribution. For multiplicative noise (e.g. dropout), we identify alternative critical initialisation strategies that depend on the second moment of the noise distribution. Simulations and experiments on real-world data confirm that our proposed initialisation is able to stably propagate signals in deep networks, while using an initialisation disregarding noise fails to do so. Furthermore, we analyse correlation dynamics between inputs. Stronger noise regularisation is shown to reduce the depth to which discriminatory information about the inputs to a noisy ReLU network is able to propagate, even when initialised at criticality. We support our theoretical predictions for these trainable depths with simulations, as well as with experiments on MNIST and CIFAR-10
Tasks
Published	2018-11-01
URL	http://arxiv.org/abs/1811.00293v2
PDF	http://arxiv.org/pdf/1811.00293v2.pdf
PWC	https://paperswithcode.com/paper/critical-initialisation-for-deep-signal
Repo	https://github.com/ElanVB/noisy_signal_prop
Framework	tf

Deep Ordinal Regression Network for Monocular Depth Estimation


Title	Deep Ordinal Regression Network for Monocular Depth Estimation
Authors	Huan Fu, Mingming Gong, Chaohui Wang, Kayhan Batmanghelich, Dacheng Tao
Abstract	Monocular depth estimation, which plays a crucial role in understanding 3D scene geometry, is an ill-posed problem. Recent methods have gained significant improvement by exploring image-level information and hierarchical features from deep convolutional neural networks (DCNNs). These methods model depth estimation as a regression problem and train the regression networks by minimizing mean squared error, which suffers from slow convergence and unsatisfactory local solutions. Besides, existing depth estimation networks employ repeated spatial pooling operations, resulting in undesirable low-resolution feature maps. To obtain high-resolution depth maps, skip-connections or multi-layer deconvolution networks are required, which complicates network training and consumes much more computations. To eliminate or at least largely reduce these problems, we introduce a spacing-increasing discretization (SID) strategy to discretize depth and recast depth network learning as an ordinal regression problem. By training the network using an ordinary regression loss, our method achieves much higher accuracy and \dd{faster convergence in synch}. Furthermore, we adopt a multi-scale network structure which avoids unnecessary spatial pooling and captures multi-scale information in parallel. The method described in this paper achieves state-of-the-art results on four challenging benchmarks, i.e., KITTI [17], ScanNet [9], Make3D [50], and NYU Depth v2 [42], and win the 1st prize in Robust Vision Challenge 2018. Code has been made available at: https://github.com/hufu6371/DORN.
Tasks	Depth Estimation, Monocular Depth Estimation
Published	2018-06-06
URL	http://arxiv.org/abs/1806.02446v1
PDF	http://arxiv.org/pdf/1806.02446v1.pdf
PWC	https://paperswithcode.com/paper/deep-ordinal-regression-network-for-monocular
Repo	https://github.com/liviniuk/DORN_depth_estimation_Pytorch
Framework	pytorch