January 31, 2020

3503 words 17 mins read

Paper Group AWR 451

A Failure of Aspect Sentiment Classifiers and an Adaptive Re-weighting Solution. Memory-efficient and fast implementation of local adaptive binarization methods. Automatic Pulmonary Lobe Segmentation Using Deep Learning. Conf-Net: Toward High-Confidence Dense 3D Point-Cloud with Error-Map Prediction. Efficient Project Gradient Descent for Ensemble …

A Failure of Aspect Sentiment Classifiers and an Adaptive Re-weighting Solution


Title	A Failure of Aspect Sentiment Classifiers and an Adaptive Re-weighting Solution
Authors	Hu Xu, Bing Liu, Lei Shu, Philip S. Yu
Abstract	Aspect-based sentiment classification (ASC) is an important task in fine-grained sentiment analysis.~Deep supervised ASC approaches typically model this task as a pair-wise classification task that takes an aspect and a sentence containing the aspect and outputs the polarity of the aspect in that sentence. However, we discovered that many existing approaches fail to learn an effective ASC classifier but more like a sentence-level sentiment classifier because they have difficulty to handle sentences with different polarities for different aspects.~This paper first demonstrates this problem using several state-of-the-art ASC models. It then proposes a novel and general adaptive re-weighting (ARW) scheme to adjust the training to dramatically improve ASC for such complex sentences. Experimental results show that the proposed framework is effective \footnote{The dataset and code are available at \url{https://github.com/howardhsu/ASC_failure}.}.
Tasks	Sentiment Analysis
Published	2019-11-04
URL	https://arxiv.org/abs/1911.01460v1
PDF	https://arxiv.org/pdf/1911.01460v1.pdf
PWC	https://paperswithcode.com/paper/a-failure-of-aspect-sentiment-classifiers-and
Repo	https://github.com/howardhsu/ASC_failure
Framework	pytorch

Memory-efficient and fast implementation of local adaptive binarization methods


Title	Memory-efficient and fast implementation of local adaptive binarization methods
Authors	Chungkwong Chan
Abstract	Binarization is widely used as an image preprocessing step to separate object especially text from background before recognition. For noisy images with uneven illumination such as degraded documents, threshold values need to be computed pixel by pixel to obtain a good segmentation. Since local threshold values typically depend on moment-based statistics such as mean and variance of gray levels inside rectangular windows, integral images which are memory consuming are commonly used to accelerate the calculation. Observed that moment-based statistics as well as quantiles in a sliding window can be computed recursively, integral images can be avoided without neglecting speed, more binarization methods can be accelerated too. In particular, given a $H\times W$ input image, Sauvola’s method and alike can run in $\Theta (HW)$ time independent of window size, while only around $6\min{H,W}$ bytes of auxiliary space is needed, which is significantly lower than the $16HW$ bytes occupied by the two integral images. Since the proposed technique enable various well-known local adaptive binarization methods to be applied in real-time use cases on devices with limited resources, it has the potential of wide application.
Tasks
Published	2019-05-30
URL	https://arxiv.org/abs/1905.13038v3
PDF	https://arxiv.org/pdf/1905.13038v3.pdf
PWC	https://paperswithcode.com/paper/memory-efficient-and-fast-implementation-of
Repo	https://github.com/chungkwong/binarizer
Framework	none

Automatic Pulmonary Lobe Segmentation Using Deep Learning


Title	Automatic Pulmonary Lobe Segmentation Using Deep Learning
Authors	Hao Tang, Chupeng Zhang, Xiaohui Xie
Abstract	Pulmonary lobe segmentation is an important task for pulmonary disease related Computer Aided Diagnosis systems (CADs). Classical methods for lobe segmentation rely on successful detection of fissures and other anatomical information such as the location of blood vessels and airways. With the success of deep learning in recent years, Deep Convolutional Neural Network (DCNN) has been widely applied to analyze medical images like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), which, however, requires a large number of ground truth annotations. In this work, we release our manually labeled 50 CT scans which are randomly chosen from the LUNA16 dataset and explore the use of deep learning on this task. We propose pre-processing CT image by cropping region that is covered by the convex hull of the lungs in order to mitigate the influence of noise from outside the lungs. Moreover, we design a hybrid loss function with dice loss to tackle extreme class imbalance issue and focal loss to force model to focus on voxels that are hard to be discriminated. To validate the robustness and performance of our proposed framework trained with a small number of training examples, we further tested our model on CT scans from an independent dataset. Experimental results show the robustness of the proposed approach, which consistently improves performance across different datasets by a maximum of $5.87%$ as compared to a baseline model.
Tasks	Computed Tomography (CT)
Published	2019-03-23
URL	http://arxiv.org/abs/1903.09879v3
PDF	http://arxiv.org/pdf/1903.09879v3.pdf
PWC	https://paperswithcode.com/paper/automatic-pulmonary-lobe-segmentation-using
Repo	https://github.com/deep-voxel/automatic_pulmonary_lobe_segmentation_using_deep_learning
Framework	none

Conf-Net: Toward High-Confidence Dense 3D Point-Cloud with Error-Map Prediction


Title	Conf-Net: Toward High-Confidence Dense 3D Point-Cloud with Error-Map Prediction
Authors	Hamid Hekmatian, Jingfu Jin, Samir Al-Stouhi
Abstract	This work proposes a method for depth completion of sparse LiDAR data using a convolutional neural network which can be used to generate semi-dense depth maps and “almost” full 3D point-clouds with significantly lower root mean squared error (RMSE) over state-of-the-art methods. We add an “Error Prediction” unit to our network and present a novel and simple end-to-end method that learns to predict an error-map of depth regression task. An “almost” dense high-confidence/low-variance point-cloud is more valuable for safety-critical applications specifically real-world autonomous driving than a full point-cloud with high error rate and high error variance. Using our predicted error-map, we demonstrate that by up-filling a LiDAR point cloud from 18,000 points to 285,000 points, versus 300,000 points for full depth, we can reduce the RMSE error from 1004 to 399. This error is approximately 60% less than the state-of-the-art and 50% less than the state-of-the-art with RGB guidance (we did not use RGB guidance in our algorithm). In addition to analyzing our results on Kitti depth completion dataset, we also demonstrate the ability of our proposed method to extend to new tasks by deploying our “Error Prediction” unit to improve upon the state-of-the-art for monocular depth estimation. Codes and demo videos are available at http://github.com/hekmak/Conf-net.
Tasks	Autonomous Driving, Depth Completion, Depth Estimation, Monocular Depth Estimation
Published	2019-07-23
URL	https://arxiv.org/abs/1907.10148v3
PDF	https://arxiv.org/pdf/1907.10148v3.pdf
PWC	https://paperswithcode.com/paper/conf-net-predicting-depth-completion-error
Repo	https://github.com/hekmak/Conf-net
Framework	tf

Efficient Project Gradient Descent for Ensemble Adversarial Attack


Title	Efficient Project Gradient Descent for Ensemble Adversarial Attack
Authors	Fanyou Wu, Rado Gazo, Eva Haviarova, Bedrich Benes
Abstract	Recent advances show that deep neural networks are not robust to deliberately crafted adversarial examples which many are generated by adding human imperceptible perturbation to clear input. Consider $l_2$ norms attacks, Project Gradient Descent (PGD) and the Carlini and Wagner (C&W) attacks are the two main methods, where PGD control max perturbation for adversarial examples while C&W approach treats perturbation as a regularization term optimized it with loss function together. If we carefully set parameters for any individual input, both methods become similar. In general, PGD attacks perform faster but obtains larger perturbation to find adversarial examples than the C&W when fixing the parameters for all inputs. In this report, we propose an efficient modified PGD method for attacking ensemble models by automatically changing ensemble weights and step size per iteration per input. This method generates smaller perturbation adversarial examples than PGD method while remains efficient as compared to C&W method. Our method won the first place in IJCAI19 Targeted Adversarial Attack competition.
Tasks	Adversarial Attack
Published	2019-06-07
URL	https://arxiv.org/abs/1906.03333v1
PDF	https://arxiv.org/pdf/1906.03333v1.pdf
PWC	https://paperswithcode.com/paper/efficient-project-gradient-descent-for
Repo	https://github.com/wufanyou/EPGD
Framework	tf

Effortless Deep Training for Traffic Sign Detection Using Templates and Arbitrary Natural Images


Title	Effortless Deep Training for Traffic Sign Detection Using Templates and Arbitrary Natural Images
Authors	Lucas Tabelini Torres, Thiago M. Paixão, Rodrigo F. Berriel, Alberto F. De Souza, Claudine Badue, Nicu Sebe, Thiago Oliveira-Santos
Abstract	Deep learning has been successfully applied to several problems related to autonomous driving. Often, these solutions rely on large networks that require databases of real image samples of the problem (i.e., real world) for proper training. The acquisition of such real-world data sets is not always possible in the autonomous driving context, and sometimes their annotation is not feasible (e.g., takes too long or is too expensive). Moreover, in many tasks, there is an intrinsic data imbalance that most learning-based methods struggle to cope with. It turns out that traffic sign detection is a problem in which these three issues are seen altogether. In this work, we propose a novel database generation method that requires only (i) arbitrary natural images, i.e., requires no real image from the domain of interest, and (ii) templates of the traffic signs, i.e., templates synthetically created to illustrate the appearance of the category of a traffic sign. The effortlessly generated training database is shown to be effective for the training of a deep detector (such as Faster R-CNN) on German traffic signs, achieving 95.66% of mAP on average. In addition, the proposed method is able to detect traffic signs with an average precision, recall and F1-score of about 94%, 91% and 93%, respectively. The experiments surprisingly show that detectors can be trained with simple data generation methods and without problem domain data for the background, which is in the opposite direction of the common sense for deep learning.
Tasks	Autonomous Driving, Data Augmentation
Published	2019-07-23
URL	https://arxiv.org/abs/1907.09679v1
PDF	https://arxiv.org/pdf/1907.09679v1.pdf
PWC	https://paperswithcode.com/paper/effortless-deep-training-for-traffic-sign
Repo	https://github.com/LCAD-UFES/publications-tabelini-ijcnn-2019
Framework	tf

Data Shapley: Equitable Valuation of Data for Machine Learning


Title	Data Shapley: Equitable Valuation of Data for Machine Learning
Authors	Amirata Ghorbani, James Zou
Abstract	As data becomes the fuel driving technological and economic growth, a fundamental challenge is how to quantify the value of data in algorithmic predictions and decisions. For example, in healthcare and consumer markets, it has been suggested that individuals should be compensated for the data that they generate, but it is not clear what is an equitable valuation for individual data. In this work, we develop a principled framework to address data valuation in the context of supervised machine learning. Given a learning algorithm trained on $n$ data points to produce a predictor, we propose data Shapley as a metric to quantify the value of each training datum to the predictor performance. Data Shapley value uniquely satisfies several natural properties of equitable data valuation. We develop Monte Carlo and gradient-based methods to efficiently estimate data Shapley values in practical settings where complex learning algorithms, including neural networks, are trained on large datasets. In addition to being equitable, extensive experiments across biomedical, image and synthetic data demonstrate that data Shapley has several other benefits: 1) it is more powerful than the popular leave-one-out or leverage score in providing insight on what data is more valuable for a given learning task; 2) low Shapley value data effectively capture outliers and corruptions; 3) high Shapley value data inform what type of new data to acquire to improve the predictor.
Tasks
Published	2019-04-05
URL	https://arxiv.org/abs/1904.02868v2
PDF	https://arxiv.org/pdf/1904.02868v2.pdf
PWC	https://paperswithcode.com/paper/data-shapley-equitable-valuation-of-data-for
Repo	https://github.com/amiratag/DataShapley
Framework	tf

Adversarial Fisher Vectors for Unsupervised Representation Learning


Title	Adversarial Fisher Vectors for Unsupervised Representation Learning
Authors	Shuangfei Zhai, Walter Talbott, Carlos Guestrin, Joshua M. Susskind
Abstract	We examine Generative Adversarial Networks (GANs) through the lens of deep Energy Based Models (EBMs), with the goal of exploiting the density model that follows from this formulation. In contrast to a traditional view where the discriminator learns a constant function when reaching convergence, here we show that it can provide useful information for downstream tasks, e.g., feature extraction for classification. To be concrete, in the EBM formulation, the discriminator learns an unnormalized density function (i.e., the negative energy term) that characterizes the data manifold. We propose to evaluate both the generator and the discriminator by deriving corresponding Fisher Score and Fisher Information from the EBM. We show that by assuming that the generated examples form an estimate of the learned density, both the Fisher Information and the normalized Fisher Vectors are easy to compute. We also show that we are able to derive a distance metric between examples and between sets of examples. We conduct experiments showing that the GAN-induced Fisher Vectors demonstrate competitive performance as unsupervised feature extractors for classification and perceptual similarity tasks. Code is available at \url{https://github.com/apple/ml-afv}.
Tasks	Representation Learning, Unsupervised Representation Learning
Published	2019-10-29
URL	https://arxiv.org/abs/1910.13101v1
PDF	https://arxiv.org/pdf/1910.13101v1.pdf
PWC	https://paperswithcode.com/paper/adversarial-fisher-vectors-for-unsupervised
Repo	https://github.com/apple/ml-afv
Framework	pytorch

On the Global Optima of Kernelized Adversarial Representation Learning


Title	On the Global Optima of Kernelized Adversarial Representation Learning
Authors	Bashir Sadeghi, Runyi Yu, Vishnu Naresh Boddeti
Abstract	Adversarial representation learning is a promising paradigm for obtaining data representations that are invariant to certain sensitive attributes while retaining the information necessary for predicting target attributes. Existing approaches solve this problem through iterative adversarial minimax optimization and lack theoretical guarantees. In this paper, we first study the “linear” form of this problem i.e., the setting where all the players are linear functions. We show that the resulting optimization problem is both non-convex and non-differentiable. We obtain an exact closed-form expression for its global optima through spectral learning and provide performance guarantees in terms of analytical bounds on the achievable utility and invariance. We then extend this solution and analysis to non-linear functions through kernel representation. Numerical experiments on UCI, Extended Yale B and CIFAR-100 datasets indicate that, (a) practically, our solution is ideal for “imparting” provable invariance to any biased pre-trained data representation, and (b) empirically, the trade-off between utility and invariance provided by our solution is comparable to iterative minimax optimization of existing deep neural network based approaches. Code is available at https://github.com/human-analysis/Kernel-ARL
Tasks	Representation Learning
Published	2019-10-16
URL	https://arxiv.org/abs/1910.07423v2
PDF	https://arxiv.org/pdf/1910.07423v2.pdf
PWC	https://paperswithcode.com/paper/on-the-global-optima-of-kernelized
Repo	https://github.com/human-analysis/Kernel-ARL
Framework	pytorch

Multi-Agent Connected Autonomous Driving using Deep Reinforcement Learning


Title	Multi-Agent Connected Autonomous Driving using Deep Reinforcement Learning
Authors	Praveen Palanisamy
Abstract	The capability to learn and adapt to changes in the driving environment is crucial for developing autonomous driving systems that are scalable beyond geo-fenced operational design domains. Deep Reinforcement Learning (RL) provides a promising and scalable framework for developing adaptive learning based solutions. Deep RL methods usually model the problem as a (Partially Observable) Markov Decision Process in which an agent acts in a stationary environment to learn an optimal behavior policy. However, driving involves complex interaction between multiple, intelligent (artificial or human) agents in a highly non-stationary environment. In this paper, we propose the use of Partially Observable Markov Games(POSG) for formulating the connected autonomous driving problems with realistic assumptions. We provide a taxonomy of multi-agent learning environments based on the nature of tasks, nature of agents and the nature of the environment to help in categorizing various autonomous driving problems that can be addressed under the proposed formulation. As our main contributions, we provide MACAD-Gym, a Multi-Agent Connected, Autonomous Driving agent learning platform for furthering research in this direction. Our MACAD-Gym platform provides an extensible set of Connected Autonomous Driving (CAD) simulation environments that enable the research and development of Deep RL- based integrated sensing, perception, planning and control algorithms for CAD systems with unlimited operational design domain under realistic, multi-agent settings. We also share the MACAD-Agents that were trained successfully using the MACAD-Gym platform to learn control policies for multiple vehicle agents in a partially observable, stop-sign controlled, 3-way urban intersection environment with raw (camera) sensor observations.
Tasks	Autonomous Driving
Published	2019-11-11
URL	https://arxiv.org/abs/1911.04175v1
PDF	https://arxiv.org/pdf/1911.04175v1.pdf
PWC	https://paperswithcode.com/paper/multi-agent-connected-autonomous-driving
Repo	https://github.com/praveen-palanisamy/macad-gym
Framework	none

A General Framework for Uncertainty Estimation in Deep Learning


Title	A General Framework for Uncertainty Estimation in Deep Learning
Authors	Antonio Loquercio, Mattia Segù, Davide Scaramuzza
Abstract	Neural networks predictions are unreliable when the input sample is out of the training distribution or corrupted by noise. Being able to detect such failures automatically is fundamental to integrate deep learning algorithms into robotics. Current approaches for uncertainty estimation of neural networks require changes to the network and optimization process, typically ignore prior knowledge about the data, and tend to make over-simplifying assumptions which underestimate uncertainty. To address these limitations, we propose a novel framework for uncertainty estimation. Based on Bayesian belief networks and Monte-Carlo sampling, our framework not only fully models the different sources of prediction uncertainty, but also incorporates prior data information, e.g. sensor noise. We show theoretically that this gives us the ability to capture uncertainty better than existing methods. In addition, our framework has several desirable properties: (i) it is agnostic to the network architecture and task; (ii) it does not require changes in the optimization process; (iii) it can be applied to already trained architectures. We thoroughly validate the proposed framework through extensive experiments on both computer vision and control tasks, where we outperform previous methods by up to 23% in accuracy.
Tasks	Autonomous Driving, Bayesian Inference
Published	2019-07-16
URL	https://arxiv.org/abs/1907.06890v4
PDF	https://arxiv.org/pdf/1907.06890v4.pdf
PWC	https://paperswithcode.com/paper/a-general-framework-for-uncertainty
Repo	https://github.com/mattiasegu/uncertainty_estimation_deep_learning
Framework	pytorch

Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer


Title	Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer
Authors	Wenzheng Chen, Jun Gao, Huan Ling, Edward J. Smith, Jaakko Lehtinen, Alec Jacobson, Sanja Fidler
Abstract	Many machine learning models operate on images, but ignore the fact that images are 2D projections formed by 3D geometry interacting with light, in a process called rendering. Enabling ML models to understand image formation might be key for generalization. However, due to an essential rasterization step involving discrete assignment operations, rendering pipelines are non-differentiable and thus largely inaccessible to gradient-based ML techniques. In this paper, we present {\emph DIB-R}, a differentiable rendering framework which allows gradients to be analytically computed for all pixels in an image. Key to our approach is to view foreground rasterization as a weighted interpolation of local properties and background rasterization as a distance-based aggregation of global geometry. Our approach allows for accurate optimization over vertex positions, colors, normals, light directions and texture coordinates through a variety of lighting models. We showcase our approach in two ML applications: single-image 3D object prediction, and 3D textured object generation, both trained using exclusively using 2D supervision. Our project website is: https://nv-tlabs.github.io/DIB-R/
Tasks
Published	2019-08-03
URL	https://arxiv.org/abs/1908.01210v2
PDF	https://arxiv.org/pdf/1908.01210v2.pdf
PWC	https://paperswithcode.com/paper/learning-to-predict-3d-objects-with-an
Repo	https://github.com/nv-tlabs/DIB-R
Framework	pytorch

PHiSeg: Capturing Uncertainty in Medical Image Segmentation


Title	PHiSeg: Capturing Uncertainty in Medical Image Segmentation
Authors	Christian F. Baumgartner, Kerem C. Tezcan, Krishna Chaitanya, Andreas M. Hötker, Urs J. Muehlematter, Khoschy Schawkat, Anton S. Becker, Olivio Donati, Ender Konukoglu
Abstract	Segmentation of anatomical structures and pathologies is inherently ambiguous. For instance, structure borders may not be clearly visible or different experts may have different styles of annotating. The majority of current state-of-the-art methods do not account for such ambiguities but rather learn a single mapping from image to segmentation. In this work, we propose a novel method to model the conditional probability distribution of the segmentations given an input image. We derive a hierarchical probabilistic model, in which separate latent variables are responsible for modelling the segmentation at different resolutions. Inference in this model can be efficiently performed using the variational autoencoder framework. We show that our proposed method can be used to generate significantly more realistic and diverse segmentation samples compared to recent related work, both, when trained with annotations from a single or multiple annotators.
Tasks	Medical Image Segmentation, Semantic Segmentation
Published	2019-06-07
URL	https://arxiv.org/abs/1906.04045v2
PDF	https://arxiv.org/pdf/1906.04045v2.pdf
PWC	https://paperswithcode.com/paper/phiseg-capturing-uncertainty-in-medical-image
Repo	https://github.com/baumgach/PHiSeg-code
Framework	tf

Water-Filling: An Efficient Algorithm for Digitized Document Shadow Removal


Title	Water-Filling: An Efficient Algorithm for Digitized Document Shadow Removal
Authors	Seungjun Jung, Muhammad Abul Hasan, Changick Kim
Abstract	In this paper, we propose a novel algorithm to rectify illumination of the digitized documents by eliminating shading artifacts. Firstly, a topographic surface of an input digitized document is created using luminance value of each pixel. Then the shading artifact on the document is estimated by simulating an immersion process. The simulation of the immersion process is modeled using a novel diffusion equation with an iterative update rule. After estimating the shading artifacts, the digitized document is reconstructed using the Lambertian surface model. In order to evaluate the performance of the proposed algorithm, we conduct rigorous experiments on a set of digitized documents which is generated using smartphones under challenging lighting conditions. According to the experimental results, it is found that the proposed method produces promising illumination correction results and outperforms the results of the state-of-the-art methods.
Tasks
Published	2019-04-22
URL	https://arxiv.org/abs/1904.09763v2
PDF	https://arxiv.org/pdf/1904.09763v2.pdf
PWC	https://paperswithcode.com/paper/water-filling-an-efficient-algorithm-for
Repo	https://github.com/seungjun45/Water-Filling
Framework	none

Exact inference under the perfect phylogeny model


Title	Exact inference under the perfect phylogeny model
Authors	Surjyendu Ray, Bei Jia, Sam Safavi, Tim van Opijnen, Ralph Isberg, Jason Rosch, José Bento
Abstract	Motivation: Many inference tools use the Perfect Phylogeny Model (PPM) to learn trees from noisy variant allele frequency (VAF) data. Learning in this setting is hard, and existing tools use approximate or heuristic algorithms. An algorithmic improvement is important to help disentangle the limitations of the PPM’s assumptions from the limitations in our capacity to learn under it. Results: We make such improvement in the scenario, where the mutations that are relevant for evolution can be clustered into a small number of groups, and the trees to be reconstructed have a small number of nodes. We use a careful combination of algorithms, software, and hardware, to develop EXACT: a tool that can explore the space of all possible phylogenetic trees, and performs exact inference under the PPM with noisy data. EXACT allows users to obtain not just the most-likely tree for some input data, but exact statistics about the distribution of trees that might explain the data. We show that EXACT outperforms several existing tools for this same task. Availability: https://github.com/surjray-repos/EXACT
Tasks
Published	2019-08-22
URL	https://arxiv.org/abs/1908.08623v1
PDF	https://arxiv.org/pdf/1908.08623v1.pdf
PWC	https://paperswithcode.com/paper/exact-inference-under-the-perfect-phylogeny
Repo	https://github.com/surjray-repos/EXACT
Framework	none