October 18, 2019

2988 words 15 mins read

Paper Group ANR 531

Deep-learning models improve on community-level diagnosis for common congenital heart disease lesions. Determination of Digital Straight Segments Using the Slope. A Data-Efficient Approach to Precise and Controlled Pushing. A New De-blurring Technique for License Plate Images with Robust Length Estimation. The Mean-Field Approximation: Information …

Deep-learning models improve on community-level diagnosis for common congenital heart disease lesions


Title	Deep-learning models improve on community-level diagnosis for common congenital heart disease lesions
Authors	Rima Arnaout, Lara Curran, Erin Chinn, Yili Zhao, Anita Moon-Grady
Abstract	Prenatal diagnosis of tetralogy of Fallot (TOF) and hypoplastic left heart syndrome (HLHS), two serious congenital heart defects, improves outcomes and can in some cases facilitate in utero interventions. In practice, however, the fetal diagnosis rate for these lesions is only 30-50 percent in community settings. Improving fetal diagnosis of congenital heart disease is therefore critical. Deep learning is a cutting-edge machine learning technique for finding patterns in images but has not yet been applied to prenatal diagnosis of congenital heart disease. Using 685 retrospectively collected echocardiograms from fetuses 18-24 weeks of gestational age from 2000-2018, we trained convolutional and fully-convolutional deep learning models in a supervised manner to (i) identify the five canonical screening views of the fetal heart and (ii) segment cardiac structures to calculate fetal cardiac biometrics. We then trained models to distinguish by view between normal hearts, TOF, and HLHS. In a holdout test set of images, F-score for identification of the five most important fetal cardiac views was 0.95. Binary classification of unannotated cardiac views of normal heart vs. TOF reached an overall sensitivity of 75% and a specificity of 76%, while normal vs. HLHS reached a sensitivity of 100% and specificity of 90%, both well above average diagnostic rates for these lesions. Furthermore, segmentation-based measurements for cardiothoracic ratio (CTR), cardiac axis (CA), and ventricular fractional area change (FAC) were compatible with clinically measured metrics for normal, TOF, and HLHS hearts. Thus, using guideline-recommended imaging, deep learning models can significantly improve detection of fetal congenital heart disease compared to the common standard of care.
Tasks
Published	2018-09-19
URL	http://arxiv.org/abs/1809.06993v1
PDF	http://arxiv.org/pdf/1809.06993v1.pdf
PWC	https://paperswithcode.com/paper/deep-learning-models-improve-on-community
Repo
Framework

Determination of Digital Straight Segments Using the Slope


Title	Determination of Digital Straight Segments Using the Slope
Authors	Alejandro Cartas, María Elena Algorri
Abstract	We present a new method for the recognition of digital straight lines based on the slope. This method combines the Freeman’s chain coding scheme and new discovered properties of the digital slope introduced in this paper. We also present the efficiency of our method from a testbed.
Tasks
Published	2018-01-20
URL	http://arxiv.org/abs/1801.06694v1
PDF	http://arxiv.org/pdf/1801.06694v1.pdf
PWC	https://paperswithcode.com/paper/determination-of-digital-straight-segments
Repo
Framework

A Data-Efficient Approach to Precise and Controlled Pushing


Title	A Data-Efficient Approach to Precise and Controlled Pushing
Authors	Maria Bauza, Francois R. Hogan, Alberto Rodriguez
Abstract	Decades of research in control theory have shown that simple controllers, when provided with timely feedback, can control complex systems. Pushing is an example of a complex mechanical system that is difficult to model accurately due to unknown system parameters such as coefficients of friction and pressure distributions. In this paper, we explore the data-complexity required for controlling, rather than modeling, such a system. Results show that a model-based control approach, where the dynamical model is learned from data, is capable of performing complex pushing trajectories with a minimal amount of training data (10 data points). The dynamics of pushing interactions are modeled using a Gaussian process (GP) and are leveraged within a model predictive control approach that linearizes the GP and imposes actuator and task constraints for a planar manipulation task.
Tasks
Published	2018-07-26
URL	http://arxiv.org/abs/1807.09904v2
PDF	http://arxiv.org/pdf/1807.09904v2.pdf
PWC	https://paperswithcode.com/paper/a-data-efficient-approach-to-precise-and
Repo
Framework

A New De-blurring Technique for License Plate Images with Robust Length Estimation


Title	A New De-blurring Technique for License Plate Images with Robust Length Estimation
Authors	P. S. Prashanth Rao, Rajesh Kumar Muthu
Abstract	Recognizing a license plate clearly while seeing a surveillance camera snapshot is often important in cases where the troublemaker vehicle(s) have to be identified. In many real world situations, these images are blurred due to fast motion of the vehicle and cannot be recognized by the human eye. For this kind of blurring, the kernel involved can be said to be a linear uniform convolution described by its angle and length. We propose a new de-blurring technique in this paper to parametrically estimate the kernel as accurately as possible with emphasis on the length estimation process. We use a technique which employs Hough transform in estimating the kernel angle. To accurately estimate the kernel length, a novel approach using the cepstral transform is introduced. We compare the de-blurred results obtained using our scheme with those of other recently introduced blind de-blurring techniques. The comparisons corroborate that our scheme can remove a large blur from the image captured by the camera to recover vital semantic information about the license plate.
Tasks
Published	2018-02-17
URL	http://arxiv.org/abs/1802.06214v1
PDF	http://arxiv.org/pdf/1802.06214v1.pdf
PWC	https://paperswithcode.com/paper/a-new-de-blurring-technique-for-license-plate
Repo
Framework

The Mean-Field Approximation: Information Inequalities, Algorithms, and Complexity


Title	The Mean-Field Approximation: Information Inequalities, Algorithms, and Complexity
Authors	Vishesh Jain, Frederic Koehler, Elchanan Mossel
Abstract	The mean field approximation to the Ising model is a canonical variational tool that is used for analysis and inference in Ising models. We provide a simple and optimal bound for the KL error of the mean field approximation for Ising models on general graphs, and extend it to higher order Markov random fields. Our bound improves on previous bounds obtained in work in the graph limit literature by Borgs, Chayes, Lov'asz, S'os, and Vesztergombi and another recent work by Basak and Mukherjee. Our bound is tight up to lower order terms. Building on the methods used to prove the bound, along with techniques from combinatorics and optimization, we study the algorithmic problem of estimating the (variational) free energy for Ising models and general Markov random fields. For a graph $G$ on $n$ vertices and interaction matrix $J$ with Frobenius norm $\ J _F$, we provide algorithms that approximate the free energy within an additive error of $\epsilon n \J_F$ in time $\exp(poly(1/\epsilon))$. We also show that approximation within $(n \J_F)^{1-\delta}$ is NP-hard for every $\delta > 0$. Finally, we provide more efficient approximation algorithms, which find the optimal mean field approximation, for ferromagnetic Ising models and for Ising models satisfying Dobrushin’s condition.
Tasks
Published	2018-02-16
URL	http://arxiv.org/abs/1802.06126v2
PDF	http://arxiv.org/pdf/1802.06126v2.pdf
PWC	https://paperswithcode.com/paper/the-mean-field-approximation-information
Repo
Framework

A Unified Framework for Mutual Improvement of SLAM and Semantic Segmentation


Title	A Unified Framework for Mutual Improvement of SLAM and Semantic Segmentation
Authors	Kai Wang, Yimin Lin, Luowei Wang, Liming Han, Minjie Hua, Xiang Wang, Shiguo Lian, Bill Huang
Abstract	This paper presents a novel framework for simultaneously implementing localization and segmentation, which are two of the most important vision-based tasks for robotics. While the goals and techniques used for them were considered to be different previously, we show that by making use of the intermediate results of the two modules, their performance can be enhanced at the same time. Our framework is able to handle both the instantaneous motion and long-term changes of instances in localization with the help of the segmentation result, which also benefits from the refined 3D pose information. We conduct experiments on various datasets, and prove that our framework works effectively on improving the precision and robustness of the two tasks and outperforms existing localization and segmentation algorithms.
Tasks	Semantic Segmentation
Published	2018-12-25
URL	http://arxiv.org/abs/1812.10016v2
PDF	http://arxiv.org/pdf/1812.10016v2.pdf
PWC	https://paperswithcode.com/paper/a-unified-framework-for-mutual-improvement-of
Repo
Framework

Reward Constrained Policy Optimization


Title	Reward Constrained Policy Optimization
Authors	Chen Tessler, Daniel J. Mankowitz, Shie Mannor
Abstract	Solving tasks in Reinforcement Learning is no easy feat. As the goal of the agent is to maximize the accumulated reward, it often learns to exploit loopholes and misspecifications in the reward signal resulting in unwanted behavior. While constraints may solve this issue, there is no closed form solution for general constraints. In this work we present a novel multi-timescale approach for constrained policy optimization, called `Reward Constrained Policy Optimization’ (RCPO), which uses an alternative penalty signal to guide the policy towards a constraint satisfying one. We prove the convergence of our approach and provide empirical evidence of its ability to train constraint satisfying policies. \|
Tasks
Published	2018-05-28
URL	http://arxiv.org/abs/1805.11074v3
PDF	http://arxiv.org/pdf/1805.11074v3.pdf
PWC	https://paperswithcode.com/paper/reward-constrained-policy-optimization
Repo
Framework

Causal Learning and Explanation of Deep Neural Networks via Autoencoded Activations


Title	Causal Learning and Explanation of Deep Neural Networks via Autoencoded Activations
Authors	Michael Harradon, Jeff Druce, Brian Ruttenberg
Abstract	Deep neural networks are complex and opaque. As they enter application in a variety of important and safety critical domains, users seek methods to explain their output predictions. We develop an approach to explaining deep neural networks by constructing causal models on salient concepts contained in a CNN. We develop methods to extract salient concepts throughout a target network by using autoencoders trained to extract human-understandable representations of network activations. We then build a bayesian causal model using these extracted concepts as variables in order to explain image classification. Finally, we use this causal model to identify and visualize features with significant causal influence on final classification.
Tasks	Image Classification
Published	2018-02-02
URL	http://arxiv.org/abs/1802.00541v1
PDF	http://arxiv.org/pdf/1802.00541v1.pdf
PWC	https://paperswithcode.com/paper/causal-learning-and-explanation-of-deep
Repo
Framework

LSALSA: Accelerated Source Separation via Learned Sparse Coding


Title	LSALSA: Accelerated Source Separation via Learned Sparse Coding
Authors	Benjamin Cowen, Apoorva Nandini Saridena, Anna Choromanska
Abstract	We propose an efficient algorithm for the generalized sparse coding (SC) inference problem. The proposed framework applies to both the single dictionary setting, where each data point is represented as a sparse combination of the columns of one dictionary matrix, as well as the multiple dictionary setting as given in morphological component analysis (MCA), where the goal is to separate a signal into additive parts such that each part has distinct sparse representation within a corresponding dictionary. Both the SC task and its generalization via MCA have been cast as $\ell_1$-regularized least-squares optimization problems. To accelerate traditional acquisition of sparse codes, we propose a deep learning architecture that constitutes a trainable time-unfolded version of the Split Augmented Lagrangian Shrinkage Algorithm (SALSA), a special case of the Alternating Direction Method of Multipliers (ADMM). We empirically validate both variants of the algorithm, that we refer to as LSALSA (learned-SALSA), on image vision tasks and demonstrate that at inference our networks achieve vast improvements in terms of the running time, the quality of estimated sparse codes, and visual clarity on both classic SC and MCA problems. Finally, we present a theoretical framework for analyzing LSALSA network: we show that the proposed approach exactly implements a truncated ADMM applied to a new, learned cost function with curvature modified by one of the learned parameterized matrices. We extend a very recent Stochastic Alternating Optimization analysis framework to show that a gradient descent step along this learned loss landscape is equivalent to a modified gradient descent step along the original loss landscape. In this framework, the acceleration achieved by LSALSA could potentially be explained by the network’s ability to learn a correction to the gradient direction of steeper descent.
Tasks
Published	2018-02-13
URL	https://arxiv.org/abs/1802.06875v2
PDF	https://arxiv.org/pdf/1802.06875v2.pdf
PWC	https://paperswithcode.com/paper/lsalsa-efficient-sparse-coding-in-single-and
Repo
Framework

Smart Device based Initial Movement Detection of Cyclists using Convolutional Neuronal Networks


Title	Smart Device based Initial Movement Detection of Cyclists using Convolutional Neuronal Networks
Authors	Jan Schneegans, Maarten Bieshaar
Abstract	For future traffic scenarios, we envision interconnected traffic participants, who exchange information about their current state, e.g., position, their predicted intentions, allowing to act in a cooperative manner. Vulnerable road users (VRUs), e.g., pedestrians and cyclists, will be equipped with smart device that can be used to detect their intentions and transmit these detected intention to approaching cars such that their drivers can be warned. In this article, we focus on detecting the initial movement of cyclist using smart devices. Smart devices provide the necessary sensors, namely accelerometer and gyroscope, and therefore pose an excellent instrument to detect movement transitions (e.g., waiting to moving) fast. Convolutional Neural Networks prove to be the state-of-the-art solution for many problems with an ever increasing range of applications. Therefore, we model the initial movement detection as a classification problem. In terms of Organic Computing (OC) it be seen as a step towards self-awareness and self-adaptation. We apply residual network architectures to the task of detecting the initial starting movement of cyclists.
Tasks
Published	2018-08-08
URL	http://arxiv.org/abs/1808.04451v1
PDF	http://arxiv.org/pdf/1808.04451v1.pdf
PWC	https://paperswithcode.com/paper/smart-device-based-initial-movement-detection
Repo
Framework

Persistence Codebooks for Topological Data Analysis


Title	Persistence Codebooks for Topological Data Analysis
Authors	Bartosz Zielinski, Michal Lipinski, Mateusz Juda, Matthias Zeppelzauer, Pawel Dlotko
Abstract	Persistent homology (PH) is a rigorous mathematical theory that provides a robust descriptor of data in the form of persistence diagrams (PDs) which are 2D multisets of points. Their variable size makes them, however, difficult to combine with typical machine learning workflows. In this paper we introduce persistence codebooks, a novel expressive and discriminative fixed-size vectorized representation of PDs. To this end, we adapt bag-of-words (BoW), vectors of locally aggregated descriptors (VLAD) and Fischer vectors (FV) for the quantization of PDs. Persistence codebooks represent PDs in a convenient way for machine learning and statistical analysis and have a number of favorable practical and theoretical properties including 1-Wasserstein stability. We evaluate the presented representations on several heterogeneous datasets and show their (high) discriminative power. Our approach achieves state-of-the-art performance and beyond in much less time than alternative approaches.
Tasks	Quantization, Topological Data Analysis
Published	2018-02-13
URL	https://arxiv.org/abs/1802.04852v4
PDF	https://arxiv.org/pdf/1802.04852v4.pdf
PWC	https://paperswithcode.com/paper/persistence-codebooks-for-topological-data
Repo
Framework

Scalable Recommender Systems through Recursive Evidence Chains


Title	Scalable Recommender Systems through Recursive Evidence Chains
Authors	Elias Tragas, Calvin Luo, Maxime Gazeau, Kevin Luk, David Duvenaud
Abstract	Recommender systems can be formulated as a matrix completion problem, predicting ratings from user and item parameter vectors. Optimizing these parameters by subsampling data becomes difficult as the number of users and items grows. We develop a novel approach to generate all latent variables on demand from the ratings matrix itself and a fixed pool of parameters. We estimate missing ratings using chains of evidence that link them to a small set of prototypical users and items. Our model automatically addresses the cold-start and online learning problems by combining information across both users and items. We investigate the scaling behavior of this model, and demonstrate competitive results with respect to current matrix factorization techniques in terms of accuracy and convergence speed.
Tasks	Matrix Completion, Recommendation Systems
Published	2018-07-05
URL	http://arxiv.org/abs/1807.02150v1
PDF	http://arxiv.org/pdf/1807.02150v1.pdf
PWC	https://paperswithcode.com/paper/scalable-recommender-systems-through
Repo
Framework

A Discriminative Approach to Bayesian Filtering with Applications to Human Neural Decoding


Title	A Discriminative Approach to Bayesian Filtering with Applications to Human Neural Decoding
Authors	Michael C. Burkhart
Abstract	Given a stationary state-space model that relates a sequence of hidden states and corresponding measurements or observations, Bayesian filtering provides a principled statistical framework for inferring the posterior distribution of the current state given all measurements up to the present time. For example, the Apollo lunar module implemented a Kalman filter to infer its location from a sequence of earth-based radar measurements and land safely on the moon. To perform Bayesian filtering, we require a measurement model that describes the conditional distribution of each observation given state. The Kalman filter takes this measurement model to be linear, Gaussian. Here we show how a nonlinear, Gaussian approximation to the distribution of state given observation can be used in conjunction with Bayes’ rule to build a nonlinear, non-Gaussian measurement model. The resulting approach, called the Discriminative Kalman Filter (DKF), retains fast closed-form updates for the posterior. We argue there are many cases where the distribution of state given measurement is better-approximated as Gaussian, especially when the dimensionality of measurements far exceeds that of states and the Bernstein-von Mises theorem applies. Online neural decoding for brain-computer interfaces provides a motivating example, where filtering incorporates increasingly detailed measurements of neural activity to provide users control over external devices. Within the BrainGate2 clinical trial, the DKF successfully enabled three volunteers with quadriplegia to control an on-screen cursor in real-time using mental imagery alone. Participant “T9” used the DKF to type out messages on a tablet PC.
Tasks
Published	2018-07-17
URL	http://arxiv.org/abs/1807.06173v1
PDF	http://arxiv.org/pdf/1807.06173v1.pdf
PWC	https://paperswithcode.com/paper/a-discriminative-approach-to-bayesian
Repo
Framework

RPNet: an End-to-End Network for Relative Camera Pose Estimation


Title	RPNet: an End-to-End Network for Relative Camera Pose Estimation
Authors	Sovann En, Alexis Lechervy, Frédéric Jurie
Abstract	This paper addresses the task of relative camera pose estimation from raw image pixels, by means of deep neural networks. The proposed RPNet network takes pairs of images as input and directly infers the relative poses, without the need of camera intrinsic/extrinsic. While state-of-the-art systems based on SIFT + RANSAC, are able to recover the translation vector only up to scale, RPNet is trained to produce the full translation vector, in an end-to-end way. Experimental results on the Cambridge Landmark dataset show very promising results regarding the recovery of the full translation vector. They also show that RPNet produces more accurate and more stable results than traditional approaches, especially for hard images (repetitive textures, textureless images, etc). To the best of our knowledge, RPNet is the first attempt to recover full translation vectors in relative pose estimation.
Tasks	Pose Estimation
Published	2018-09-22
URL	http://arxiv.org/abs/1809.08402v1
PDF	http://arxiv.org/pdf/1809.08402v1.pdf
PWC	https://paperswithcode.com/paper/rpnet-an-end-to-end-network-for-relative
Repo
Framework

$HS^2$: Active Learning over Hypergraphs


Title	$HS^2$: Active Learning over Hypergraphs
Authors	I Chien, Huozhi Zhou, Pan Li
Abstract	We propose a hypergraph-based active learning scheme which we term $HS^2$, $HS^2$ generalizes the previously reported algorithm $S^2$ originally proposed for graph-based active learning with pointwise queries [Dasarathy et al., COLT 2015]. Our $HS^2$ method can accommodate hypergraph structures and allows one to ask both pointwise queries and pairwise queries. Based on a novel parametric system particularly designed for hypergraphs, we derive theoretical results on the query complexity of $HS^2$ for the above described generalized settings. Both the theoretical and empirical results show that $HS^2$ requires a significantly fewer number of queries than $S^2$ when one uses $S^2$ over a graph obtained from the corresponding hypergraph via clique expansion.
Tasks	Active Learning
Published	2018-11-25
URL	http://arxiv.org/abs/1811.11549v1
PDF	http://arxiv.org/pdf/1811.11549v1.pdf
PWC	https://paperswithcode.com/paper/hs2-active-learning-over-hypergraphs
Repo
Framework