July 27, 2019

3192 words 15 mins read

Paper Group ANR 718

Feature Tracking Cardiac Magnetic Resonance via Deep Learning and Spline Optimization. From Zero-shot Learning to Conventional Supervised Classification: Unseen Visual Data Synthesis. Event Stream-Based Process Discovery using Abstract Representations. High Voltage Insulator Surface Evaluation Using Image Processing. Factoring Exogenous State for M …

Feature Tracking Cardiac Magnetic Resonance via Deep Learning and Spline Optimization


Title	Feature Tracking Cardiac Magnetic Resonance via Deep Learning and Spline Optimization
Authors	Davis M. Vigneault, Weidi Xie, David A. Bluemke, J. Alison Noble
Abstract	Feature tracking Cardiac Magnetic Resonance (CMR) has recently emerged as an area of interest for quantification of regional cardiac function from balanced, steady state free precession (SSFP) cine sequences. However, currently available techniques lack full automation, limiting reproducibility. We propose a fully automated technique whereby a CMR image sequence is first segmented with a deep, fully convolutional neural network (CNN) architecture, and quadratic basis splines are fitted simultaneously across all cardiac frames using least squares optimization. Experiments are performed using data from 42 patients with hypertrophic cardiomyopathy (HCM) and 21 healthy control subjects. In terms of segmentation, we compared state-of-the-art CNN frameworks, U-Net and dilated convolution architectures, with and without temporal context, using cross validation with three folds. Performance relative to expert manual segmentation was similar across all networks: pixel accuracy was ~97%, intersection-over-union (IoU) across all classes was ~87%, and IoU across foreground classes only was ~85%. Endocardial left ventricular circumferential strain calculated from the proposed pipeline was significantly different in control and disease subjects (-25.3% vs -29.1%, p = 0.006), in agreement with the current clinical literature.
Tasks
Published	2017-04-12
URL	http://arxiv.org/abs/1704.03660v1
PDF	http://arxiv.org/pdf/1704.03660v1.pdf
PWC	https://paperswithcode.com/paper/feature-tracking-cardiac-magnetic-resonance
Repo
Framework

From Zero-shot Learning to Conventional Supervised Classification: Unseen Visual Data Synthesis


Title	From Zero-shot Learning to Conventional Supervised Classification: Unseen Visual Data Synthesis
Authors	Yang Long, Li Liu, Ling Shao, Fumin Shen, Guiguang Ding, Jungong Han
Abstract	Robust object recognition systems usually rely on powerful feature extraction mechanisms from a large number of real images. However, in many realistic applications, collecting sufficient images for ever-growing new classes is unattainable. In this paper, we propose a new Zero-shot learning (ZSL) framework that can synthesise visual features for unseen classes without acquiring real images. Using the proposed Unseen Visual Data Synthesis (UVDS) algorithm, semantic attributes are effectively utilised as an intermediate clue to synthesise unseen visual features at the training stage. Hereafter, ZSL recognition is converted into the conventional supervised problem, i.e. the synthesised visual features can be straightforwardly fed to typical classifiers such as SVM. On four benchmark datasets, we demonstrate the benefit of using synthesised unseen data. Extensive experimental results suggest that our proposed approach significantly improve the state-of-the-art results.
Tasks	Object Recognition, Zero-Shot Learning
Published	2017-05-04
URL	http://arxiv.org/abs/1705.01782v1
PDF	http://arxiv.org/pdf/1705.01782v1.pdf
PWC	https://paperswithcode.com/paper/from-zero-shot-learning-to-conventional
Repo
Framework

Event Stream-Based Process Discovery using Abstract Representations


Title	Event Stream-Based Process Discovery using Abstract Representations
Authors	Sebastiaan J. van Zelst, Boudewijn F. van Dongen, Wil M. P. van der Aalst
Abstract	The aim of process discovery, originating from the area of process mining, is to discover a process model based on business process execution data. A majority of process discovery techniques relies on an event log as an input. An event log is a static source of historical data capturing the execution of a business process. In this paper we focus on process discovery relying on online streams of business process execution events. Learning process models from event streams poses both challenges and opportunities, i.e. we need to handle unlimited amounts of data using finite memory and, preferably, constant time. We propose a generic architecture that allows for adopting several classes of existing process discovery techniques in context of event streams. Moreover, we provide several instantiations of the architecture, accompanied by implementations in the process mining tool-kit ProM (http://promtools.org). Using these instantiations, we evaluate several dimensions of stream-based process discovery. The evaluation shows that the proposed architecture allows us to lift process discovery to the streaming domain.
Tasks
Published	2017-04-25
URL	http://arxiv.org/abs/1704.08101v1
PDF	http://arxiv.org/pdf/1704.08101v1.pdf
PWC	https://paperswithcode.com/paper/event-stream-based-process-discovery-using
Repo
Framework

High Voltage Insulator Surface Evaluation Using Image Processing


Title	High Voltage Insulator Surface Evaluation Using Image Processing
Authors	Damira Pernebayeva, Mehdi Bagheri, Alex Pappachen James
Abstract	High voltage insulators are widely deployed in power systems to isolate the live- and dead-part of overhead lines as well as to support the power line conductors mechanically. Permanent, secure and safe operation of power transmission lines require that the high voltage insulators are inspected and monitor, regularly. Severe environment conditions will influence insulator surface and change creepage distance. Consequently, power utilities and transmission companies face significant problem in operation due to insulator damage or contamination. In this study, a new technique is developed for real-time inspection of insulator and estimating the snow, ice and water over the insulator surface which can be a potential risk of operation breakdown. To examine the proposed system, practical experiment is conducted using ceramic insulator for capturing the images with snow, ice and wet surface conditions. Gabor and Standard deviation filters are utilized for image feature extraction. The best achieved recognition accuracy rate was 87% using statistical approach the Standard deviation.
Tasks
Published	2017-08-19
URL	http://arxiv.org/abs/1708.05828v1
PDF	http://arxiv.org/pdf/1708.05828v1.pdf
PWC	https://paperswithcode.com/paper/high-voltage-insulator-surface-evaluation
Repo
Framework

Factoring Exogenous State for Model-Free Monte Carlo


Title	Factoring Exogenous State for Model-Free Monte Carlo
Authors	Sean McGregor, Rachel Houtman, Claire Montgomery, Ronald Metoyer, Thomas G. Dietterich
Abstract	Policy analysts wish to visualize a range of policies for large simulator-defined Markov Decision Processes (MDPs). One visualization approach is to invoke the simulator to generate on-policy trajectories and then visualize those trajectories. When the simulator is expensive, this is not practical, and some method is required for generating trajectories for new policies without invoking the simulator. The method of Model-Free Monte Carlo (MFMC) can do this by stitching together state transitions for a new policy based on previously-sampled trajectories from other policies. This “off-policy Monte Carlo simulation” method works well when the state space has low dimension but fails as the dimension grows. This paper describes a method for factoring out some of the state and action variables so that MFMC can work in high-dimensional MDPs. The new method, MFMCi, is evaluated on a very challenging wildfire management MDP.
Tasks
Published	2017-03-28
URL	http://arxiv.org/abs/1703.09390v2
PDF	http://arxiv.org/pdf/1703.09390v2.pdf
PWC	https://paperswithcode.com/paper/factoring-exogenous-state-for-model-free
Repo
Framework

Belief Tree Search for Active Object Recognition


Title	Belief Tree Search for Active Object Recognition
Authors	Mohsen Malmir, Garrison W. Cottrell
Abstract	Active Object Recognition (AOR) has been approached as an unsupervised learning problem, in which optimal trajectories for object inspection are not known and are to be discovered by reducing label uncertainty measures or training with reinforcement learning. Such approaches have no guarantees of the quality of their solution. In this paper, we treat AOR as a Partially Observable Markov Decision Process (POMDP) and find near-optimal policies on training data using Belief Tree Search (BTS) on the corresponding belief Markov Decision Process (MDP). AOR then reduces to the problem of knowledge transfer from near-optimal policies on training set to the test set. We train a Long Short Term Memory (LSTM) network to predict the best next action on the training set rollouts. We sho that the proposed AOR method generalizes well to novel views of familiar objects and also to novel objects. We compare this supervised scheme against guided policy search, and find that the LSTM network reaches higher recognition accuracy compared to the guided policy method. We further look into optimizing the observation function to increase the total collected reward of optimal policy. In AOR, the observation function is known only approximately. We propose a gradient-based method update to this approximate observation function to increase the total reward of any policy. We show that by optimizing the observation function and retraining the supervised LSTM network, the AOR performance on the test set improves significantly.
Tasks	Object Recognition, Transfer Learning
Published	2017-08-13
URL	http://arxiv.org/abs/1708.03901v1
PDF	http://arxiv.org/pdf/1708.03901v1.pdf
PWC	https://paperswithcode.com/paper/belief-tree-search-for-active-object
Repo
Framework

Exploiting generalization in the subspaces for faster model-based learning


Title	Exploiting generalization in the subspaces for faster model-based learning
Authors	Maryam Hashemzadeh, Reshad Hosseini, Majid Nili Ahmadabadi
Abstract	Due to the lack of enough generalization in the state-space, common methods in Reinforcement Learning (RL) suffer from slow learning speed especially in the early learning trials. This paper introduces a model-based method in discrete state-spaces for increasing learning speed in terms of required experience (but not required computational time) by exploiting generalization in the experiences of the subspaces. A subspace is formed by choosing a subset of features in the original state representation (full-space). Generalization and faster learning in a subspace are due to many-to-one mapping of experiences from the full-space to each state in the subspace. Nevertheless, due to inherent perceptual aliasing in the subspaces, the policy suggested by each subspace does not generally converge to the optimal policy. Our approach, called Model Based Learning with Subspaces (MoBLeS), calculates confidence intervals of the estimated Q-values in the full-space and in the subspaces. These confidence intervals are used in the decision making, such that the agent benefits the most from the possible generalization while avoiding from detriment of the perceptual aliasing in the subspaces. Convergence of MoBLeS to the optimal policy is theoretically investigated. Additionally, we show through several experiments that MoBLeS improves the learning speed in the early trials.
Tasks	Decision Making
Published	2017-10-22
URL	http://arxiv.org/abs/1710.08012v2
PDF	http://arxiv.org/pdf/1710.08012v2.pdf
PWC	https://paperswithcode.com/paper/exploiting-generalization-in-the-subspaces
Repo
Framework

Speech recognition for medical conversations


Title	Speech recognition for medical conversations
Authors	Chung-Cheng Chiu, Anshuman Tripathi, Katherine Chou, Chris Co, Navdeep Jaitly, Diana Jaunzeikare, Anjuli Kannan, Patrick Nguyen, Hasim Sak, Ananth Sankar, Justin Tansuwan, Nathan Wan, Yonghui Wu, Xuedong Zhang
Abstract	In this work we explored building automatic speech recognition models for transcribing doctor patient conversation. We collected a large scale dataset of clinical conversations ($14,000$ hr), designed the task to represent the real word scenario, and explored several alignment approaches to iteratively improve data quality. We explored both CTC and LAS systems for building speech recognition models. The LAS was more resilient to noisy data and CTC required more data clean up. A detailed analysis is provided for understanding the performance for clinical tasks. Our analysis showed the speech recognition models performed well on important medical utterances, while errors occurred in causal conversations. Overall we believe the resulting models can provide reasonable quality in practice.
Tasks	Speech Recognition
Published	2017-11-20
URL	http://arxiv.org/abs/1711.07274v2
PDF	http://arxiv.org/pdf/1711.07274v2.pdf
PWC	https://paperswithcode.com/paper/speech-recognition-for-medical-conversations
Repo
Framework


Title	Churn Prediction in Mobile Social Games: Towards a Complete Assessment Using Survival Ensembles
Authors	África Periáñez, Alain Saas, Anna Guitart, Colin Magne
Abstract	Reducing user attrition, i.e. churn, is a broad challenge faced by several industries. In mobile social games, decreasing churn is decisive to increase player retention and rise revenues. Churn prediction models allow to understand player loyalty and to anticipate when they will stop playing a game. Thanks to these predictions, several initiatives can be taken to retain those players who are more likely to churn. Survival analysis focuses on predicting the time of occurrence of a certain event, churn in our case. Classical methods, like regressions, could be applied only when all players have left the game. The challenge arises for datasets with incomplete churning information for all players, as most of them still connect to the game. This is called a censored data problem and is in the nature of churn. Censoring is commonly dealt with survival analysis techniques, but due to the inflexibility of the survival statistical algorithms, the accuracy achieved is often poor. In contrast, novel ensemble learning techniques, increasingly popular in a variety of scientific fields, provide high-class prediction results. In this work, we develop, for the first time in the social games domain, a survival ensemble model which provides a comprehensive analysis together with an accurate prediction of churn. For each player, we predict the probability of churning as function of time, which permits to distinguish various levels of loyalty profiles. Additionally, we assess the risk factors that explain the predicted player survival times. Our results show that churn prediction by survival ensembles significantly improves the accuracy and robustness of traditional analyses, like Cox regression.
Tasks	Survival Analysis
Published	2017-10-06
URL	http://arxiv.org/abs/1710.02264v1
PDF	http://arxiv.org/pdf/1710.02264v1.pdf
PWC	https://paperswithcode.com/paper/churn-prediction-in-mobile-social-games
Repo
Framework

Scavenger 0.1: A Theorem Prover Based on Conflict Resolution


Title	Scavenger 0.1: A Theorem Prover Based on Conflict Resolution
Authors	Daniyar Itegulov, John Slaney, Bruno Woltzenlogel Paleo
Abstract	This paper introduces Scavenger, the first theorem prover for pure first-order logic without equality based on the new conflict resolution calculus. Conflict resolution has a restricted resolution inference rule that resembles (a first-order generalization of) unit propagation as well as a rule for assuming decision literals and a rule for deriving new clauses by (a first-order generalization of) conflict-driven clause learning.
Tasks
Published	2017-04-11
URL	http://arxiv.org/abs/1704.03275v2
PDF	http://arxiv.org/pdf/1704.03275v2.pdf
PWC	https://paperswithcode.com/paper/scavenger-01-a-theorem-prover-based-on
Repo
Framework

Local Deep Neural Networks for Age and Gender Classification


Title	Local Deep Neural Networks for Age and Gender Classification
Authors	Zukang Liao, Stavros Petridis, Maja Pantic
Abstract	Local deep neural networks have been recently introduced for gender recognition. Although, they achieve very good performance they are very computationally expensive to train. In this work, we introduce a simplified version of local deep neural networks which significantly reduces the training time. Instead of using hundreds of patches per image, as suggested by the original method, we propose to use 9 overlapping patches per image which cover the entire face region. This results in a much reduced training time, since just 9 patches are extracted per image instead of hundreds, at the expense of a slightly reduced performance. We tested the proposed modified local deep neural networks approach on the LFW and Adience databases for the task of gender and age classification. For both tasks and both databases the performance is up to 1% lower compared to the original version of the algorithm. We have also investigated which patches are more discriminative for age and gender classification. It turns out that the mouth and eyes regions are useful for age classification, whereas just the eye region is useful for gender classification.
Tasks	Age And Gender Classification
Published	2017-03-24
URL	http://arxiv.org/abs/1703.08497v1
PDF	http://arxiv.org/pdf/1703.08497v1.pdf
PWC	https://paperswithcode.com/paper/local-deep-neural-networks-for-age-and-gender
Repo
Framework

Exact Tensor Completion from Sparsely Corrupted Observations via Convex Optimization


Title	Exact Tensor Completion from Sparsely Corrupted Observations via Convex Optimization
Authors	Jonathan Q. Jiang, Michael K. Ng
Abstract	This paper conducts a rigorous analysis for provable estimation of multidimensional arrays, in particular third-order tensors, from a random subset of its corrupted entries. Our study rests heavily on a recently proposed tensor algebraic framework in which we can obtain tensor singular value decomposition (t-SVD) that is similar to the SVD for matrices, and define a new notion of tensor rank referred to as the tubal rank. We prove that by simply solving a convex program, which minimizes a weighted combination of tubal nuclear norm, a convex surrogate for the tubal rank, and the $\ell_1$-norm, one can recover an incoherent tensor exactly with overwhelming probability, provided that its tubal rank is not too large and that the corruptions are reasonably sparse. Interestingly, our result includes the recovery guarantees for the problems of tensor completion (TC) and tensor principal component analysis (TRPCA) under the same algebraic setup as special cases. An alternating direction method of multipliers (ADMM) algorithm is presented to solve this optimization problem. Numerical experiments verify our theory and real-world applications demonstrate the effectiveness of our algorithm.
Tasks
Published	2017-08-02
URL	http://arxiv.org/abs/1708.00601v1
PDF	http://arxiv.org/pdf/1708.00601v1.pdf
PWC	https://paperswithcode.com/paper/exact-tensor-completion-from-sparsely
Repo
Framework


Title	Optimal Policies for Observing Time Series and Related Restless Bandit Problems
Authors	Christopher R. Dance, Tomi Silander
Abstract	The trade-off between the cost of acquiring and processing data, and uncertainty due to a lack of data is fundamental in machine learning. A basic instance of this trade-off is the problem of deciding when to make noisy and costly observations of a discrete-time Gaussian random walk, so as to minimise the posterior variance plus observation costs. We present the first proof that a simple policy, which observes when the posterior variance exceeds a threshold, is optimal for this problem. The proof generalises to a wide range of cost functions other than the posterior variance. This result implies that optimal policies for linear-quadratic-Gaussian control with costly observations have a threshold structure. It also implies that the restless bandit problem of observing multiple such time series, has a well-defined Whittle index. We discuss computation of that index, give closed-form formulae for it, and compare the performance of the associated index policy with heuristic policies. The proof is based on a new verification theorem that demonstrates threshold structure for Markov decision processes, and on the relation between binary sequences known as mechanical words and the dynamics of discontinuous nonlinear maps, which frequently arise in physics, control and biology.
Tasks	Time Series
Published	2017-03-29
URL	http://arxiv.org/abs/1703.10010v1
PDF	http://arxiv.org/pdf/1703.10010v1.pdf
PWC	https://paperswithcode.com/paper/optimal-policies-for-observing-time-series
Repo
Framework

Exploiting saliency for object segmentation from image level labels


Title	Exploiting saliency for object segmentation from image level labels
Authors	Seong Joon Oh, Rodrigo Benenson, Anna Khoreva, Zeynep Akata, Mario Fritz, Bernt Schiele
Abstract	There have been remarkable improvements in the semantic labelling task in the recent years. However, the state of the art methods rely on large-scale pixel-level annotations. This paper studies the problem of training a pixel-wise semantic labeller network from image-level annotations of the present object classes. Recently, it has been shown that high quality seeds indicating discriminative object regions can be obtained from image-level labels. Without additional information, obtaining the full extent of the object is an inherently ill-posed problem due to co-occurrences. We propose using a saliency model as additional information and hereby exploit prior knowledge on the object extent and image statistics. We show how to combine both information sources in order to recover 80% of the fully supervised performance - which is the new state of the art in weakly supervised training for pixel-wise semantic labelling. The code is available at https://goo.gl/KygSeb.
Tasks	Semantic Segmentation
Published	2017-01-28
URL	http://arxiv.org/abs/1701.08261v2
PDF	http://arxiv.org/pdf/1701.08261v2.pdf
PWC	https://paperswithcode.com/paper/exploiting-saliency-for-object-segmentation
Repo
Framework

Probabilistically Safe Policy Transfer


Title	Probabilistically Safe Policy Transfer
Authors	David Held, Zoe McCarthy, Michael Zhang, Fred Shentu, Pieter Abbeel
Abstract	Although learning-based methods have great potential for robotics, one concern is that a robot that updates its parameters might cause large amounts of damage before it learns the optimal policy. We formalize the idea of safe learning in a probabilistic sense by defining an optimization problem: we desire to maximize the expected return while keeping the expected damage below a given safety limit. We study this optimization for the case of a robot manipulator with safety-based torque limits. We would like to ensure that the damage constraint is maintained at every step of the optimization and not just at convergence. To achieve this aim, we introduce a novel method which predicts how modifying the torque limit, as well as how updating the policy parameters, might affect the robot’s safety. We show through a number of experiments that our approach allows the robot to improve its performance while ensuring that the expected damage constraint is not violated during the learning process.
Tasks
Published	2017-05-15
URL	http://arxiv.org/abs/1705.05394v1
PDF	http://arxiv.org/pdf/1705.05394v1.pdf
PWC	https://paperswithcode.com/paper/probabilistically-safe-policy-transfer
Repo
Framework