January 27, 2020

3316 words 16 mins read

Paper Group ANR 1133

Separable Convolutional LSTMs for Faster Video Segmentation. Risk-Averse Action Selection Using Extreme Value Theory Estimates of the CVaR. Quantifying and Leveraging Classification Uncertainty for Chest Radiograph Assessment. An analysis of the cost of hyper-parameter selection via split-sample validation, with applications to penalized regression …

Separable Convolutional LSTMs for Faster Video Segmentation


Title	Separable Convolutional LSTMs for Faster Video Segmentation
Authors	Andreas Pfeuffer, Klaus Dietmayer
Abstract	Semantic Segmentation is an important module for autonomous robots such as self-driving cars. The advantage of video segmentation approaches compared to single image segmentation is that temporal image information is considered, and their performance increases due to this. Hence, single image segmentation approaches are extended by recurrent units such as convolutional LSTM (convLSTM) cells, which are placed at suitable positions in the basic network architecture. However, a major critique of video segmentation approaches based on recurrent neural networks is their large parameter count and their computational complexity, and so, their inference time of one video frame takes up to 66 percent longer than their basic version. Inspired by the success of the spatial and depthwise separable convolutional neural networks, we generalize these techniques for convLSTMs in this work, so that the number of parameters and the required FLOPs are reduced significantly. Experiments on different datasets show that the segmentation approaches using the proposed, modified convLSTM cells achieve similar or slightly worse accuracy, but are up to 15 percent faster on a GPU than the ones using the standard convLSTM cells. Furthermore, a new evaluation metric is introduced, which measures the amount of flickering pixels in the segmented video sequence.
Tasks	Self-Driving Cars, Semantic Segmentation, Video Semantic Segmentation
Published	2019-07-16
URL	https://arxiv.org/abs/1907.06876v1
PDF	https://arxiv.org/pdf/1907.06876v1.pdf
PWC	https://paperswithcode.com/paper/separable-convolutional-lstms-for-faster
Repo
Framework

Risk-Averse Action Selection Using Extreme Value Theory Estimates of the CVaR


Title	Risk-Averse Action Selection Using Extreme Value Theory Estimates of the CVaR
Authors	Dylan Troop, Frédéric Godin, Jia Yuan Yu
Abstract	The Conditional Value-at-Risk (CVaR) is a useful risk measure in machine learning, finance, insurance, energy, etc. When the CVaR confidence parameter is very high, estimation by sample averaging exhibits high variance due to the limited number of samples above the corresponding threshold. To mitigate this problem, we present an estimation procedure for the CVaR that combines extreme value theory and a recently introduced method of automated threshold selection by Bader et al. (2018). Under appropriate conditions, we estimate the tail risk using a generalized Pareto distribution. We compare empirically this estimation procedure with the naive method of sample averaging, and show an improvement in accuracy for some specific cases. We also show how the estimation procedure can be used in reinforcement learning by applying our method to the multi-armed bandit problem where the goal is to avoid catastrophic risk.
Tasks
Published	2019-12-03
URL	https://arxiv.org/abs/1912.01718v1
PDF	https://arxiv.org/pdf/1912.01718v1.pdf
PWC	https://paperswithcode.com/paper/risk-averse-action-selection-using-extreme
Repo
Framework

Quantifying and Leveraging Classification Uncertainty for Chest Radiograph Assessment


Title	Quantifying and Leveraging Classification Uncertainty for Chest Radiograph Assessment
Authors	Florin C. Ghesu, Bogdan Georgescu, Eli Gibson, Sebastian Guendel, Mannudeep K. Kalra, Ramandeep Singh, Subba R. Digumarthy, Sasa Grbic, Dorin Comaniciu
Abstract	The interpretation of chest radiographs is an essential task for the detection of thoracic diseases and abnormalities. However, it is a challenging problem with high inter-rater variability and inherent ambiguity due to inconclusive evidence in the data, limited data quality or subjective definitions of disease appearance. Current deep learning solutions for chest radiograph abnormality classification are typically limited to providing probabilistic predictions, relying on the capacity of learning models to adapt to the high degree of label noise and become robust to the enumerated causal factors. In practice, however, this leads to overconfident systems with poor generalization on unseen data. To account for this, we propose an automatic system that learns not only the probabilistic estimate on the presence of an abnormality, but also an explicit uncertainty measure which captures the confidence of the system in the predicted output. We argue that explicitly learning the classification uncertainty as an orthogonal measure to the predicted output, is essential to account for the inherent variability characteristic of this data. Experiments were conducted on two datasets of chest radiographs of over 85,000 patients. Sample rejection based on the predicted uncertainty can significantly improve the ROC-AUC, e.g., by 8% to 0.91 with an expected rejection rate of under 25%. Eliminating training samples using uncertainty-driven bootstrapping, enables a significant increase in robustness and accuracy. In addition, we present a multi-reader study showing that the predictive uncertainty is indicative of reader errors.
Tasks
Published	2019-06-18
URL	https://arxiv.org/abs/1906.07775v1
PDF	https://arxiv.org/pdf/1906.07775v1.pdf
PWC	https://paperswithcode.com/paper/quantifying-and-leveraging-classification
Repo
Framework

An analysis of the cost of hyper-parameter selection via split-sample validation, with applications to penalized regression


Title	An analysis of the cost of hyper-parameter selection via split-sample validation, with applications to penalized regression
Authors	Jean Feng, Noah Simon
Abstract	In the regression setting, given a set of hyper-parameters, a model-estimation procedure constructs a model from training data. The optimal hyper-parameters that minimize generalization error of the model are usually unknown. In practice they are often estimated using split-sample validation. Up to now, there is an open question regarding how the generalization error of the selected model grows with the number of hyper-parameters to be estimated. To answer this question, we establish finite-sample oracle inequalities for selection based on a single training/test split and based on cross-validation. We show that if the model-estimation procedures are smoothly parameterized by the hyper-parameters, the error incurred from tuning hyper-parameters shrinks at nearly a parametric rate. Hence for semi- and non-parametric model-estimation procedures with a fixed number of hyper-parameters, this additional error is negligible. For parametric model-estimation procedures, adding a hyper-parameter is roughly equivalent to adding a parameter to the model itself. In addition, we specialize these ideas for penalized regression problems with multiple penalty parameters. We establish that the fitted models are Lipschitz in the penalty parameters and thus our oracle inequalities apply. This result encourages development of regularization methods with many penalty parameters.
Tasks
Published	2019-03-28
URL	http://arxiv.org/abs/1903.12297v1
PDF	http://arxiv.org/pdf/1903.12297v1.pdf
PWC	https://paperswithcode.com/paper/an-analysis-of-the-cost-of-hyper-parameter
Repo
Framework

Experimentation on the motion of an obstacle avoiding robot


Title	Experimentation on the motion of an obstacle avoiding robot
Authors	Rakhmanov Ochilbek, Nzurumike Obianuju, Amina Sani, Rukayya Umar
Abstract	An intelligent robot can be used for applications where a human is at significant risk (like nuclear, space, military), the economics or menial nature of the application result in inefficient use of human workers (service industry, agriculture), for humanitarian uses where there is great risk (demining an area of land mines, urban search and rescue). This paper implements an experiment on one of important fields of AI Searching Algorithms, to find shortest possible solution by searching the produced tree. We will concentrate on Hill climbing algorithm, which is one of simplest searching algorithms in AI. This algorithm is one of most suitable searching methods to help expert system to make decision at every state, at every node. The experimental robot will traverse the maze by using sensors plugged on it. The robot used is E.V.3 Lego Mind storms, with native software for programming LabView. The reason we chose this robot is that it interacts quickly with sensors and can be reconstructed in many ways. This programmed robot will calculate the best possibilities to find way out of maze. The maze is made of wood, and it is adjustable, as robot should be able to leave the maze in any design.
Tasks
Published	2019-07-25
URL	https://arxiv.org/abs/1907.11021v1
PDF	https://arxiv.org/pdf/1907.11021v1.pdf
PWC	https://paperswithcode.com/paper/experimentation-on-the-motion-of-an-obstacle
Repo
Framework

Soft-Bayes: Prod for Mixtures of Experts with Log-Loss


Title	Soft-Bayes: Prod for Mixtures of Experts with Log-Loss
Authors	Laurent Orseau, Tor Lattimore, Shane Legg
Abstract	We consider prediction with expert advice under the log-loss with the goal of deriving efficient and robust algorithms. We argue that existing algorithms such as exponentiated gradient, online gradient descent and online Newton step do not adequately satisfy both requirements. Our main contribution is an analysis of the Prod algorithm that is robust to any data sequence and runs in linear time relative to the number of experts in each round. Despite the unbounded nature of the log-loss, we derive a bound that is independent of the largest loss and of the largest gradient, and depends only on the number of experts and the time horizon. Furthermore we give a Bayesian interpretation of Prod and adapt the algorithm to derive a tracking regret.
Tasks
Published	2019-01-08
URL	http://arxiv.org/abs/1901.02230v1
PDF	http://arxiv.org/pdf/1901.02230v1.pdf
PWC	https://paperswithcode.com/paper/soft-bayes-prod-for-mixtures-of-experts-with
Repo
Framework

Image Super Resolution via Bilinear Pooling: Application to Confocal Endomicroscopy


Title	Image Super Resolution via Bilinear Pooling: Application to Confocal Endomicroscopy
Authors	Saeed Izadi, Darren Sutton, Ghassan Hamarneh
Abstract	Recent developments in image acquisition literature have miniaturized the confocal laser endomicroscopes to improve usability and flexibility of the apparatus in actual clinical settings. However, miniaturized devices collect less light and have fewer optical components, resulting in pixelation artifacts and low resolution images. Owing to the strength of deep networks, many supervised methods known as super resolution have achieved considerable success in restoring low resolution images by generating the missing high frequency details. In this work, we propose a novel attention mechanism that, for the first time, combines 1st- and 2nd-order statistics for pooling operation, in the spatial and channel-wise dimensions. We compare the efficacy of our method to 11 other existing single image super resolution techniques that compensate for the reduction in image quality caused by the necessity of endomicroscope miniaturization. All evaluations are carried out on three publicly available datasets. Experimental results show that our method can produce competitive results against state-of-the-art in terms of PSNR, SSIM, and IFC metrics. Additionally, our proposed method contains small number of parameters, which makes it lightweight and fast for real-time applications.
Tasks	Image Super-Resolution, Super-Resolution
Published	2019-06-18
URL	https://arxiv.org/abs/1906.07802v2
PDF	https://arxiv.org/pdf/1906.07802v2.pdf
PWC	https://paperswithcode.com/paper/image-super-resolution-via-bilinear-pooling
Repo
Framework

Kidney Recognition in CT Using YOLOv3


Title	Kidney Recognition in CT Using YOLOv3
Authors	Andréanne Lemay
Abstract	Organ localization can be challenging considering the heterogeneity of medical images and the biological diversity from one individual to another. The contribution of this paper is to overview the performance of the object detection model, YOLOv3, on kidney localization in 2D and in 3D from CT scans. The model obtained a 0.851 Dice score in 2D and 0.742 in 3D. The SSD, a similar state-of-the-art object detection model, showed similar scores on the test set. YOLOv3 and SSD demonstrated the ability to detect kidneys on a wide variety of CT scans including patients suffering from different renal conditions.
Tasks	Object Detection
Published	2019-10-03
URL	https://arxiv.org/abs/1910.01268v1
PDF	https://arxiv.org/pdf/1910.01268v1.pdf
PWC	https://paperswithcode.com/paper/kidney-recognition-in-ct-using-yolov3
Repo
Framework

Using ConceptNet to Teach Common Sense to an Automated Theorem Prover


Title	Using ConceptNet to Teach Common Sense to an Automated Theorem Prover
Authors	Claudia Schon, Sophie Siebert, Frieder Stolzenburg
Abstract	The CoRg system is a system to solve commonsense reasoning problems. The core of the CoRg system is the automated theorem prover Hyper that is fed with large amounts of background knowledge. This background knowledge plays a crucial role in solving commonsense reasoning problems. In this paper we present different ways to use knowledge graphs as background knowledge and discuss challenges that arise.
Tasks	Common Sense Reasoning, Knowledge Graphs
Published	2019-12-30
URL	https://arxiv.org/abs/1912.12957v1
PDF	https://arxiv.org/pdf/1912.12957v1.pdf
PWC	https://paperswithcode.com/paper/using-conceptnet-to-teach-common-sense-to-an
Repo
Framework

ODE-Inspired Analysis for the Biological Version of Oja’s Rule in Solving Streaming PCA


Title	ODE-Inspired Analysis for the Biological Version of Oja’s Rule in Solving Streaming PCA
Authors	Chi-Ning Chou, Mien Brabeeba Wang
Abstract	Oja’s rule [Oja, Journal of mathematical biology 1982] is a well-known biologically-plausible algorithm using a Hebbian-type synaptic update rule to solve streaming principal component analysis (PCA). Computational neuroscientists have known that this biological version of Oja’s rule converges to the top eigenvector of the covariance matrix of the input in the limit. However, prior to this work, it was open to prove any convergence rate guarantee. In this work, we give the first convergence rate analysis for the biological version of Oja’s rule in solving streaming PCA. Moreover, our convergence rate matches the information theoretical lower bound up to logarithmic factors and outperforms the state-of-the-art upper bound for streaming PCA. Furthermore, we develop a novel framework inspired by ordinary differential equations (ODE) to analyze general stochastic dynamics. The framework abandons the traditional step-by-step analysis and instead analyzes a stochastic dynamic in one-shot by giving a closed-form solution to the entire dynamic. The one-shot framework allows us to apply stopping time and martingale techniques to have a flexible and precise control on the dynamic. We believe that this general framework is powerful and should lead to effective yet simple analysis for a large class of problems with stochastic dynamics.
Tasks
Published	2019-11-04
URL	https://arxiv.org/abs/1911.02363v1
PDF	https://arxiv.org/pdf/1911.02363v1.pdf
PWC	https://paperswithcode.com/paper/ode-inspired-analysis-for-the-biological
Repo
Framework

A Unified Analytical Framework for Trustable Machine Learning and Automation Running with Blockchain


Title	A Unified Analytical Framework for Trustable Machine Learning and Automation Running with Blockchain
Authors	Tao Wang
Abstract	Traditional machine learning algorithms use data from databases that are mutable, and therefore the data cannot be fully trusted. Also, the machine learning process is difficult to automate. This paper proposes building a trustable machine learning system by using blockchain technology, which can store data in a permanent and immutable way. In addition, smart contracts are used to automate the machine learning process. This paper makes three contributions. First, it establishes a link between machine learning technology and blockchain technology. Previously, machine learning and blockchain have been considered two independent technologies without an obvious link. Second, it proposes a unified analytical framework for trustable machine learning by using blockchain technology. This unified framework solves both the trustability and automation issues in machine learning. Third, it enables a computer to translate core machine learning implementation from a single thread on a single machine to multiple threads on multiple machines running with blockchain by using a unified approach. The paper uses association rule mining as an example to demonstrate how trustable machine learning can be implemented with blockchain, and it shows how this approach can be used to analyze opioid prescriptions to help combat the opioid crisis.
Tasks
Published	2019-03-21
URL	http://arxiv.org/abs/1903.08801v1
PDF	http://arxiv.org/pdf/1903.08801v1.pdf
PWC	https://paperswithcode.com/paper/a-unified-analytical-framework-for-trustable
Repo
Framework

ST-UNet: A Spatio-Temporal U-Network for Graph-structured Time Series Modeling


Title	ST-UNet: A Spatio-Temporal U-Network for Graph-structured Time Series Modeling
Authors	Bing Yu, Haoteng Yin, Zhanxing Zhu
Abstract	The spatio-temporal graph learning is becoming an increasingly important object of graph study. Many application domains involve highly dynamic graphs where temporal information is crucial, e.g. traffic networks and financial transaction graphs. Despite the constant progress made on learning structured data, there is still a lack of effective means to extract dynamic complex features from spatio-temporal structures. Particularly, conventional models such as convolutional networks or recurrent neural networks are incapable of revealing the temporal patterns in short or long terms and exploring the spatial properties in local or global scope from spatio-temporal graphs simultaneously. To tackle this problem, we design a novel multi-scale architecture, Spatio-Temporal U-Net (ST-UNet), for graph-structured time series modeling. In this U-shaped network, a paired sampling operation is proposed in spacetime domain accordingly: the pooling (ST-Pool) coarsens the input graph in spatial from its deterministic partition while abstracts multi-resolution temporal dependencies through dilated recurrent skip connections; based on previous settings in the downsampling, the unpooling (ST-Unpool) restores the original structure of spatio-temporal graphs and resumes regular intervals within graph sequences. Experiments on spatio-temporal prediction tasks demonstrate that our model effectively captures comprehensive features in multiple scales and achieves substantial improvements over mainstream methods on several real-world datasets.
Tasks	Time Series, Traffic Prediction
Published	2019-03-13
URL	http://arxiv.org/abs/1903.05631v1
PDF	http://arxiv.org/pdf/1903.05631v1.pdf
PWC	https://paperswithcode.com/paper/st-unet-a-spatio-temporal-u-network-for-graph
Repo
Framework

Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation


Title	Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation
Authors	Hanyang Kong, Jian Zhao, Xiaoguang Tu, Junliang Xing, Shengmei Shen, Jiashi Feng
Abstract	Recent deep learning based face recognition methods have achieved great performance, but it still remains challenging to recognize very low-resolution query face like 28x28 pixels when CCTV camera is far from the captured subject. Such face with very low-resolution is totally out of detail information of the face identity compared to normal resolution in a gallery and hard to find corresponding faces therein. To this end, we propose a Resolution Invariant Model (RIM) for addressing such cross-resolution face recognition problems, with three distinct novelties. First, RIM is a novel and unified deep architecture, containing a Face Hallucination sub-Net (FHN) and a Heterogeneous Recognition sub-Net (HRN), which are jointly learned end to end. Second, FHN is a well-designed tri-path Generative Adversarial Network (GAN) which simultaneously perceives facial structure and geometry prior information, i.e. landmark heatmaps and parsing maps, incorporated with an unsupervised cross-domain adversarial training strategy to super-resolve very low-resolution query image to its 8x larger ones without requiring them to be well aligned. Third, HRN is a generic Convolutional Neural Network (CNN) for heterogeneous face recognition with our proposed residual knowledge distillation strategy for learning discriminative yet generalized feature representation. Quantitative and qualitative experiments on several benchmarks demonstrate the superiority of the proposed model over the state-of-the-arts. Codes and models will be released upon acceptance.
Tasks	Face Hallucination, Face Recognition, Heterogeneous Face Recognition
Published	2019-05-26
URL	https://arxiv.org/abs/1905.10777v1
PDF	https://arxiv.org/pdf/1905.10777v1.pdf
PWC	https://paperswithcode.com/paper/cross-resolution-face-recognition-via-prior
Repo
Framework

Carving out the low surface brightness universe with NoiseChisel


Title	Carving out the low surface brightness universe with NoiseChisel
Authors	Mohammad Akhlaghi
Abstract	NoiseChisel is a program to detect very low signal-to-noise ratio (S/N) features with minimal assumptions on their morphology. It was introduced in 2015 and released within a collection of data analysis programs and libraries known as GNU Astronomy Utilities (Gnuastro). Over the last ten stable releases of Gnuastro, NoiseChisel has significantly improved: detecting even fainter signal, enabling better user control over its inner workings, and many bug fixes. The most important change may be that NoiseChisel’s segmentation features have been moved into a new program called Segment. Another major change is the final growth strategy of its true detections, for example NoiseChisel is able to detect the outer wings of M51 down to S/N of 0.25, or 28.27 mag/arcsec2 on a single-exposure SDSS image (r-band). Segment is also able to detect the localized HII regions as “clumps” much more successfully. Finally, to orchestrate a controlled analysis, the concept of a “reproducible paper” is discussed: this paper itself is exactly reproducible (snapshot v4-0-g8505cfd).
Tasks
Published	2019-09-24
URL	https://arxiv.org/abs/1909.11230v1
PDF	https://arxiv.org/pdf/1909.11230v1.pdf
PWC	https://paperswithcode.com/paper/carving-out-the-low-surface-brightness
Repo
Framework

Low-Power Neuromorphic Hardware for Signal Processing Applications


Title	Low-Power Neuromorphic Hardware for Signal Processing Applications
Authors	Bipin Rajendran, Abu Sebastian, Michael Schmuker, Narayan Srinivasa, Evangelos Eleftheriou
Abstract	Machine learning has emerged as the dominant tool for implementing complex cognitive tasks that require supervised, unsupervised, and reinforcement learning. While the resulting machines have demonstrated in some cases even super-human performance, their energy consumption has often proved to be prohibitive in the absence of costly super-computers. Most state-of-the-art machine learning solutions are based on memory-less models of neurons. This is unlike the neurons in the human brain, which encode and process information using temporal information in spike events. The different computing principles underlying biological neurons and how they combine together to efficiently process information is believed to be a key factor behind their superior efficiency compared to current machine learning systems. Inspired by the time-encoding mechanism used by the brain, third generation spiking neural networks (SNNs) are being studied for building a new class of information processing engines. Modern computing systems based on the von Neumann architecture, however, are ill-suited for efficiently implementing SNNs, since their performance is limited by the need to constantly shuttle data between physically separated logic and memory units. Hence, novel computational architectures that address the von Neumann bottleneck are necessary in order to build systems that can implement SNNs with low energy budgets. In this paper, we review some of the architectural and system level design aspects involved in developing a new class of brain-inspired information processing engines that mimic the time-based information encoding and processing aspects of the brain.
Tasks
Published	2019-01-11
URL	https://arxiv.org/abs/1901.03690v3
PDF	https://arxiv.org/pdf/1901.03690v3.pdf
PWC	https://paperswithcode.com/paper/low-power-neuromorphic-hardware-for-signal
Repo
Framework