October 20, 2019

3340 words 16 mins read

Paper Group ANR 62

Image Denoising and Super-Resolution using Residual Learning of Deep Convolutional Network. Radius-margin bounds for deep neural networks. Novel Methods for Enhancing the Performance of Genetic Algorithms. A Unified Framework for Planning in Adversarial and Cooperative Environments. CT organ segmentation using GPU data augmentation, unsupervised la …

Image Denoising and Super-Resolution using Residual Learning of Deep Convolutional Network


Title	Image Denoising and Super-Resolution using Residual Learning of Deep Convolutional Network
Authors	Rohit Pardasani, Utkarsh Shreemali
Abstract	Image super-resolution and denoising are two important tasks in image processing that can lead to improvement in image quality. Image super-resolution is the task of mapping a low resolution image to a high resolution image whereas denoising is the task of learning a clean image from a noisy input. We propose and train a single deep learning network that we term as SuRDCNN (super-resolution and denoising convolutional neural network), to perform these two tasks simultaneously . Our model nearly replicates the architecture of existing state-of-the-art deep learning models for super-resolution and denoising. We use the proven strategy of residual learning, as supported by state-of-the-art networks in this domain. Our trained SuRDCNN is capable of super-resolving image in the presence of Gaussian noise, Poisson noise or any random combination of both of these noises.
Tasks	Denoising, Image Denoising, Image Super-Resolution, Super-Resolution
Published	2018-09-21
URL	http://arxiv.org/abs/1809.08229v1
PDF	http://arxiv.org/pdf/1809.08229v1.pdf
PWC	https://paperswithcode.com/paper/image-denoising-and-super-resolution-using
Repo
Framework

Radius-margin bounds for deep neural networks


Title	Radius-margin bounds for deep neural networks
Authors	Mayank Sharma, Jayadeva, Sumit Soman
Abstract	Explaining the unreasonable effectiveness of deep learning has eluded researchers around the globe. Various authors have described multiple metrics to evaluate the capacity of deep architectures. In this paper, we allude to the radius margin bounds described for a support vector machine (SVM) with hinge loss, apply the same to the deep feed-forward architectures and derive the Vapnik-Chervonenkis (VC) bounds which are different from the earlier bounds proposed in terms of number of weights of the network. In doing so, we also relate the effectiveness of techniques like Dropout and Dropconnect in bringing down the capacity of the network. Finally, we describe the effect of maximizing the input as well as the output margin to achieve an input noise-robust deep architecture.
Tasks
Published	2018-11-03
URL	http://arxiv.org/abs/1811.01171v1
PDF	http://arxiv.org/pdf/1811.01171v1.pdf
PWC	https://paperswithcode.com/paper/radius-margin-bounds-for-deep-neural-networks
Repo
Framework

Novel Methods for Enhancing the Performance of Genetic Algorithms


Title	Novel Methods for Enhancing the Performance of Genetic Algorithms
Authors	Esra’a O Alkafaween
Abstract	In this thesis we propose new methods for crossover operator namely: cut on worst gene (COWGC), cut on worst L+R gene (COWLRGC) and Collision Crossovers. And also we propose several types of mutation operator such as: worst gene with random gene mutation (WGWRGM) , worst LR gene with random gene mutation (WLRGWRGM), worst gene with worst gene mutation (WGWWGM), worst gene with nearest neighbour mutation (WGWNNM), worst gene with the worst around the nearest neighbour mutation (WGWWNNM), worst gene inserted beside nearest neighbour mutation (WGIBNNM), random gene inserted beside nearest neighbour mutation (RGIBNNM), Swap worst gene locally mutation (SWGLM), Insert best random gene before worst gene mutation (IBRGBWGM) and Insert best random gene before random gene mutation (IBRGBRGM). In addition to proposing four selection strategies, namely: select any crossover (SAC), select any mutation (SAM), select best crossover (SBC) and select best mutation (SBM). The first two are based on selection of the best crossover and mutation operator respectively, and the other two strategies randomly select any operator. So we investigate the use of more than one crossover/mutation operator (based on the proposed strategies) to enhance the performance of genetic algorithms. Our experiments, conducted on several Travelling Salesman Problems (TSP), show the superiority of some of the proposed methods in crossover and mutation over some of the well-known crossover and mutation operators described in the literature. In addition, using any of the four strategies (SAC, SAM, SBC and SBM), found to be better than using one crossover/mutation operator in general, because those allow the GA to avoid local optima, or the so-called premature convergence. Keywords: GAs, Collision crossover, Multi crossovers, Multi mutations, TSP.
Tasks
Published	2018-01-09
URL	http://arxiv.org/abs/1801.02827v3
PDF	http://arxiv.org/pdf/1801.02827v3.pdf
PWC	https://paperswithcode.com/paper/novel-methods-for-enhancing-the-performance
Repo
Framework

A Unified Framework for Planning in Adversarial and Cooperative Environments


Title	A Unified Framework for Planning in Adversarial and Cooperative Environments
Authors	Anagha Kulkarni, Siddharth Srivastava, Subbarao Kambhampati
Abstract	Users of AI systems may rely upon them to produce plans for achieving desired objectives. Such AI systems should be able to compute obfuscated plans whose execution in adversarial situations protects privacy, as well as legible plans which are easy for team members to understand in cooperative situations. We develop a unified framework that addresses these dual problems by computing plans with a desired level of comprehensibility from the point of view of a partially informed observer. For adversarial settings, our approach produces obfuscated plans with observations that are consistent with at least k goals from a set of decoy goals. By slightly varying our framework, we present an approach for goal legibility in cooperative settings which produces plans that achieve a goal while being consistent with at most j goals from a set of confounding goals. In addition, we show how the observability of the observer can be controlled to either obfuscate or clarify the next actions in a plan when the goal is known to the observer. We present theoretical results on the complexity analysis of our problems. We demonstrate the execution of obfuscated and legible plans in a cooking domain using a physical robot Fetch. We also provide an empirical evaluation to show the feasibility and usefulness of our approaches using IPC domains.
Tasks
Published	2018-02-16
URL	http://arxiv.org/abs/1802.06137v3
PDF	http://arxiv.org/pdf/1802.06137v3.pdf
PWC	https://paperswithcode.com/paper/a-unified-framework-for-planning-in
Repo
Framework

CT organ segmentation using GPU data augmentation, unsupervised labels and IOU loss


Title	CT organ segmentation using GPU data augmentation, unsupervised labels and IOU loss
Authors	Blaine Rister, Darvin Yi, Kaushik Shivakumar, Tomomi Nobashi, Daniel L. Rubin
Abstract	Fully-convolutional neural networks have achieved superior performance in a variety of image segmentation tasks. However, their training requires laborious manual annotation of large datasets, as well as acceleration by parallel processors with high-bandwidth memory, such as GPUs. We show that simple models can achieve competitive accuracy for organ segmentation on CT images when trained with extensive data augmentation, which leverages existing graphics hardware to quickly apply geometric and photometric transformations to 3D image data. On 3 mm^3 CT volumes, our GPU implementation is 2.6-8X faster than a widely-used CPU version, including communication overhead. We also show how to automatically generate training labels using rudimentary morphological operations, which are efficiently computed by 3D Fourier transforms. We combined fully-automatic labels for the lungs and bone with semi-automatic ones for the liver, kidneys and bladder, to create a dataset of 130 labeled CT scans. To achieve the best results from data augmentation, our model uses the intersection-over-union (IOU) loss function, a close relative of the Dice loss. We discuss its mathematical properties and explain why it outperforms the usual weighted cross-entropy loss for unbalanced segmentation tasks. We conclude that there is no unique IOU loss function, as the naive one belongs to a broad family of functions with the same essential properties. When combining data augmentation with the IOU loss, our model achieves a Dice score of 78-92% for each organ. The trained model, code and dataset will be made publicly available, to further medical imaging research.
Tasks	Data Augmentation, Semantic Segmentation, Unbalanced Segmentation
Published	2018-11-27
URL	http://arxiv.org/abs/1811.11226v1
PDF	http://arxiv.org/pdf/1811.11226v1.pdf
PWC	https://paperswithcode.com/paper/ct-organ-segmentation-using-gpu-data
Repo
Framework

Fast Policy Learning through Imitation and Reinforcement


Title	Fast Policy Learning through Imitation and Reinforcement
Authors	Ching-An Cheng, Xinyan Yan, Nolan Wagener, Byron Boots
Abstract	Imitation learning (IL) consists of a set of tools that leverage expert demonstrations to quickly learn policies. However, if the expert is suboptimal, IL can yield policies with inferior performance compared to reinforcement learning (RL). In this paper, we aim to provide an algorithm that combines the best aspects of RL and IL. We accomplish this by formulating several popular RL and IL algorithms in a common mirror descent framework, showing that these algorithms can be viewed as a variation on a single approach. We then propose LOKI, a strategy for policy learning that first performs a small but random number of IL iterations before switching to a policy gradient RL method. We show that if the switching time is properly randomized, LOKI can learn to outperform a suboptimal expert and converge faster than running policy gradient from scratch. Finally, we evaluate the performance of LOKI experimentally in several simulated environments.
Tasks	Imitation Learning
Published	2018-05-26
URL	http://arxiv.org/abs/1805.10413v1
PDF	http://arxiv.org/pdf/1805.10413v1.pdf
PWC	https://paperswithcode.com/paper/fast-policy-learning-through-imitation-and
Repo
Framework

Circuit designs for superconducting optoelectronic loop neurons


Title	Circuit designs for superconducting optoelectronic loop neurons
Authors	Jeffrey M. Shainline, Sonia M. Buckley, Adam N. McCaughan, Jeff Chiles, Richard P. Mirin, Sae Woo Nam
Abstract	Optical communication achieves high fanout and short delay advantageous for information integration in neural systems. Superconducting detectors enable signaling with single photons for maximal energy efficiency. We present designs of superconducting optoelectronic neurons based on superconducting single-photon detectors, Josephson junctions, semiconductor light sources, and multi-planar dielectric waveguides. These circuits achieve complex synaptic and neuronal functions with high energy efficiency, leveraging the strengths of light for communication and superconducting electronics for computation. The neurons send few-photon signals to synaptic connections. These signals communicate neuronal firing events as well as update synaptic weights. Spike-timing-dependent plasticity is implemented with a single photon triggering each step of the process. Microscale light-emitting diodes and waveguide networks enable connectivity from a neuron to thousands of synaptic connections, and the use of light for communication enables synchronization of neurons across an area limited only by the distance light can travel within the period of a network oscillation. Experimentally, each of the requisite circuit elements has been demonstrated, yet a hardware platform combining them all has not been attempted. Compared to digital logic or quantum computing, device tolerances are relaxed. For this neural application, optical sources providing incoherent pulses with 10,000 photons produced with efficiency of 10$^{-3}$ operating at 20,MHz at 4.2,K are sufficient to enable a massively scalable neural computing platform with connectivity comparable to the brain and thirty thousand times higher speed.
Tasks
Published	2018-05-04
URL	http://arxiv.org/abs/1805.01947v2
PDF	http://arxiv.org/pdf/1805.01947v2.pdf
PWC	https://paperswithcode.com/paper/circuit-designs-for-superconducting
Repo
Framework

Scalar Arithmetic Multiple Data: Customizable Precision for Deep Neural Networks


Title	Scalar Arithmetic Multiple Data: Customizable Precision for Deep Neural Networks
Authors	Andrew Anderson, David Gregg
Abstract	Quantization of weights and activations in Deep Neural Networks (DNNs) is a powerful technique for network compression, and has enjoyed significant attention and success. However, much of the inference-time benefit of quantization is accessible only through the use of customized hardware accelerators or by providing an FPGA implementation of quantized arithmetic. Building on prior work, we show how to construct arbitrary bit-precise signed and unsigned integer operations using a software technique which logically \emph{embeds} a vector architecture with custom bit-width lanes in universally available fixed-width scalar arithmetic. We evaluate our approach on a high-end Intel Haswell processor, and an embedded ARM processor. Our approach yields very fast implementations of bit-precise custom DNN operations, which often match or exceed the performance of operations quantized to the sizes supported in native arithmetic. At the strongest level of quantization, our approach yields a maximum speedup of $\thicksim6\times$ on the Intel platform, and $\thicksim10\times$ on the ARM platform versus quantization to native 8-bit integers.
Tasks	Quantization
Published	2018-09-27
URL	https://arxiv.org/abs/1809.10572v2
PDF	https://arxiv.org/pdf/1809.10572v2.pdf
PWC	https://paperswithcode.com/paper/scalar-arithmetic-multiple-data-customizable
Repo
Framework

Human Activity Recognition using Recurrent Neural Networks


Title	Human Activity Recognition using Recurrent Neural Networks
Authors	Deepika Singh, Erinc Merdivan, Ismini Psychoula, Johannes Kropf, Sten Hanke, Matthieu Geist, Andreas Holzinger
Abstract	Human activity recognition using smart home sensors is one of the bases of ubiquitous computing in smart environments and a topic undergoing intense research in the field of ambient assisted living. The increasingly large amount of data sets calls for machine learning methods. In this paper, we introduce a deep learning model that learns to classify human activities without using any prior knowledge. For this purpose, a Long Short Term Memory (LSTM) Recurrent Neural Network was applied to three real world smart home datasets. The results of these experiments show that the proposed approach outperforms the existing ones in terms of accuracy and performance.
Tasks	Activity Recognition, Human Activity Recognition
Published	2018-04-19
URL	http://arxiv.org/abs/1804.07144v1
PDF	http://arxiv.org/pdf/1804.07144v1.pdf
PWC	https://paperswithcode.com/paper/human-activity-recognition-using-recurrent
Repo
Framework

Towards radiologist-level cancer risk assessment in CT lung screening using deep learning


Title	Towards radiologist-level cancer risk assessment in CT lung screening using deep learning
Authors	Stojan Trajanovski, Dimitrios Mavroeidis, Christine Leon Swisher, Binyam Gebrekidan Gebre, Bastiaan S. Veeling, Rafael Wiemker, Tobias Klinder, Amir Tahmasebi, Shawn M. Regis, Christoph Wald, Brady J. McKee, Sebastian Flacke, Heber MacMahon, Homer Pien
Abstract	Importance: Lung cancer is the leading cause of cancer mortality in the US, responsible for more deaths than breast, prostate, colon and pancreas cancer combined and it has been recently demonstrated that low-dose computed tomography (CT) screening of the chest can significantly reduce this death rate. Objective: To compare the performance of a deep learning model to state-of-the-art automated algorithms and radiologists as well as assessing the robustness of the algorithm in heterogeneous datasets. Design, Setting, and Participants: Three low-dose CT lung cancer screening datasets from heterogeneous sources were used, including National Lung Screening Trial (NLST, n=3410), Lahey Hospital and Medical Center (LHMC, n=3174) data, Kaggle competition data (from both stages, n=1595+505) and the University of Chicago data (UCM, a subset of NLST, annotated by radiologists, n=197). Relevant works on automated methods for Lung Cancer malignancy estimation have used significantly less data in size and diversity. At the first stage, our framework employs a nodule detector; while in the second stage, we use both the image area around the nodules and nodule features as inputs to a neural network that estimates the malignancy risk for the entire CT scan. We trained our two-stage algorithm on a part of the NLST dataset, and validated it on the other datasets. Results, Conclusions, and Relevance: The proposed deep learning model: (a) generalizes well across all three data sets, achieving AUC between 86% to 94%; (b) has better performance than the widely accepted PanCan Risk Model, achieving 11% better AUC score; (c) has improved performance compared to the state-of-the-art represented by the winners of the Kaggle Data Science Bowl 2017 competition on lung cancer screening; (d) has comparable performance to radiologists in estimating cancer risk at a patient level.
Tasks	Computed Tomography (CT)
Published	2018-04-05
URL	http://arxiv.org/abs/1804.01901v2
PDF	http://arxiv.org/pdf/1804.01901v2.pdf
PWC	https://paperswithcode.com/paper/towards-radiologist-level-cancer-risk
Repo
Framework

Inductive Framework for Multi-Aspect Streaming Tensor Completion with Side Information


Title	Inductive Framework for Multi-Aspect Streaming Tensor Completion with Side Information
Authors	Madhav Nimishakavi, Bamdev Mishra, Manish Gupta, Partha Talukdar
Abstract	Low rank tensor completion is a well studied problem and has applications in various fields. However, in many real world applications the data is dynamic, i.e., new data arrives at different time intervals. As a result, the tensors used to represent the data grow in size. Besides the tensors, in many real world scenarios, side information is also available in the form of matrices which also grow in size with time. The problem of predicting missing values in the dynamically growing tensor is called dynamic tensor completion. Most of the previous work in dynamic tensor completion make an assumption that the tensor grows only in one mode. To the best of our Knowledge, there is no previous work which incorporates side information with dynamic tensor completion. We bridge this gap in this paper by proposing a dynamic tensor completion framework called Side Information infused Incremental Tensor Analysis (SIITA), which incorporates side information and works for general incremental tensors. We also show how non-negative constraints can be incorporated with SIITA, which is essential for mining interpretable latent clusters. We carry out extensive experiments on multiple real world datasets to demonstrate the effectiveness of SIITA in various different settings.
Tasks
Published	2018-02-18
URL	http://arxiv.org/abs/1802.06371v3
PDF	http://arxiv.org/pdf/1802.06371v3.pdf
PWC	https://paperswithcode.com/paper/inductive-framework-for-multi-aspect
Repo
Framework

Investigating Enactive Learning for Autonomous Intelligent Agents


Title	Investigating Enactive Learning for Autonomous Intelligent Agents
Authors	Rafik Hadfi
Abstract	The enactive approach to cognition is typically proposed as a viable alternative to traditional cognitive science. Enactive cognition displaces the explanatory focus from the internal representations of the agent to the direct sensorimotor interaction with its environment. In this paper, we investigate enactive learning through means of artificial agent simulations. We compare the performances of the enactive agent to an agent operating on classical reinforcement learning in foraging tasks within maze environments. The characteristics of the agents are analysed in terms of the accessibility of the environmental states, goals, and exploration/exploitation tradeoffs. We confirm that the enactive agent can successfully interact with its environment and learn to avoid unfavourable interactions using intrinsically defined goals. The performance of the enactive agent is shown to be limited by the number of affordable actions.
Tasks
Published	2018-10-09
URL	http://arxiv.org/abs/1810.04535v1
PDF	http://arxiv.org/pdf/1810.04535v1.pdf
PWC	https://paperswithcode.com/paper/investigating-enactive-learning-for
Repo
Framework

Let’s do it “again”: A First Computational Approach to Detecting Adverbial Presupposition Triggers


Title	Let’s do it “again”: A First Computational Approach to Detecting Adverbial Presupposition Triggers
Authors	Andre Cianflone, Yulan Feng, Jad Kabbara, Jackie Chi Kit Cheung
Abstract	We introduce the task of predicting adverbial presupposition triggers such as also and again. Solving such a task requires detecting recurring or similar events in the discourse context, and has applications in natural language generation tasks such as summarization and dialogue systems. We create two new datasets for the task, derived from the Penn Treebank and the Annotated English Gigaword corpora, as well as a novel attention mechanism tailored to this task. Our attention mechanism augments a baseline recurrent neural network without the need for additional trainable parameters, minimizing the added computational cost of our mechanism. We demonstrate that our model statistically outperforms a number of baselines, including an LSTM-based language model.
Tasks	Language Modelling, Text Generation
Published	2018-06-11
URL	http://arxiv.org/abs/1806.04262v1
PDF	http://arxiv.org/pdf/1806.04262v1.pdf
PWC	https://paperswithcode.com/paper/lets-do-it-again-a-first-computational
Repo
Framework

On the Effect of Task-to-Worker Assignment in Distributed Computing Systems with Stragglers


Title	On the Effect of Task-to-Worker Assignment in Distributed Computing Systems with Stragglers
Authors	Amir Behrouzi-Far, Emina Soljanin
Abstract	We study the expected completion time of some recently proposed algorithms for distributed computing which redundantly assign computing tasks to multiple machines in order to tolerate a certain number of machine failures. We analytically show that not only the amount of redundancy but also the task-to-machine assignments affect the latency in a distributed system. We study systems with a fixed number of computing tasks that are split in possibly overlapping batches, and independent exponentially distributed machine service times. We show that, for such systems, the uniform replication of non- overlapping (disjoint) batches of computing tasks achieves the minimum expected computing time.
Tasks
Published	2018-08-08
URL	http://arxiv.org/abs/1808.02838v1
PDF	http://arxiv.org/pdf/1808.02838v1.pdf
PWC	https://paperswithcode.com/paper/on-the-effect-of-task-to-worker-assignment-in
Repo
Framework

Deep learning and its application to medical image segmentation


Title	Deep learning and its application to medical image segmentation
Authors	Holger R. Roth, Chen Shen, Hirohisa Oda, Masahiro Oda, Yuichiro Hayashi, Kazunari Misawa, Kensaku Mori
Abstract	One of the most common tasks in medical imaging is semantic segmentation. Achieving this segmentation automatically has been an active area of research, but the task has been proven very challenging due to the large variation of anatomy across different patients. However, recent advances in deep learning have made it possible to significantly improve the performance of image recognition and semantic segmentation methods in the field of computer vision. Due to the data driven approaches of hierarchical feature learning in deep learning frameworks, these advances can be translated to medical images without much difficulty. Several variations of deep convolutional neural networks have been successfully applied to medical images. Especially fully convolutional architectures have been proven efficient for segmentation of 3D medical images. In this article, we describe how to build a 3D fully convolutional network (FCN) that can process 3D images in order to produce automatic semantic segmentations. The model is trained and evaluated on a clinical computed tomography (CT) dataset and shows state-of-the-art performance in multi-organ segmentation.
Tasks	Computed Tomography (CT), Medical Image Segmentation, Semantic Segmentation
Published	2018-03-23
URL	http://arxiv.org/abs/1803.08691v1
PDF	http://arxiv.org/pdf/1803.08691v1.pdf
PWC	https://paperswithcode.com/paper/deep-learning-and-its-application-to-medical
Repo
Framework