July 28, 2019

3332 words 16 mins read

Paper Group ANR 198

MATIC: Learning Around Errors for Efficient Low-Voltage Neural Network Accelerators. Deep Text Classification Can be Fooled. Hi, how can I help you?: Automating enterprise IT support help desks. Unsupervised Joint Mining of Deep Features and Image Labels for Large-scale Radiology Image Categorization and Scene Recognition. Modelling Energy Consumpt …

MATIC: Learning Around Errors for Efficient Low-Voltage Neural Network Accelerators


Title	MATIC: Learning Around Errors for Efficient Low-Voltage Neural Network Accelerators
Authors	Sung Kim, Patrick Howe, Thierry Moreau, Armin Alaghi, Luis Ceze, Visvesh Sathe
Abstract	As a result of the increasing demand for deep neural network (DNN)-based services, efforts to develop dedicated hardware accelerators for DNNs are growing rapidly. However,while accelerators with high performance and efficiency on convolutional deep neural networks (Conv-DNNs) have been developed, less progress has been made with regards to fully-connected DNNs (FC-DNNs). In this paper, we propose MATIC (Memory Adaptive Training with In-situ Canaries), a methodology that enables aggressive voltage scaling of accelerator weight memories to improve the energy-efficiency of DNN accelerators. To enable accurate operation with voltage overscaling, MATIC combines the characteristics of destructive SRAM reads with the error resilience of neural networks in a memory-adaptive training process. Furthermore, PVT-related voltage margins are eliminated using bit-cells from synaptic weights as in-situ canaries to track runtime environmental variation. Demonstrated on a low-power DNN accelerator that we fabricate in 65 nm CMOS, MATIC enables up to 60-80 mV of voltage overscaling (3.3x total energy reduction versus the nominal voltage), or 18.6x application error reduction.
Tasks
Published	2017-06-14
URL	http://arxiv.org/abs/1706.04332v3
PDF	http://arxiv.org/pdf/1706.04332v3.pdf
PWC	https://paperswithcode.com/paper/matic-learning-around-errors-for-efficient
Repo
Framework

Deep Text Classification Can be Fooled


Title	Deep Text Classification Can be Fooled
Authors	Bin Liang, Hongcheng Li, Miaoqiang Su, Pan Bian, Xirong Li, Wenchang Shi
Abstract	In this paper, we present an effective method to craft text adversarial samples, revealing one important yet underestimated fact that DNN-based text classifiers are also prone to adversarial sample attack. Specifically, confronted with different adversarial scenarios, the text items that are important for classification are identified by computing the cost gradients of the input (white-box attack) or generating a series of occluded test samples (black-box attack). Based on these items, we design three perturbation strategies, namely insertion, modification, and removal, to generate adversarial samples. The experiment results show that the adversarial samples generated by our method can successfully fool both state-of-the-art character-level and word-level DNN-based text classifiers. The adversarial samples can be perturbed to any desirable classes without compromising their utilities. At the same time, the introduced perturbation is difficult to be perceived.
Tasks	Text Classification
Published	2017-04-26
URL	http://arxiv.org/abs/1704.08006v2
PDF	http://arxiv.org/pdf/1704.08006v2.pdf
PWC	https://paperswithcode.com/paper/deep-text-classification-can-be-fooled
Repo
Framework

Hi, how can I help you?: Automating enterprise IT support help desks


Title	Hi, how can I help you?: Automating enterprise IT support help desks
Authors	Senthil Mani, Neelamadhav Gantayat, Rahul Aralikatte, Monika Gupta, Sampath Dechu, Anush Sankaran, Shreya Khare, Barry Mitchell, Hemamalini Subramanian, Hema Venkatarangan
Abstract	Question answering is one of the primary challenges of natural language understanding. In realizing such a system, providing complex long answers to questions is a challenging task as opposed to factoid answering as the former needs context disambiguation. The different methods explored in the literature can be broadly classified into three categories namely: 1) classification based, 2) knowledge graph based and 3) retrieval based. Individually, none of them address the need of an enterprise wide assistance system for an IT support and maintenance domain. In this domain the variance of answers is large ranging from factoid to structured operating procedures; the knowledge is present across heterogeneous data sources like application specific documentation, ticket management systems and any single technique for a general purpose assistance is unable to scale for such a landscape. To address this, we have built a cognitive platform with capabilities adopted for this domain. Further, we have built a general purpose question answering system leveraging the platform that can be instantiated for multiple products, technologies in the support domain. The system uses a novel hybrid answering model that orchestrates across a deep learning classifier, a knowledge graph based context disambiguation module and a sophisticated bag-of-words search system. This orchestration performs context switching for a provided question and also does a smooth hand-off of the question to a human expert if none of the automated techniques can provide a confident answer. This system has been deployed across 675 internal enterprise IT support and maintenance projects.
Tasks	Question Answering
Published	2017-11-02
URL	http://arxiv.org/abs/1711.02012v1
PDF	http://arxiv.org/pdf/1711.02012v1.pdf
PWC	https://paperswithcode.com/paper/hi-how-can-i-help-you-automating-enterprise
Repo
Framework

Unsupervised Joint Mining of Deep Features and Image Labels for Large-scale Radiology Image Categorization and Scene Recognition


Title	Unsupervised Joint Mining of Deep Features and Image Labels for Large-scale Radiology Image Categorization and Scene Recognition
Authors	Xiaosong Wang, Le Lu, Hoo-chang Shin, Lauren Kim, Mohammadhadi Bagheri, Isabella Nogues, Jianhua Yao, Ronald M. Summers
Abstract	The recent rapid and tremendous success of deep convolutional neural networks (CNN) on many challenging computer vision tasks largely derives from the accessibility of the well-annotated ImageNet and PASCAL VOC datasets. Nevertheless, unsupervised image categorization (i.e., without the ground-truth labeling) is much less investigated, yet critically important and difficult when annotations are extremely hard to obtain in the conventional way of “Google Search” and crowd sourcing. We address this problem by presenting a looped deep pseudo-task optimization (LDPO) framework for joint mining of deep CNN features and image labels. Our method is conceptually simple and rests upon the hypothesized “convergence” of better labels leading to better trained CNN models which in turn feed more discriminative image representations to facilitate more meaningful clusters/labels. Our proposed method is validated in tackling two important applications: 1) Large-scale medical image annotation has always been a prohibitively expensive and easily-biased task even for well-trained radiologists. Significantly better image categorization results are achieved via our proposed approach compared to the previous state-of-the-art method. 2) Unsupervised scene recognition on representative and publicly available datasets with our proposed technique is examined. The LDPO achieves excellent quantitative scene classification results. On the MIT indoor scene dataset, it attains a clustering accuracy of 75.3%, compared to the state-of-the-art supervised classification accuracy of 81.0% (when both are based on the VGG-VD model).
Tasks	Image Categorization, Scene Classification, Scene Recognition
Published	2017-01-23
URL	http://arxiv.org/abs/1701.06599v2
PDF	http://arxiv.org/pdf/1701.06599v2.pdf
PWC	https://paperswithcode.com/paper/unsupervised-joint-mining-of-deep-features
Repo
Framework

Modelling Energy Consumption based on Resource Utilization


Title	Modelling Energy Consumption based on Resource Utilization
Authors	Lucas Venezian Povoa, Cesar Marcondes, Hermes Senger
Abstract	Power management is an expensive and important issue for large computational infrastructures such as datacenters, large clusters, and computational grids. However, measuring energy consumption of scalable systems may be impractical due to both cost and complexity for deploying power metering devices on a large number of machines. In this paper, we propose the use of information about resource utilization (e.g. processor, memory, disk operations, and network traffic) as proxies for estimating power consumption. We employ machine learning techniques to estimate power consumption using such information which are provided by common operating systems. Experiments with linear regression, regression tree, and multilayer perceptron on data from different hardware resulted into a model with 99.94% of accuracy and 6.32 watts of error in the best case.
Tasks
Published	2017-09-15
URL	http://arxiv.org/abs/1709.06076v1
PDF	http://arxiv.org/pdf/1709.06076v1.pdf
PWC	https://paperswithcode.com/paper/modelling-energy-consumption-based-on
Repo
Framework

Intraoperative Organ Motion Models with an Ensemble of Conditional Generative Adversarial Networks


Title	Intraoperative Organ Motion Models with an Ensemble of Conditional Generative Adversarial Networks
Authors	Yipeng Hu, Eli Gibson, Tom Vercauteren, Hashim U. Ahmed, Mark Emberton, Caroline M. Moore, J. Alison Noble, Dean C. Barratt
Abstract	In this paper, we describe how a patient-specific, ultrasound-probe-induced prostate motion model can be directly generated from a single preoperative MR image. Our motion model allows for sampling from the conditional distribution of dense displacement fields, is encoded by a generative neural network conditioned on a medical image, and accepts random noise as additional input. The generative network is trained by a minimax optimisation with a second discriminative neural network, tasked to distinguish generated samples from training motion data. In this work, we propose that 1) jointly optimising a third conditioning neural network that pre-processes the input image, can effectively extract patient-specific features for conditioning; and 2) combining multiple generative models trained separately with heuristically pre-disjointed training data sets can adequately mitigate the problem of mode collapse. Trained with diagnostic T2-weighted MR images from 143 real patients and 73,216 3D dense displacement fields from finite element simulations of intraoperative prostate motion due to transrectal ultrasound probe pressure, the proposed models produced physically-plausible patient-specific motion of prostate glands. The ability to capture biomechanically simulated motion was evaluated using two errors representing generalisability and specificity of the model. The median values, calculated from a 10-fold cross-validation, were 2.8+/-0.3 mm and 1.7+/-0.1 mm, respectively. We conclude that the introduced approach demonstrates the feasibility of applying state-of-the-art machine learning algorithms to generate organ motion models from patient images, and shows significant promise for future research.
Tasks
Published	2017-09-05
URL	http://arxiv.org/abs/1709.02255v1
PDF	http://arxiv.org/pdf/1709.02255v1.pdf
PWC	https://paperswithcode.com/paper/intraoperative-organ-motion-models-with-an
Repo
Framework

The quality of priority ratios estimation in relation to a selected prioritization procedure and consistency measure for a Pairwise Comparison Matrix


Title	The quality of priority ratios estimation in relation to a selected prioritization procedure and consistency measure for a Pairwise Comparison Matrix
Authors	Paul Thaddeus Kazibudzki
Abstract	An overview of current debates and contemporary research devoted to the modeling of decision making processes and their facilitation directs attention to the Analytic Hierarchy Process (AHP). At the core of the AHP are various prioritization procedures (PPs) and consistency measures (CMs) for a Pairwise Comparison Matrix (PCM) which, in a sense, reflects preferences of decision makers. Certainly, when judgments about these preferences are perfectly consistent (cardinally transitive), all PPs coincide and the quality of the priority ratios (PRs) estimation is exemplary. However, human judgments are very rarely consistent, thus the quality of PRs estimation may significantly vary. The scale of these variations depends on the applied PP and utilized CM for a PCM. This is why it is important to find out which PPs and which CMs for a PCM lead directly to an improvement of the PRs estimation accuracy. The main goal of this research is realized through the properly designed, coded and executed seminal and sophisticated simulation algorithms in Wolfram Mathematica 8.0. These research results convince that the embedded in the AHP and commonly applied, both genuine PP and CM for PCM may significantly deteriorate the quality of PRs estimation; however, solutions proposed in this paper can significantly improve the methodology.
Tasks	Decision Making
Published	2017-04-06
URL	http://arxiv.org/abs/1704.01944v2
PDF	http://arxiv.org/pdf/1704.01944v2.pdf
PWC	https://paperswithcode.com/paper/the-quality-of-priority-ratios-estimation-in
Repo
Framework

On the Evaluation of Video Keyframe Summaries using User Ground Truth


Title	On the Evaluation of Video Keyframe Summaries using User Ground Truth
Authors	Ludmila I. Kuncheva, Paria Yousefi, Iain A. D. Gunn
Abstract	Given the great interest in creating keyframe summaries from video, it is surprising how little has been done to formalise their evaluation and comparison. User studies are often carried out to demonstrate that a proposed method generates a more appealing summary than one or two rival methods. But larger comparison studies cannot feasibly use such user surveys. Here we propose a discrimination capacity measure as a formal way to quantify the improvement over the uniform baseline, assuming that one or more ground truth summaries are available. Using the VSUMM video collection, we examine 10 video feature types, including CNN and SURF, and 6 methods for matching frames from two summaries. Our results indicate that a simple frame representation through hue histograms suffices for the purposes of comparing keyframe summaries. We subsequently propose a formal protocol for comparing summaries when ground truth is available.
Tasks
Published	2017-12-19
URL	http://arxiv.org/abs/1712.06899v1
PDF	http://arxiv.org/pdf/1712.06899v1.pdf
PWC	https://paperswithcode.com/paper/on-the-evaluation-of-video-keyframe-summaries
Repo
Framework

Indoor UAV scheduling with Restful Task Assignment Algorithm


Title	Indoor UAV scheduling with Restful Task Assignment Algorithm
Authors	Yohanes Khosiawan, Izabela Nielsen
Abstract	Research in UAV scheduling has obtained an emerging interest from scientists in the optimization field. When the scheduling itself has established a strong root since the 19th century, works on UAV scheduling in indoor environment has come forth in the latest decade. Several works on scheduling UAV operations in indoor (two and three dimensional) and outdoor environments are reported. In this paper, a further study on UAV scheduling in three dimensional indoor environment is investigated. Dealing with indoor environment\textemdash where humans, UAVs, and other elements or infrastructures are likely to coexist in the same space\textemdash draws attention towards the safety of the operations. In relation to the battery level, a preserved battery level leads to safer operations, promoting the UAV to have a decent remaining power level. A methodology which consists of a heuristic approach based on Restful Task Assignment Algorithm, incorporated with Particle Swarm Optimization Algorithm, is proposed. The motivation is to preserve the battery level throughout the operations, which promotes less possibility in having failed UAVs on duty. This methodology is tested with 54 benchmark datasets stressing on 4 different aspects: geographical distance, number of tasks, number of predecessors, and slack time. The test results and their characteristics in regard to the proposed methodology are discussed and presented.
Tasks
Published	2017-06-29
URL	http://arxiv.org/abs/1706.09737v1
PDF	http://arxiv.org/pdf/1706.09737v1.pdf
PWC	https://paperswithcode.com/paper/indoor-uav-scheduling-with-restful-task
Repo
Framework

A-Lamp: Adaptive Layout-Aware Multi-Patch Deep Convolutional Neural Network for Photo Aesthetic Assessment


Title	A-Lamp: Adaptive Layout-Aware Multi-Patch Deep Convolutional Neural Network for Photo Aesthetic Assessment
Authors	Shuang Ma, Jing Liu, Chang Wen Chen
Abstract	Deep convolutional neural networks (CNN) have recently been shown to generate promising results for aesthetics assessment. However, the performance of these deep CNN methods is often compromised by the constraint that the neural network only takes the fixed-size input. To accommodate this requirement, input images need to be transformed via cropping, warping, or padding, which often alter image composition, reduce image resolution, or cause image distortion. Thus the aesthetics of the original images is impaired because of potential loss of fine grained details and holistic image layout. However, such fine grained details and holistic image layout is critical for evaluating an image’s aesthetics. In this paper, we present an Adaptive Layout-Aware Multi-Patch Convolutional Neural Network (A-Lamp CNN) architecture for photo aesthetic assessment. This novel scheme is able to accept arbitrary sized images, and learn from both fined grained details and holistic image layout simultaneously. To enable training on these hybrid inputs, we extend the method by developing a dedicated double-subnet neural network structure, i.e. a Multi-Patch subnet and a Layout-Aware subnet. We further construct an aggregation layer to effectively combine the hybrid features from these two subnets. Extensive experiments on the large-scale aesthetics assessment benchmark (AVA) demonstrate significant performance improvement over the state-of-the-art in photo aesthetic assessment.
Tasks	Aesthetics Quality Assessment
Published	2017-04-02
URL	http://arxiv.org/abs/1704.00248v1
PDF	http://arxiv.org/pdf/1704.00248v1.pdf
PWC	https://paperswithcode.com/paper/a-lamp-adaptive-layout-aware-multi-patch-deep
Repo
Framework

Understanding the Changing Roles of Scientific Publications via Citation Embeddings


Title	Understanding the Changing Roles of Scientific Publications via Citation Embeddings
Authors	Jiangen He, Chaomei Chen
Abstract	Researchers may describe different aspects of past scientific publications in their publications and the descriptions may keep changing in the evolution of science. The diverse and changing descriptions (i.e., citation context) on a publication characterize the impact and contributions of the past publication. In this article, we aim to provide an approach to understanding the changing and complex roles of a publication characterized by its citation context. We described a method to represent the publications’ dynamic roles in science community in different periods as a sequence of vectors by training temporal embedding models. The temporal representations can be used to quantify how much the roles of publications changed and interpret how they changed. Our study in the biomedical domain shows that our metric on the changes of publications’ roles is stable over time at the population level but significantly distinguish individuals. We also show the interpretability of our methods by a concrete example.
Tasks
Published	2017-11-15
URL	http://arxiv.org/abs/1711.05822v1
PDF	http://arxiv.org/pdf/1711.05822v1.pdf
PWC	https://paperswithcode.com/paper/understanding-the-changing-roles-of
Repo
Framework

LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task


Title	LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task
Authors	Alexandre Berard, Olivier Pietquin, Laurent Besacier
Abstract	This paper presents the LIG-CRIStAL submission to the shared Automatic Post- Editing task of WMT 2017. We propose two neural post-editing models: a monosource model with a task-specific attention mechanism, which performs particularly well in a low-resource scenario; and a chained architecture which makes use of the source sentence to provide extra context. This latter architecture manages to slightly improve our results when more training data is available. We present and discuss our results on two datasets (en-de and de-en) that are made available for the task.
Tasks	Automatic Post-Editing
Published	2017-07-17
URL	http://arxiv.org/abs/1707.05118v1
PDF	http://arxiv.org/pdf/1707.05118v1.pdf
PWC	https://paperswithcode.com/paper/lig-cristal-system-for-the-wmt17-automatic
Repo
Framework

Zero-Shot Learning to Manage a Large Number of Place-Specific Compressive Change Classifiers


Title	Zero-Shot Learning to Manage a Large Number of Place-Specific Compressive Change Classifiers
Authors	Tanaka Kanji
Abstract	With recent progress in large-scale map maintenance and long-term map learning, the task of change detection on a large-scale map from a visual image captured by a mobile robot has become a problem of increasing criticality. Previous approaches for change detection are typically based on image differencing and require the memorization of a prohibitively large number of mapped images in the above context. In contrast, this study follows the recent, efficient paradigm of change-classifier-learning and specifically employs a collection of place-specific change classifiers. Our change-classifier-learning algorithm is based on zero-shot learning (ZSL) and represents a place-specific change classifier by its training examples mined from an external knowledge base (EKB). The proposed algorithm exhibits several advantages. First, we are required to memorize only training examples (rather than the classifier itself), which can be further compressed in the form of bag-of-words (BoW). Secondly, we can incorporate the most recent map into the classifiers by straightforwardly adding or deleting a few training examples that correspond to these classifiers. Thirdly, we can share the BoW vocabulary with other related task scenarios (e.g., BoW-based self-localization), wherein the vocabulary is generally designed as a rich, continuously growing, and domain-adaptive knowledge base. In our contribution, the proposed algorithm is applied and evaluated on a practical long-term cross-season change detection system that consists of a large number of place-specific object-level change classifiers.
Tasks	Zero-Shot Learning
Published	2017-09-15
URL	http://arxiv.org/abs/1709.05397v1
PDF	http://arxiv.org/pdf/1709.05397v1.pdf
PWC	https://paperswithcode.com/paper/zero-shot-learning-to-manage-a-large-number
Repo
Framework

Low-Precision Batch-Normalized Activations


Title	Low-Precision Batch-Normalized Activations
Authors	Benjamin Graham
Abstract	Artificial neural networks can be trained with relatively low-precision floating-point and fixed-point arithmetic, using between one and 16 bits. Previous works have focused on relatively wide-but-shallow, feed-forward networks. We introduce a quantization scheme that is compatible with training very deep neural networks. Quantizing the network activations in the middle of each batch-normalization module can greatly reduce the amount of memory and computational power needed, with little loss in accuracy.
Tasks	Quantization
Published	2017-02-27
URL	http://arxiv.org/abs/1702.08231v1
PDF	http://arxiv.org/pdf/1702.08231v1.pdf
PWC	https://paperswithcode.com/paper/low-precision-batch-normalized-activations
Repo
Framework

Recurrent Topic-Transition GAN for Visual Paragraph Generation


Title	Recurrent Topic-Transition GAN for Visual Paragraph Generation
Authors	Xiaodan Liang, Zhiting Hu, Hao Zhang, Chuang Gan, Eric P. Xing
Abstract	A natural image usually conveys rich semantic content and can be viewed from different angles. Existing image description methods are largely restricted by small sets of biased visual paragraph annotations, and fail to cover rich underlying semantics. In this paper, we investigate a semi-supervised paragraph generative framework that is able to synthesize diverse and semantically coherent paragraph descriptions by reasoning over local semantic regions and exploiting linguistic knowledge. The proposed Recurrent Topic-Transition Generative Adversarial Network (RTT-GAN) builds an adversarial framework between a structured paragraph generator and multi-level paragraph discriminators. The paragraph generator generates sentences recurrently by incorporating region-based visual and language attention mechanisms at each step. The quality of generated paragraph sentences is assessed by multi-level adversarial discriminators from two aspects, namely, plausibility at sentence level and topic-transition coherence at paragraph level. The joint adversarial training of RTT-GAN drives the model to generate realistic paragraphs with smooth logical transition between sentence topics. Extensive quantitative experiments on image and video paragraph datasets demonstrate the effectiveness of our RTT-GAN in both supervised and semi-supervised settings. Qualitative results on telling diverse stories for an image also verify the interpretability of RTT-GAN.
Tasks	Image Paragraph Captioning
Published	2017-03-21
URL	http://arxiv.org/abs/1703.07022v2
PDF	http://arxiv.org/pdf/1703.07022v2.pdf
PWC	https://paperswithcode.com/paper/recurrent-topic-transition-gan-for-visual
Repo
Framework