April 3, 2020

3442 words 17 mins read

Paper Group AWR 33

Attractive or Faithful? Popularity-Reinforced Learning for Inspired Headline Generation. Lane Detection in Low-light Conditions Using an Efficient Data Enhancement : Light Conditions Style Transfer. Subclass Distillation. Exploring and Improving Robustness of Multi Task Deep Neural Networks via Domain Agnostic Defenses. Relaxing Relationship Querie …

Attractive or Faithful? Popularity-Reinforced Learning for Inspired Headline Generation


Title	Attractive or Faithful? Popularity-Reinforced Learning for Inspired Headline Generation
Authors	Yun-Zhu Song, Hong-Han Shuai, Sung-Lin Yeh, Yi-Lun Wu, Lun-Wei Ku, Wen-Chih Peng
Abstract	With the rapid proliferation of online media sources and published news, headlines have become increasingly important for attracting readers to news articles, since users may be overwhelmed with the massive information. In this paper, we generate inspired headlines that preserve the nature of news articles and catch the eye of the reader simultaneously. The task of inspired headline generation can be viewed as a specific form of Headline Generation (HG) task, with the emphasis on creating an attractive headline from a given news article. To generate inspired headlines, we propose a novel framework called POpularity-Reinforced Learning for inspired Headline Generation (PORL-HG). PORL-HG exploits the extractive-abstractive architecture with 1) Popular Topic Attention (PTA) for guiding the extractor to select the attractive sentence from the article and 2) a popularity predictor for guiding the abstractor to rewrite the attractive sentence. Moreover, since the sentence selection of the extractor is not differentiable, techniques of reinforcement learning (RL) are utilized to bridge the gap with rewards obtained from a popularity score predictor. Through quantitative and qualitative experiments, we show that the proposed PORL-HG significantly outperforms the state-of-the-art headline generation models in terms of attractiveness evaluated by both human (71.03%) and the predictor (at least 27.60%), while the faithfulness of PORL-HG is also comparable to the state-of-the-art generation model.
Tasks
Published	2020-02-06
URL	https://arxiv.org/abs/2002.02095v1
PDF	https://arxiv.org/pdf/2002.02095v1.pdf
PWC	https://paperswithcode.com/paper/attractive-or-faithful-popularity-reinforced
Repo	https://github.com/yunzhusong/AAAI20-PORLHG
Framework	none

Lane Detection in Low-light Conditions Using an Efficient Data Enhancement : Light Conditions Style Transfer


Title	Lane Detection in Low-light Conditions Using an Efficient Data Enhancement : Light Conditions Style Transfer
Authors	Tong Liu, Zhaowei Chen, Yi Yang, Zehao Wu, Haowei Li
Abstract	Nowadays, deep learning techniques are widely used for lane detection, but application in low-light conditions remains a challenge until this day. Although multi-task learning and contextual information based methods have been proposed to solve the problem, they either require additional manual annotations or introduce extra inference computation respectively. In this paper, we propose a style-transfer-based data enhancement method, which uses Generative Adversarial Networks (GANs) to generate images in low-light conditions, that increases the environmental adaptability of the lane detector. Our solution consists of three models: the proposed Better-CycleGAN, light conditions style transfer network and lane detection network. It does not require additional manual annotations nor extra inference computation. We validated our methods on the lane detection benchmark CULane using ERFNet. Empirically, lane detection model trained using our method demonstrated adaptability in low-light conditions and robustness in complex scenarios. Our code for this paper will be publicly available.
Tasks	Lane Detection, Multi-Task Learning, Style Transfer
Published	2020-02-04
URL	https://arxiv.org/abs/2002.01177v1
PDF	https://arxiv.org/pdf/2002.01177v1.pdf
PWC	https://paperswithcode.com/paper/lane-detection-in-low-light-conditions-using
Repo	https://github.com/Chenzhaowei13/Light-Condition-Style-Transfer
Framework	pytorch

Subclass Distillation


Title	Subclass Distillation
Authors	Rafael Müller, Simon Kornblith, Geoffrey Hinton
Abstract	After a large “teacher” neural network has been trained on labeled data, the probabilities that the teacher assigns to incorrect classes reveal a lot of information about the way in which the teacher generalizes. By training a small “student” model to match these probabilities, it is possible to transfer most of the generalization ability of the teacher to the student, often producing a much better small model than directly training the student on the training data. The transfer works best when there are many possible classes because more is then revealed about the function learned by the teacher, but in cases where there are only a few possible classes we show that we can improve the transfer by forcing the teacher to divide each class into many subclasses that it invents during the supervised training. The student is then trained to match the subclass probabilities. For datasets where there are known, natural subclasses we demonstrate that the teacher learns similar subclasses and these improve distillation. For clickthrough datasets where the subclasses are unknown we demonstrate that subclass distillation allows the student to learn faster and better.
Tasks
Published	2020-02-10
URL	https://arxiv.org/abs/2002.03936v1
PDF	https://arxiv.org/pdf/2002.03936v1.pdf
PWC	https://paperswithcode.com/paper/subclass-distillation
Repo	https://github.com/pvgladkov/knowledge-distillation
Framework	pytorch

Exploring and Improving Robustness of Multi Task Deep Neural Networks via Domain Agnostic Defenses


Title	Exploring and Improving Robustness of Multi Task Deep Neural Networks via Domain Agnostic Defenses
Authors	Kashyap Coimbatore Murali
Abstract	In this paper, we explore the robustness of the Multi-Task Deep Neural Networks (MT-DNN) against non-targeted adversarial attacks across Natural Language Understanding (NLU) tasks as well as some possible ways to defend against them. Liu et al., have shown that the Multi-Task Deep Neural Network, due to the regularization effect produced when training as a result of its cross task data, is more robust than a vanilla BERT model trained only on one task (1.1%-1.5% absolute difference). We further show that although the MT-DNN has generalized better, making it easily transferable across domains and tasks, it can still be compromised as after only 2 attacks (1-character and 2-character) the accuracy drops by 42.05% and 32.24% for the SNLI and SciTail tasks. Finally, we propose a domain agnostic defense which restores the model’s accuracy (36.75% and 25.94% respectively) as opposed to a general-purpose defense or an off-the-shelf spell checker.
Tasks
Published	2020-01-11
URL	https://arxiv.org/abs/2001.05286v1
PDF	https://arxiv.org/pdf/2001.05286v1.pdf
PWC	https://paperswithcode.com/paper/exploring-and-improving-robustness-of-multi
Repo	https://github.com/katchu11/Robustness-of-MT-DNNs
Framework	pytorch

Relaxing Relationship Queries on Graph Data


Title	Relaxing Relationship Queries on Graph Data
Authors	Shuxin Li, Gong Cheng, Chengkai Li
Abstract	In many domains we have witnessed the need to search a large entity-relation graph for direct and indirect relationships between a set of entities specified in a query. A search result, called a semantic association (SA), is typically a compact (e.g., diameter-constrained) connected subgraph containing all the query entities. For this problem of SA search, efficient algorithms exist but will return empty results if some query entities are distant in the graph. To reduce the occurrence of failing query and provide alternative results, we study the problem of query relaxation in the context of SA search. Simply relaxing the compactness constraint will sacrifice the compactness of an SA, and more importantly, may lead to performance issues and be impracticable. Instead, we focus on removing the smallest number of entities from the original failing query, to form a maximum successful sub-query which minimizes the loss of result quality caused by relaxation. We prove that verifying the success of a sub-query turns into finding an entity (called a certificate) that satisfies a distance-based condition about the query entities. To efficiently find a certificate of the success of a maximum sub-query, we propose a best-first search algorithm that leverages distance-based estimation to effectively prune the search space. We further improve its performance by adding two fine-grained heuristics: one based on degree and the other based on distance. Extensive experiments over popular RDF datasets demonstrate the efficiency of our algorithm, which is more scalable than baselines.
Tasks
Published	2020-02-24
URL	https://arxiv.org/abs/2002.10181v1
PDF	https://arxiv.org/pdf/2002.10181v1.pdf
PWC	https://paperswithcode.com/paper/relaxing-relationship-queries-on-graph-data
Repo	https://github.com/nju-websoft/CertQR
Framework	none

Gated Fusion Network for Degraded Image Super Resolution


Title	Gated Fusion Network for Degraded Image Super Resolution
Authors	Xinyi Zhang, Hang Dong, Zhe Hu, Wei-Sheng Lai, Fei Wang, Ming-Hsuan Yang
Abstract	Single image super resolution aims to enhance image quality with respect to spatial content, which is a fundamental task in computer vision. In this work, we address the task of single frame super resolution with the presence of image degradation, e.g., blur, haze, or rain streaks. Due to the limitations of frame capturing and formation processes, image degradation is inevitable, and the artifacts would be exacerbated by super resolution methods. To address this problem, we propose a dual-branch convolutional neural network to extract base features and recovered features separately. The base features contain local and global information of the input image. On the other hand, the recovered features focus on the degraded regions and are used to remove the degradation. Those features are then fused through a recursive gate module to obtain sharp features for super resolution. By decomposing the feature extraction step into two task-independent streams, the dual-branch model can facilitate the training process by avoiding learning the mixed degradation all-in-one and thus enhance the final high-resolution prediction results. We evaluate the proposed method in three degradation scenarios. Experiments on these scenarios demonstrate that the proposed method performs more efficiently and favorably against the state-of-the-art approaches on benchmark datasets.
Tasks	Image Super-Resolution, Super-Resolution
Published	2020-03-02
URL	https://arxiv.org/abs/2003.00893v2
PDF	https://arxiv.org/pdf/2003.00893v2.pdf
PWC	https://paperswithcode.com/paper/gated-fusion-network-for-degraded-image-super
Repo	https://github.com/BookerDeWitt/GFN-IJCV
Framework	pytorch

DDet: Dual-path Dynamic Enhancement Network for Real-World Image Super-Resolution


Title	DDet: Dual-path Dynamic Enhancement Network for Real-World Image Super-Resolution
Authors	Yukai Shi, Haoyu Zhong, Zhijing Yang, Xiaojun Yang, Liang Lin
Abstract	Different from traditional image super-resolution task, real image super-resolution(Real-SR) focus on the relationship between real-world high-resolution(HR) and low-resolution(LR) image. Most of the traditional image SR obtains the LR sample by applying a fixed down-sampling operator. Real-SR obtains the LR and HR image pair by incorporating different quality optical sensors. Generally, Real-SR has more challenges as well as broader application scenarios. Previous image SR methods fail to exhibit similar performance on Real-SR as the image data is not aligned inherently. In this article, we propose a Dual-path Dynamic Enhancement Network(DDet) for Real-SR, which addresses the cross-camera image mapping by realizing a dual-way dynamic sub-pixel weighted aggregation and refinement. Unlike conventional methods which stack up massive convolutional blocks for feature representation, we introduce a content-aware framework to study non-inherently aligned image pair in image SR issue. First, we use a content-adaptive component to exhibit the Multi-scale Dynamic Attention(MDA). Second, we incorporate a long-term skip connection with a Coupled Detail Manipulation(CDM) to perform collaborative compensation and manipulation. The above dual-path model is joint into a unified model and works collaboratively. Extensive experiments on the challenging benchmarks demonstrate the superiority of our model.
Tasks	Image Super-Resolution, Super-Resolution
Published	2020-02-25
URL	https://arxiv.org/abs/2002.11079v1
PDF	https://arxiv.org/pdf/2002.11079v1.pdf
PWC	https://paperswithcode.com/paper/ddet-dual-path-dynamic-enhancement-network
Repo	https://github.com/ykshi/DDet
Framework	none

PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling


Title	PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
Authors	Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui
Abstract	Raw point clouds data inevitably contains outliers or noise through acquisition from 3D sensors or reconstruction algorithms. In this paper, we present a novel end-to-end network for robust point clouds processing, named PointASNL, which can deal with point clouds with noise effectively. The key component in our approach is the adaptive sampling (AS) module. It first re-weights the neighbors around the initial sampled points from farthest point sampling (FPS), and then adaptively adjusts the sampled points beyond the entire point cloud. Our AS module can not only benefit the feature learning of point clouds, but also ease the biased effect of outliers. To further capture the neighbor and long-range dependencies of the sampled point, we proposed a local-nonlocal (L-NL) module inspired by the nonlocal operation. Such L-NL module enables the learning process insensitive to noise. Extensive experiments verify the robustness and superiority of our approach in point clouds processing tasks regardless of synthesis data, indoor data, and outdoor data with or without noise. Specifically, PointASNL achieves state-of-the-art robust performance for classification and segmentation tasks on all datasets, and significantly outperforms previous methods on real-world outdoor SemanticKITTI dataset with considerate noise. Our code is released through https://github.com/yanx27/PointASNL.
Tasks
Published	2020-03-01
URL	https://arxiv.org/abs/2003.00492v2
PDF	https://arxiv.org/pdf/2003.00492v2.pdf
PWC	https://paperswithcode.com/paper/pointasnl-robust-point-clouds-processing
Repo	https://github.com/yanx27/PointASNL
Framework	tf

Learning multiview 3D point cloud registration


Title	Learning multiview 3D point cloud registration
Authors	Zan Gojcic, Caifa Zhou, Jan D. Wegner, Leonidas J. Guibas, Tolga Birdal
Abstract	We present a novel, end-to-end learnable, multiview 3D point cloud registration algorithm. Registration of multiple scans typically follows a two-stage pipeline: the initial pairwise alignment and the globally consistent refinement. The former is often ambiguous due to the low overlap of neighboring point clouds, symmetries and repetitive scene parts. Therefore, the latter global refinement aims at establishing the cyclic consistency across multiple scans and helps in resolving the ambiguous cases. In this paper we propose, to the best of our knowledge, the first end-to-end algorithm for joint learning of both parts of this two-stage problem. Experimental evaluation on well accepted benchmark datasets shows that our approach outperforms the state-of-the-art by a significant margin, while being end-to-end trainable and computationally less costly. Moreover, we present detailed analysis and an ablation study that validate the novel components of our approach. The source code and pretrained models are publicly available under https://github.com/zgojcic/3D_multiview_reg.
Tasks	Point Cloud Registration
Published	2020-01-15
URL	https://arxiv.org/abs/2001.05119v2
PDF	https://arxiv.org/pdf/2001.05119v2.pdf
PWC	https://paperswithcode.com/paper/learning-multiview-3d-point-cloud
Repo	https://github.com/zgojcic/3D_multiview_reg
Framework	none

Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network


Title	Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network
Authors	Jakaria Rabbi, Nilanjan Ray, Matthias Schubert, Subir Chowdhury, Dennis Chao
Abstract	The detection performance of small objects in remote sensing images is not satisfactory compared to large objects, especially in low-resolution and noisy images. A generative adversarial network (GAN)-based model called enhanced super-resolution GAN (ESRGAN) shows remarkable image enhancement performance, but reconstructed images miss high-frequency edge information. Therefore, object detection performance degrades for the small objects on recovered noisy and low-resolution remote sensing images. Inspired by the success of edge enhanced GAN (EEGAN) and ESRGAN, we apply a new edge-enhanced super-resolution GAN (EESRGAN) to improve the image quality of remote sensing images and used different detector networks in an end-to-end manner where detector loss is backpropagated into the EESRGAN to improve the detection performance. We propose an architecture with three components: ESRGAN, Edge Enhancement Network (EEN), and Detection network. We use residual-in-residual dense blocks (RRDB) for both the GAN and EEN, and for the detector network, we use the faster region-based convolutional network (FRCNN) (two-stage detector) and single-shot multi-box detector (SSD) (one stage detector). Extensive experiments on car overhead with context and oil and gas storage tank (created by us) data sets show superior performance of our method compared to the standalone state-of-the-art object detectors.
Tasks	Image Enhancement, Object Detection, Small Object Detection, Super-Resolution
Published	2020-03-20
URL	https://arxiv.org/abs/2003.09085v1
PDF	https://arxiv.org/pdf/2003.09085v1.pdf
PWC	https://paperswithcode.com/paper/small-object-detection-in-remote-sensing
Repo	https://github.com/Jakaria08/Filter_Enhance_Detect
Framework	none

Deep Graph Matching Consensus


Title	Deep Graph Matching Consensus
Authors	Matthias Fey, Jan E. Lenssen, Christopher Morris, Jonathan Masci, Nils M. Kriege
Abstract	This work presents a two-stage neural architecture for learning and refining structural correspondences between graphs. First, we use localized node embeddings computed by a graph neural network to obtain an initial ranking of soft correspondences between nodes. Secondly, we employ synchronous message passing networks to iteratively re-rank the soft correspondences to reach a matching consensus in local neighborhoods between graphs. We show, theoretically and empirically, that our message passing scheme computes a well-founded measure of consensus for corresponding neighborhoods, which is then used to guide the iterative re-ranking process. Our purely local and sparsity-aware architecture scales well to large, real-world inputs while still being able to recover global correspondences consistently. We demonstrate the practical effectiveness of our method on real-world tasks from the fields of computer vision and entity alignment between knowledge graphs, on which we improve upon the current state-of-the-art. Our source code is available under https://github.com/rusty1s/ deep-graph-matching-consensus.
Tasks	Entity Alignment, Graph Matching, Knowledge Graphs
Published	2020-01-27
URL	https://arxiv.org/abs/2001.09621v1
PDF	https://arxiv.org/pdf/2001.09621v1.pdf
PWC	https://paperswithcode.com/paper/deep-graph-matching-consensus-1
Repo	https://github.com/rusty1s/deep-graph-matching-consensus
Framework	pytorch

Ultra Efficient Transfer Learning with Meta Update for Cross Subject EEG Classification


Title	Ultra Efficient Transfer Learning with Meta Update for Cross Subject EEG Classification
Authors	Tiehang Duan, Mihir Chauhan, Mohammad Abuzar Shaikh, Sargur Srihari
Abstract	Electroencephalogram (EEG) signal is widely used in brain computer interfaces (BCI), the pattern of which differs significantly across different subjects, and poses a major challenge for real world application of EEG classifiers. We found an efficient transfer learning method, named Meta UPdate Strategy (MUPS), boosts cross subject classification performance of EEG signals, and only need a small amount of data from target subject. The model tackles the problem with a two step process: (1) extract versatile features that are effective across all source subjects, and (2) adapt the model to target subject. The proposed model, which originates from meta learning, aims to find feature representation that is broadly suitable for different subjects, and maximizes sensitivity of the loss function on new subject such that one or a small number of gradient steps can lead to effective adaptation. The method can be applied to all deep learning oriented models. We performed extensive experiments on two public datasets, the proposed MUPS model outperforms current state of the arts by a large margin on accuracy and AUC-ROC when only a small amount of target data is used.
Tasks	EEG, Meta-Learning, Transfer Learning
Published	2020-03-13
URL	https://arxiv.org/abs/2003.06113v1
PDF	https://arxiv.org/pdf/2003.06113v1.pdf
PWC	https://paperswithcode.com/paper/ultra-efficient-transfer-learning-with-meta
Repo	https://github.com/tiehangd/MUPS
Framework	pytorch

Dynamic Parameter Allocation in Parameter Servers


Title	Dynamic Parameter Allocation in Parameter Servers
Authors	Alexander Renz-Wieland, Rainer Gemulla, Steffen Zeuch, Volker Markl
Abstract	To keep up with increasing dataset sizes and model complexity, distributed training has become a necessity for large machine learning tasks. Parameter servers ease the implementation of distributed parameter management—a key concern in distributed training—, but can induce severe communication overhead. To reduce communication overhead, distributed machine learning algorithms use techniques to increase parameter access locality (PAL), achieving up to linear speed-ups. We found that existing parameter servers provide only limited support for PAL techniques, however, and therefore prevent efficient training. In this paper, we explore whether and to what extent PAL techniques can be supported, and whether such support is beneficial. We propose to integrate dynamic parameter allocation into parameter servers, describe an efficient implementation of such a parameter server called Lapse, and experimentally compare its performance to existing parameter servers across a number of machine learning tasks. We found that Lapse provides near linear scaling and can be orders of magnitude faster than existing parameter servers.
Tasks
Published	2020-02-03
URL	https://arxiv.org/abs/2002.00655v1
PDF	https://arxiv.org/pdf/2002.00655v1.pdf
PWC	https://paperswithcode.com/paper/dynamic-parameter-allocation-in-parameter
Repo	https://github.com/alexrenz/lapse-ps
Framework	none

Level Three Synthetic Fingerprint Generation


Title	Level Three Synthetic Fingerprint Generation
Authors	André Brasil Vieira Wyzykowski, Mauricio Pamplona Segundo, Rubisley de Paula Lemes
Abstract	Today’s legal restrictions that protect the privacy of biometric data are hampering fingerprint recognition researches. For instance, all public databases of high-resolution fingerprints ceased to be publicly available. To address this problem, we present an approach to creating high-resolution synthetic fingerprints. We modified a state-of-the-art fingerprint generator to create ridge maps with sweat pores and trained a CycleGAN to transform these maps into realistic prints. We also create a synthetic database of high-resolution fingerprints using the proposed approach to propel further studies in this field without raising any legal issues. We test this database with two existing fingerprint matchers without adjustments to confirm the realism of the generated images. Besides, we provide a visual analysis that highlights the quality of our results compared to the state-of-the-art.
Tasks
Published	2020-02-05
URL	https://arxiv.org/abs/2002.03809v2
PDF	https://arxiv.org/pdf/2002.03809v2.pdf
PWC	https://paperswithcode.com/paper/level-three-synthetic-fingerprint-generation
Repo	https://github.com/andrewyzy/L3-SF
Framework	none

MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers


Title	MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers
Authors	Wenhui Wang, Furu Wei, Li Dong, Hangbo Bao, Nan Yang, Ming Zhou
Abstract	Pre-trained language models (e.g., BERT (Devlin et al., 2018) and its variants) have achieved remarkable success in varieties of NLP tasks. However, these models usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and online serving in real-life applications due to latency and capacity constraints. In this work, we present a simple and effective approach to compress large Transformer (Vaswani et al., 2017) based pre-trained models, termed as deep self-attention distillation. The small model (student) is trained by deeply mimicking the self-attention module, which plays a vital role in Transformer networks, of the large model (teacher). Specifically, we propose distilling the self-attention module of the last Transformer layer of the teacher, which is effective and flexible for the student. Furthermore, we introduce the scaled dot-product between values in the self-attention module as the new deep self-attention knowledge, in addition to the attention distributions (i.e., the scaled dot-product of queries and keys) that have been used in existing works. Moreover, we show that introducing a teacher assistant (Mirzadeh et al., 2019) also helps the distillation of large pre-trained Transformer models. Experimental results demonstrate that our model outperforms state-of-the-art baselines in different parameter size of student models. In particular, it retains more than 99% accuracy on SQuAD 2.0 and several GLUE benchmark tasks using 50% of the Transformer parameters and computations of the teacher model. The code and models are publicly available at https://github.com/microsoft/unilm/tree/master/minilm
Tasks
Published	2020-02-25
URL	https://arxiv.org/abs/2002.10957v1
PDF	https://arxiv.org/pdf/2002.10957v1.pdf
PWC	https://paperswithcode.com/paper/200210957
Repo	https://github.com/microsoft/unilm
Framework	pytorch