January 25, 2020

3004 words 15 mins read

Paper Group ANR 1742

A Simple but Effective BERT Model for Dialog State Tracking on Resource-Limited Systems. Using BERT for Word Sense Disambiguation. HeteSpaceyWalk: A Heterogeneous Spacey Random Walk for Heterogeneous Information Network Embedding. Learning to Modulate for Non-coherent MIMO. PoshakNet: Framework for matching dresses from real-life photos using GAN a …

A Simple but Effective BERT Model for Dialog State Tracking on Resource-Limited Systems


Title	A Simple but Effective BERT Model for Dialog State Tracking on Resource-Limited Systems
Authors	Tuan Manh Lai, Quan Hung Tran, Trung Bui, Daisuke Kihara
Abstract	In a task-oriented dialog system, the goal of dialog state tracking (DST) is to monitor the state of the conversation from the dialog history. Recently, many deep learning based methods have been proposed for the task. Despite their impressive performance, current neural architectures for DST are typically heavily-engineered and conceptually complex, making it difficult to implement, debug, and maintain them in a production setting. In this work, we propose a simple but effective DST model based on BERT. In addition to its simplicity, our approach also has a number of other advantages: (a) the number of parameters does not grow with the ontology size (b) the model can operate in situations where the domain ontology may change dynamically. Experimental results demonstrate that our BERT-based model outperforms previous methods by a large margin, achieving new state-of-the-art results on the standard WoZ 2.0 dataset. Finally, to make the model small and fast enough for resource-restricted systems, we apply the knowledge distillation method to compress our model. The final compressed model achieves comparable results with the original model while being 8x smaller and 7x faster.
Tasks
Published	2019-10-28
URL	https://arxiv.org/abs/1910.12995v3
PDF	https://arxiv.org/pdf/1910.12995v3.pdf
PWC	https://paperswithcode.com/paper/a-simple-but-effective-bert-model-for-dialog
Repo
Framework

Using BERT for Word Sense Disambiguation


Title	Using BERT for Word Sense Disambiguation
Authors	Jiaju Du, Fanchao Qi, Maosong Sun
Abstract	Word Sense Disambiguation (WSD), which aims to identify the correct sense of a given polyseme, is a long-standing problem in NLP. In this paper, we propose to use BERT to extract better polyseme representations for WSD and explore several ways of combining BERT and the classifier. We also utilize sense definitions to train a unified classifier for all words, which enables the model to disambiguate unseen polysemes. Experiments show that our model achieves the state-of-the-art results on the standard English All-word WSD evaluation.
Tasks	Word Sense Disambiguation
Published	2019-09-18
URL	https://arxiv.org/abs/1909.08358v1
PDF	https://arxiv.org/pdf/1909.08358v1.pdf
PWC	https://paperswithcode.com/paper/using-bert-for-word-sense-disambiguation
Repo
Framework

HeteSpaceyWalk: A Heterogeneous Spacey Random Walk for Heterogeneous Information Network Embedding


Title	HeteSpaceyWalk: A Heterogeneous Spacey Random Walk for Heterogeneous Information Network Embedding
Authors	Yu He, Yangqiu Song, Jianxin Li, Cheng Ji, Jian Peng, Hao Peng
Abstract	Heterogeneous information network (HIN) embedding has gained increasing interests recently. However, the current way of random-walk based HIN embedding methods have paid few attention to the higher-order Markov chain nature of meta-path guided random walks, especially to the stationarity issue. In this paper, we systematically formalize the meta-path guided random walk as a higher-order Markov chain process, and present a heterogeneous personalized spacey random walk to efficiently and effectively attain the expected stationary distribution among nodes. Then we propose a generalized scalable framework to leverage the heterogeneous personalized spacey random walk to learn embeddings for multiple types of nodes in an HIN guided by a meta-path, a meta-graph, and a meta-schema respectively. We conduct extensive experiments in several heterogeneous networks and demonstrate that our methods substantially outperform the existing state-of-the-art network embedding algorithms.
Tasks	Network Embedding
Published	2019-09-07
URL	https://arxiv.org/abs/1909.03228v1
PDF	https://arxiv.org/pdf/1909.03228v1.pdf
PWC	https://paperswithcode.com/paper/hetespaceywalk-a-heterogeneous-spacey-random
Repo
Framework

Learning to Modulate for Non-coherent MIMO


Title	Learning to Modulate for Non-coherent MIMO
Authors	Ye Wang, Toshiaki Koike-Akino
Abstract	The deep learning trend has recently impacted a variety of fields, including communication systems, where various approaches have explored the application of neural networks in place of traditional designs. Neural networks flexibly allow for data/simulation-driven optimization, but are often employed as black boxes detached from direct application of domain knowledge. Our work considers learning-based approaches addressing modulation and signal detection design for the non-coherent MIMO channel. We demonstrate that simulation-driven optimization can be performed while entirely avoiding neural networks, yet still perform comparably. Additionally, we show the feasibility of MIMO communications over extremely short coherence windows (i.e., channel coefficient stability period), with as few as two time slots.
Tasks
Published	2019-03-09
URL	http://arxiv.org/abs/1903.03711v1
PDF	http://arxiv.org/pdf/1903.03711v1.pdf
PWC	https://paperswithcode.com/paper/learning-to-modulate-for-non-coherent-mimo
Repo
Framework

PoshakNet: Framework for matching dresses from real-life photos using GAN and Siamese Network


Title	PoshakNet: Framework for matching dresses from real-life photos using GAN and Siamese Network
Authors	Abhigyan Khaund, Daksh Thapar, Aditya Nigam
Abstract	Online garment shopping has gained many customers in recent years. Describing a dress using keywords does not always yield the proper results, which in turn leads to dissatisfaction of customers. A visual search based system will be enormously beneficent to the industry. Hence, we propose a framework that can retrieve similar clothes that can be found in an image. The first task is to extract the garment from the input image (street photo). There are various challenges for that, including pose, illumination, and background clutter. We use a Generative Adversarial Network for the task of retrieving the garment that the person in the image was wearing. It has been shown that GAN can retrieve the garment very efficiently despite the challenges of street photos. Finally, a siamese based matching system takes the retrieved cloth image and matches it with the clothes in the dataset, giving us the top k matches. We take a pre-trained inception-ResNet v1 module as a siamese network (trained using triplet loss for face detection) and fine-tune it on the shopping dataset using center loss. The dataset has been collected inhouse. For training the GAN, we use the LookBook dataset, which is publically available.
Tasks	Face Detection
Published	2019-11-11
URL	https://arxiv.org/abs/1911.04237v1
PDF	https://arxiv.org/pdf/1911.04237v1.pdf
PWC	https://paperswithcode.com/paper/poshaknet-framework-for-matching-dresses-from
Repo
Framework

Face Detection on Surveillance Images


Title	Face Detection on Surveillance Images
Authors	Mohammad Iqbal Nouyed, Guodong Guo
Abstract	In last few decades, a lot of progress has been made in the field of face detection. Various face detection methods have been proposed by numerous researchers working in this area. The two well-known benchmarking platform: the FDDB and WIDER face detection provide quite challenging scenarios to assess the efficacy of the detection methods. These benchmarking data sets are mostly created using images from the public network ie. the Internet. A recent, face detection and open-set recognition challenge has shown that those same face detection algorithms produce high false alarms for images taken in surveillance scenario. This shows the difficult nature of the surveillance environment. Our proposed body pose based face detection method was one of the top performers in this competition. In this paper, we perform a comparative performance analysis of some of the well known face detection methods including the few used in that competition, and, compare them to our proposed body pose based face detection method. Experiment results show that, our proposed method that leverages body information to detect faces, is the most realistic approach in terms of accuracy, false alarms and average detection time, when surveillance scenario is in consideration.
Tasks	Face Detection, Open Set Learning
Published	2019-10-22
URL	https://arxiv.org/abs/1910.11121v1
PDF	https://arxiv.org/pdf/1910.11121v1.pdf
PWC	https://paperswithcode.com/paper/face-detection-on-surveillance-images
Repo
Framework

Adversarial Examples Are a Natural Consequence of Test Error in Noise


Title	Adversarial Examples Are a Natural Consequence of Test Error in Noise
Authors	Nic Ford, Justin Gilmer, Nicolas Carlini, Dogus Cubuk
Abstract	Over the last few years, the phenomenon of adversarial examples — maliciously constructed inputs that fool trained machine learning models — has captured the attention of the research community, especially when the adversary is restricted to small modifications of a correctly handled input. Less surprisingly, image classifiers also lack human-level performance on randomly corrupted images, such as images with additive Gaussian noise. In this paper we provide both empirical and theoretical evidence that these are two manifestations of the same underlying phenomenon, establishing close connections between the adversarial robustness and corruption robustness research programs. This suggests that improving adversarial robustness should go hand in hand with improving performance in the presence of more general and realistic image corruptions. Based on our results we recommend that future adversarial defenses consider evaluating the robustness of their methods to distributional shift with benchmarks such as Imagenet-C.
Tasks
Published	2019-01-29
URL	http://arxiv.org/abs/1901.10513v1
PDF	http://arxiv.org/pdf/1901.10513v1.pdf
PWC	https://paperswithcode.com/paper/adversarial-examples-are-a-natural
Repo
Framework

Enhancing self-supervised monocular depth estimation with traditional visual odometry


Title	Enhancing self-supervised monocular depth estimation with traditional visual odometry
Authors	Lorenzo Andraghetti, Panteleimon Myriokefalitakis, Pier Luigi Dovesi, Belen Luque, Matteo Poggi, Alessandro Pieropan, Stefano Mattoccia
Abstract	Estimating depth from a single image represents an attractive alternative to more traditional approaches leveraging multiple cameras. In this field, deep learning yielded outstanding results at the cost of needing large amounts of data labeled with precise depth measurements for training. An issue softened by self-supervised approaches leveraging monocular sequences or stereo pairs in place of expensive ground truth depth annotations. This paper enables to further improve monocular depth estimation by integrating into existing self-supervised networks a geometrical prior. Specifically, we propose a sparsity-invariant autoencoder able to process the output of conventional visual odometry algorithms working in synergy with depth-from-mono networks. Experimental results on the KITTI dataset show that by exploiting the geometrical prior, our proposal: i) outperforms existing approaches in the literature and ii) couples well with both compact and complex depth-from-mono architectures, allowing for its deployment on high-end GPUs as well as on embedded devices (e.g., NVIDIA Jetson TX2).
Tasks	Depth And Camera Motion, Depth Estimation, Monocular Depth Estimation, Visual Odometry
Published	2019-08-08
URL	https://arxiv.org/abs/1908.03127v2
PDF	https://arxiv.org/pdf/1908.03127v2.pdf
PWC	https://paperswithcode.com/paper/enhancing-self-supervised-monocular-depth
Repo
Framework

Creating Lightweight Object Detectors with Model Compression for Deployment on Edge Devices


Title	Creating Lightweight Object Detectors with Model Compression for Deployment on Edge Devices
Authors	Yiwu Yao, Weiqiang Yang, Haoqi Zhu
Abstract	To achieve lightweight object detectors for deployment on the edge devices, an effective model compression pipeline is proposed in this paper. The compression pipeline consists of automatic channel pruning for the backbone, fixed channel deletion for the branch layers and knowledge distillation for the guidance learning. As results, the Resnet50-v1d is auto-pruned and fine-tuned on ImageNet to attain a compact base model as the backbone of object detector. Then, lightweight object detectors are implemented with proposed compression pipeline. For instance, the SSD-300 with model size=16.3MB, FLOPS=2.31G, and mAP=71.2 is created, revealing a better result than SSD-300-MobileNet.
Tasks	Model Compression
Published	2019-05-06
URL	https://arxiv.org/abs/1905.01787v1
PDF	https://arxiv.org/pdf/1905.01787v1.pdf
PWC	https://paperswithcode.com/paper/creating-lightweight-object-detectors-with
Repo
Framework

Causal relationship between eWOM topics and profit of rural tourism at Japanese Roadside Stations “MICHINOEKI”


Title	Causal relationship between eWOM topics and profit of rural tourism at Japanese Roadside Stations “MICHINOEKI”
Authors	Elisa Claire Alemán Carreón, Tetsuro Ito, Hirofumi Nonaka, Minoru Kumano, Toru Hiraoka, Masaharu Hirota
Abstract	Affected by urbanization, centralization and the decrease of overall population, Japan has been making efforts to revitalize the rural areas across the country. One particular effort is to increase tourism to these rural areas via regional branding, using local farm products as tourist attractions across Japan. Particularly, a program subsidized by the government called Michinoeki, which stands for ‘roadside station’, was created 20 years ago and it strives to provide a safe and comfortable space for cultural interaction between road travelers and the local community, as well as offering refreshment, and relevant information to travelers. However, despite its importance in the revitalization of the Japanese economy, studies with newer technologies and methodologies are lacking. Using sales data from establishments in the Kyushu area of Japan, we used Support Vector to classify content from Twitter into relevant topics and studied their causal relationship to the sales for each establishment using LiNGAM, a linear non-gaussian acyclic model built for causal structure analysis, to perform an improved market analysis considering more than just correlation. Under the hypotheses stated by the LiNGAM model, we discovered a positive causal relationship between the number of tweets mentioning those establishments, specially mentioning deserts, a need for better access and traf^ic options, and a potentially untapped customer base in motorcycle biker groups.
Tasks
Published	2019-04-25
URL	http://arxiv.org/abs/1904.12039v2
PDF	http://arxiv.org/pdf/1904.12039v2.pdf
PWC	https://paperswithcode.com/paper/190412039
Repo
Framework

Predicting intelligence based on cortical WM/GM contrast, cortical thickness and volumetry


Title	Predicting intelligence based on cortical WM/GM contrast, cortical thickness and volumetry
Authors	Juan Miguel Valverde, Vandad Imani, John D. Lewis, Jussi Tohka
Abstract	We propose a four-layer fully-connected neural network (FNN) for predicting fluid intelligence scores from T1-weighted MR images for the ABCD-challenge. In addition to the volumes of brain structures, the FNN uses cortical WM/GM contrast and cortical thickness at 78 cortical regions. These last two measurements were derived from the T1-weighted MR images using cortical surfaces produced by the CIVET pipeline. The age and gender of the subjects and the scanner manufacturer are also used as features for the learning algorithm. This yielded 283 features provided to the FNN with two hidden layers of 20 and 15 nodes. The method was applied to the data from the ABCD study. Trained with a training set of 3736 subjects, the proposed method achieved a MSE of 71.596 and a correlation of 0.151 in the validation set of 415 subjects. For the final submission, the model was trained with 3568 subjects and it achieved a MSE of 94.0270 in the test set comprised of 4383 subjects.
Tasks
Published	2019-09-09
URL	https://arxiv.org/abs/1909.05660v1
PDF	https://arxiv.org/pdf/1909.05660v1.pdf
PWC	https://paperswithcode.com/paper/predicting-intelligence-based-on-cortical
Repo
Framework

Spoofing and Anti-Spoofing with Wax Figure Faces


Title	Spoofing and Anti-Spoofing with Wax Figure Faces
Authors	Shan Jia, Xin Li, Chuanbo Hu, Zhengquan Xu
Abstract	We have witnessed rapid advances in both face presentation attack models and presentation attack detection (PAD) in recent years. Compared to widely studied 2D face presentation attacks (e.g. printed photos and video replays), 3D face presentation attacks are more challenging because face recognition systems (FRS) is more easily confused by the 3D characteristics of materials similar to real faces. Existing 3D face spoofing databases, mostly based on 3D facial masks, are restricted to small data size and suffer from poor authenticity due to the difficulty and expense of mask production. In this work, we introduce a wax figure face database (WFFD) as a novel and super-realistic 3D face presentation attack. This database contains 2300 image pairs (totally 4600) and 745 subjects including both real and wax figure faces with high diversity from online collections. On one hand, our experiments have demonstrated the spoofing potential of WFFD on three popular FRSs. On the other hand, we have developed a multi-feature voting scheme for wax figure face detection (anti-spoofing), which combines three discriminative features at the decision level. The proposed detection method was compared against several face PAD approaches and found to outperform other competing methods. Surprisingly, our fusion-based detection method achieves an Average Classification Error Rate (ACER) of 11.73% on the WFFD database, which is even better than human-based detection.
Tasks	Face Detection, Face Recognition
Published	2019-10-12
URL	https://arxiv.org/abs/1910.05457v1
PDF	https://arxiv.org/pdf/1910.05457v1.pdf
PWC	https://paperswithcode.com/paper/spoofing-and-anti-spoofing-with-wax-figure
Repo
Framework

Deep Message Passing on Sets


Title	Deep Message Passing on Sets
Authors	Yifeng Shi, Junier Oliva, Marc Niethammer
Abstract	Modern methods for learning over graph input data have shown the fruitfulness of accounting for relationships among elements in a collection. However, most methods that learn over set input data use only rudimentary approaches to exploit intra-collection relationships. In this work we introduce Deep Message Passing on Sets (DMPS), a novel method that incorporates relational learning for sets. DMPS not only connects learning on graphs with learning on sets via deep kernel learning, but it also bridges message passing on sets and traditional diffusion dynamics commonly used in denoising models. Based on these connections, we develop two new blocks for relational learning on sets: the set-denoising block and the set-residual block. The former is motivated by the connection between message passing on general graphs and diffusion-based denoising models, whereas the latter is inspired by the well-known residual network. In addition to demonstrating the interpretability of our model by learning the true underlying relational structure experimentally, we also show the effectiveness of our approach on both synthetic and real-world datasets by achieving results that are competitive with or outperform the state-of-the-art.
Tasks	Denoising, Relational Reasoning
Published	2019-09-21
URL	https://arxiv.org/abs/1909.09877v1
PDF	https://arxiv.org/pdf/1909.09877v1.pdf
PWC	https://paperswithcode.com/paper/190909877
Repo
Framework

LucidDream: Controlled Temporally-Consistent DeepDream on Videos


Title	LucidDream: Controlled Temporally-Consistent DeepDream on Videos
Authors	Joel Ruben Antony Moniz, Eunsu Kang, Barnabás Póczos
Abstract	In this work, we aim to propose a set of techniques to improve the controllability and aesthetic appeal when DeepDream, which uses a pre-trained neural network to modify images by hallucinating objects into them, is applied to videos. In particular, we demonstrate a simple modification that improves control over the class of object that DeepDream is induced to hallucinate. We also show that the flickering artifacts which frequently appear when DeepDream is applied on videos can be mitigated by the use of an additional temporal consistency loss term.
Tasks
Published	2019-11-27
URL	https://arxiv.org/abs/1911.11960v1
PDF	https://arxiv.org/pdf/1911.11960v1.pdf
PWC	https://paperswithcode.com/paper/luciddream-controlled-temporally-consistent
Repo
Framework

The Impact of Extraneous Variables on the Performance of Recurrent Neural Network Models in Clinical Tasks


Title	The Impact of Extraneous Variables on the Performance of Recurrent Neural Network Models in Clinical Tasks
Authors	Eugene Laksana, Melissa Aczon, Long Ho, Cameron Carlin, David Ledbetter, Randall Wetzel
Abstract	Electronic Medical Records (EMR) are a rich source of patient information, including measurements reflecting physiologic signs and administered therapies. Identifying which variables are useful in predicting clinical outcomes can be challenging. Advanced algorithms such as deep neural networks were designed to process high-dimensional inputs containing variables in their measured form, thus bypass separate feature selection or engineering steps. We investigated the effect of extraneous input variables on the predictive performance of Recurrent Neural Networks (RNN) by including in the input vector extraneous variables randomly drawn from theoretical and empirical distributions. RNN models using different input vectors (EMR variables; EMR and extraneous variables; extraneous variables only) were trained to predict three clinical outcomes: in-ICU mortality, 72-hour ICU re-admission, and 30-day ICU-free days. The measured degradations of the RNN’s predictive performance with the addition of extraneous variables to EMR variables were negligible.
Tasks	Feature Selection
Published	2019-04-01
URL	http://arxiv.org/abs/1904.01125v1
PDF	http://arxiv.org/pdf/1904.01125v1.pdf
PWC	https://paperswithcode.com/paper/the-impact-of-extraneous-variables-on-the
Repo
Framework