Paper Group NAWR 13
TF-LM: TensorFlow-based Language Modeling Toolkit. The ACoLi CoNLL Libraries: Beyond Tab-Separated Values. Predcnn: Predictive learning with cascade convolutions. Angiogenic Factors produced by Hypoxic Cells are a leading driver of Anastomoses in Sprouting Angiogenesis–a computational study. End-to-End Differentiable Physics for Learning and Contro …
TF-LM: TensorFlow-based Language Modeling Toolkit
| Title | TF-LM: TensorFlow-based Language Modeling Toolkit |
| Authors | Lyan Verwimp, Hugo Van hamme, Patrick Wambacq |
| Abstract | |
| Tasks | Language Modelling, Machine Translation, Optical Character Recognition, Speech Recognition |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1470/ |
| https://www.aclweb.org/anthology/L18-1470 | |
| PWC | https://paperswithcode.com/paper/tf-lm-tensorflow-based-language-modeling |
| Repo | https://github.com/lverwimp/tf-lm |
| Framework | tf |
The ACoLi CoNLL Libraries: Beyond Tab-Separated Values
| Title | The ACoLi CoNLL Libraries: Beyond Tab-Separated Values |
| Authors | Christian Chiarcos, Niko Schenk |
| Abstract | |
| Tasks | Coreference Resolution, Semantic Role Labeling, Tokenization |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1090/ |
| https://www.aclweb.org/anthology/L18-1090 | |
| PWC | https://paperswithcode.com/paper/the-acoli-conll-libraries-beyond-tab |
| Repo | https://github.com/acoli-repo/conll |
| Framework | none |
Predcnn: Predictive learning with cascade convolutions
| Title | Predcnn: Predictive learning with cascade convolutions |
| Authors | Ziru Xu, Yunbo Wang, Mingsheng Long, Jianmin Wang |
| Abstract | Predicting future frames in videos remains an unsolved but challenging problem. Mainstream recurrent models suffer from huge memory usage and computation cost, while convolutional models are unable to effectively capture the temporal dependencies between consecutive video frames. To tackle this problem, we introduce an entirely CNN-based architecture, PredCNN, that models the dependencies between the next frame and the sequential video inputs. Inspired by the core idea of recurrent models that previous states have more transition operations than future states, we design a cascade multiplicative unit (CMU) that provides relatively more operations for previous video frames. This newly proposed unit enables PredCNN to predict future spatiotemporal data without any recurrent chain structures, which eases gradient propagation and enables a fully paralleled optimization. We show that PredCNN outperforms the state-of-the-art recurrent models for video prediction on the standard Moving MNIST dataset and two challenging crowd flow prediction datasets, and achieves a faster training speed and lower memory footprint. |
| Tasks | Pose Prediction, Video Prediction |
| Published | 2018-07-01 |
| URL | https://doi.org/10.24963/ijcai.2018/408 |
| https://www.researchgate.net/publication/326201620_PredCNN_Predictive_Learning_with_Cascade_Convolutions | |
| PWC | https://paperswithcode.com/paper/predcnn-predictive-learning-with-cascade |
| Repo | https://github.com/xzr12/PredCNN |
| Framework | tf |
Angiogenic Factors produced by Hypoxic Cells are a leading driver of Anastomoses in Sprouting Angiogenesis–a computational study
| Title | Angiogenic Factors produced by Hypoxic Cells are a leading driver of Anastomoses in Sprouting Angiogenesis–a computational study |
| Authors | Maurício Moreira-Soares, Rita Coimbra, Luís Rebelo, João Carvalho, Rui D. M. Travasso |
| Abstract | Angiogenesis - the growth of new blood vessels from a pre-existing vasculature - is key in both physiological processes and on several pathological scenarios such as cancer progression or diabetic retinopathy. For the new vascular networks to be functional, it is required that the growing sprouts merge either with an existing functional mature vessel or with another growing sprout. This process is called anastomosis. We present a systematic 2D and 3D computational study of vessel growth in a tissue to address the capability of angiogenic factor gradients to drive anastomosis formation. We consider that these growth factors are produced only by tissue cells in hypoxia, i.e. until nearby vessels merge and become capable of carrying blood and irrigating their vicinity. We demonstrate that this increased production of angiogenic factors by hypoxic cells is able to promote vessel anastomoses events in both 2D and 3D. The simulations also verify that the morphology of these networks has an increased resilience toward variations in the endothelial cell’s proliferation and chemotactic response. The distribution of tissue cells and the concentration of the growth factors they produce are the major factors in determining the final morphology of the network. |
| Tasks | |
| Published | 2018-06-07 |
| URL | https://www.nature.com/articles/s41598-018-27034-8 |
| https://www.nature.com/articles/s41598-018-27034-8.pdf | |
| PWC | https://paperswithcode.com/paper/angiogenic-factors-produced-by-hypoxic-cells |
| Repo | https://github.com/phydev/angio |
| Framework | none |
End-to-End Differentiable Physics for Learning and Control
| Title | End-to-End Differentiable Physics for Learning and Control |
| Authors | Filipe De Avila Belbute-Peres, Kevin Smith, Kelsey Allen, Josh Tenenbaum, J. Zico Kolter |
| Abstract | We present a differentiable physics engine that can be integrated as a module in deep neural networks for end-to-end learning. As a result, structured physics knowledge can be embedded into larger systems, allowing them, for example, to match observations by performing precise simulations, while achieves high sample efficiency. Specifically, in this paper we demonstrate how to perform backpropagation analytically through a physical simulator defined via a linear complementarity problem. Unlike traditional finite difference methods, such gradients can be computed analytically, which allows for greater flexibility of the engine. Through experiments in diverse domains, we highlight the system’s ability to learn physical parameters from data, efficiently match and simulate observed visual behavior, and readily enable control via gradient-based planning methods. Code for the engine and experiments is included with the paper. |
| Tasks | |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/7948-end-to-end-differentiable-physics-for-learning-and-control |
| http://papers.nips.cc/paper/7948-end-to-end-differentiable-physics-for-learning-and-control.pdf | |
| PWC | https://paperswithcode.com/paper/end-to-end-differentiable-physics-for |
| Repo | https://github.com/locuslab/lcp-physics |
| Framework | pytorch |
Signal mixture estimation for degenerate heavy Higgses using a deep neural network
| Title | Signal mixture estimation for degenerate heavy Higgses using a deep neural network |
| Authors | Anders Kvellestad, Steffen Maeland, Inga Strümke |
| Abstract | If a new signal is established in future LHC data, a next question will be to determine the signal composition, in particular whether the signal is due to multiple near-degenerate states. We investigate the performance of a deep learning approach to signal mixture estimation for the challenging scenario of a ditau signal coming from a pair of degenerate Higgs bosons of opposite CP charge. This constitutes a parameter estimation problem for a mixture model with highly overlapping features. We use an unbinned maximum likelihood fit to a neural network output, and compare the results to mixture estimation via a fit to a single kinematic variable. For our benchmark scenarios we find a ∼20% improvement in the estimate uncertainty. |
| Tasks | |
| Published | 2018-12-12 |
| URL | https://link.springer.com/article/10.1140/epjc/s10052-018-6455-z |
| https://link.springer.com/content/pdf/10.1140%2Fepjc%2Fs10052-018-6455-z.pdf | |
| PWC | https://paperswithcode.com/paper/signal-mixture-estimation-for-degenerate |
| Repo | https://github.com/smaeland/ML-2HDM |
| Framework | tf |
CartoonGAN: Generative Adversarial Networks for Photo Cartoonization
| Title | CartoonGAN: Generative Adversarial Networks for Photo Cartoonization |
| Authors | Yang Chen, Yu-Kun Lai, Yong-Jin Liu |
| Abstract | In this paper, we propose a solution to transforming photos of real-world scenes into cartoon style images, which is valuable and challenging in computer vision and computer graphics. Our solution belongs to learning based methods, which have recently become popular to stylize images in artistic forms such as painting. However, existing methods do not produce satisfactory results for cartoonization, due to the fact that (1) cartoon styles have unique characteristics with high level simplification and abstraction, and (2) cartoon images tend to have clear edges, smooth color shading and relatively simple textures, which exhibit significant challenges for texture-descriptor-based loss functions used in existing methods. In this paper, we propose CartoonGAN, a generative adversarial network (GAN) framework for cartoon stylization. Our method takes unpaired photos and cartoon images for training, which is easy to use. Two novel losses suitable for cartoonization are proposed: (1) a semantic content loss, which is formulated as a sparse regularization in the high-level feature maps of the VGG network to cope with substantial style variation between photos and cartoons, and (2) an edge-promoting adversarial loss for preserving clear edges. We further introduce an initialization phase, to improve the convergence of the network to the target manifold. Our method is also much more efficient to train than existing methods. Experimental results show that our method is able to generate high-quality cartoon images from real-world photos (i.e., following specific artists’ styles and with clear edges and smooth shading) and outperforms state-of-the-art methods. |
| Tasks | Image-to-Image Translation, Real-to-Cartoon translation |
| Published | 2018-06-01 |
| URL | http://openaccess.thecvf.com/content_cvpr_2018/html/Chen_CartoonGAN_Generative_Adversarial_CVPR_2018_paper.html |
| http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_CartoonGAN_Generative_Adversarial_CVPR_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/cartoongan-generative-adversarial-networks |
| Repo | https://github.com/Yijunmaverick/CartoonGAN-Test-Pytorch-Torch |
| Framework | pytorch |
RUSE: Regressor Using Sentence Embeddings for Automatic Machine Translation Evaluation
| Title | RUSE: Regressor Using Sentence Embeddings for Automatic Machine Translation Evaluation |
| Authors | Hiroki Shimanaka, Tomoyuki Kajiwara, Mamoru Komachi |
| Abstract | We introduce the RUSE metric for the WMT18 metrics shared task. Sentence embeddings can capture global information that cannot be captured by local features based on character or word N-grams. Although training sentence embeddings using small-scale translation datasets with manual evaluation is difficult, sentence embeddings trained from large-scale data in other tasks can improve the automatic evaluation of machine translation. We use a multi-layer perceptron regressor based on three types of sentence embeddings. The experimental results of the WMT16 and WMT17 datasets show that the RUSE metric achieves a state-of-the-art performance in both segment- and system-level metrics tasks with embedding features only. |
| Tasks | Machine Translation, Sentence Embeddings |
| Published | 2018-10-01 |
| URL | https://www.aclweb.org/anthology/W18-6456/ |
| https://www.aclweb.org/anthology/W18-6456 | |
| PWC | https://paperswithcode.com/paper/ruse-regressor-using-sentence-embeddings-for |
| Repo | https://github.com/Shi-ma/RUSE |
| Framework | pytorch |
GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation
| Title | GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation |
| Authors | Xiaojuan Qi, Renjie Liao, Zhengzhe Liu, Raquel Urtasun, Jiaya Jia |
| Abstract | In this paper, we propose Geometric Neural Network (GeoNet) to jointly predict depth and surface normal maps from a single image. Building on top of two-stream CNNs, our GeoNet incorporates geometric relation between depth and surface normal via the new depth-to-normal and normal- to-depth networks. Depth-to-normal network exploits the least square solution of surface normal from depth and im- proves its quality with a residual module. Normal-to-depth network, contrarily, refines the depth map based on the con- straints from the surface normal through a kernel regression module, which has no parameter to learn. These two net- works enforce the underlying model to efficiently predict depth and surface normal for high consistency and corre- sponding accuracy. Our experiments on NYU v2 dataset verify that our GeoNet is able to predict geometrically con- sistent depth and normal maps. It achieves top performance on surface normal estimation and is on par with state-of-the- art depth estimation methods. |
| Tasks | Depth Estimation |
| Published | 2018-06-01 |
| URL | http://openaccess.thecvf.com/content_cvpr_2018/html/Qi_GeoNet_Geometric_Neural_CVPR_2018_paper.html |
| http://openaccess.thecvf.com/content_cvpr_2018/papers/Qi_GeoNet_Geometric_Neural_CVPR_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/geonet-geometric-neural-network-for-joint |
| Repo | https://github.com/xjqi/GeoNet.git |
| Framework | none |
Efficient Generation and Processing of Word Co-occurrence Networks Using corpus2graph
| Title | Efficient Generation and Processing of Word Co-occurrence Networks Using corpus2graph |
| Authors | Zheng Zhang, Pierre Zweigenbaum, Ruiqing Yin |
| Abstract | Corpus2graph is an open-source NLP-application-oriented tool that generates a word co-occurrence network from a large corpus. It not only contains different built-in methods to preprocess words, analyze sentences, extract word pairs and define edge weights, but also supports user-customized functions. By using parallelization techniques, it can generate a large word co-occurrence network of the whole English Wikipedia data within hours. And thanks to its nodes-edges-weight three-level progressive calculation design, rebuilding networks with different configurations is even faster as it does not need to start all over again. This tool also works with other graph libraries such as igraph, NetworkX and graph-tool as a front end providing data to boost network generation speed. |
| Tasks | Keyword Extraction |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/W18-1702/ |
| https://www.aclweb.org/anthology/W18-1702 | |
| PWC | https://paperswithcode.com/paper/efficient-generation-and-processing-of-word |
| Repo | https://github.com/zzcoolj/corpus2graph |
| Framework | none |
Polarity Computations in Flexible Categorial Grammar
| Title | Polarity Computations in Flexible Categorial Grammar |
| Authors | Hai Hu, Larry Moss |
| Abstract | This paper shows how to take parse trees in CCG and algorithmically find the polarities of all the constituents. Our work uses the well-known polarization principle corresponding to function application, and we have extended this with principles for type raising and composition. We provide an algorithm, extending the polarity marking algorithm of van Benthem. We discuss how our system works in practice, taking input from the C{&}C parser. |
| Tasks | |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-2015/ |
| https://www.aclweb.org/anthology/S18-2015 | |
| PWC | https://paperswithcode.com/paper/polarity-computations-in-flexible-categorial |
| Repo | https://github.com/huhailinguist/ccg2mono |
| Framework | none |
Learning Single-View 3D Reconstruction with Limited Pose Supervision
| Title | Learning Single-View 3D Reconstruction with Limited Pose Supervision |
| Authors | Guandao Yang, Yin Cui, Serge Belongie, Bharath Hariharan |
| Abstract | It is expensive to label images with 3D structure or precise camera pose. Yet, this is precisely the kind of annotation required to train single-view 3D reconstruction models. In contrast, unlabeled images or images with just category labels are easy to acquire, but few current models can use this weak supervision. We present a unified framework that can combine both types of supervision: a small amount of camera pose annotations are used to enforce pose-invariance and view-point consistency, and unlabeled images combined with an adversarial loss are used to enforce the realism of rendered, generated models. We use this unified framework to measure the impact of each form of supervision in three paradigms: semi-supervised, multi-task, and transfer learning. We show that with a combination of these ideas, we can train single-view reconstruction models that improve up to 7 points in performance (AP) when using only 1% pose annotated training data. |
| Tasks | 3D Reconstruction, Single-View 3D Reconstruction, Transfer Learning |
| Published | 2018-09-01 |
| URL | http://openaccess.thecvf.com/content_ECCV_2018/html/Guandao_Yang_A_Unified_Framework_ECCV_2018_paper.html |
| http://openaccess.thecvf.com/content_ECCV_2018/papers/Guandao_Yang_A_Unified_Framework_ECCV_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/learning-single-view-3d-reconstruction-with |
| Repo | https://github.com/stevenygd/3d-recon |
| Framework | tf |
Scale Aggregation Network for Accurate and Efficient Crowd Counting
| Title | Scale Aggregation Network for Accurate and Efficient Crowd Counting |
| Authors | Xinkun Cao, Zhipeng Wang, Yanyun Zhao, Fei Su |
| Abstract | In this paper, we propose a novel encoder-decoder network, called extit{Scale Aggregation Network (SANet)}, for accurate and efficient crowd counting. The encoder extracts multi-scale features with scale aggregation modules and the decoder generates high-resolution density maps by using a set of transposed convolutions. Moreover, we find that most existing works use only Euclidean loss which assumes independence among each pixel but ignores the local correlation in density maps. Therefore, we propose a novel training loss, combining of Euclidean loss and local pattern consistency loss, which improves the performance of the model in our experiments. In addition, we use normalization layers to ease the training process and apply a patch-based test scheme to reduce the impact of statistic shift problem. To demonstrate the effectiveness of the proposed method, we conduct extensive experiments on four major crowd counting datasets and our method achieves superior performance to state-of-the-art methods while with much less parameters. |
| Tasks | Crowd Counting |
| Published | 2018-09-01 |
| URL | http://openaccess.thecvf.com/content_ECCV_2018/html/Xinkun_Cao_Scale_Aggregation_Network_ECCV_2018_paper.html |
| http://openaccess.thecvf.com/content_ECCV_2018/papers/Xinkun_Cao_Scale_Aggregation_Network_ECCV_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/scale-aggregation-network-for-accurate-and |
| Repo | https://github.com/ZhengPeng7/SANet-Keras |
| Framework | tf |
An evaluation metric for object detection algorithms in autonomous navigation systems and its application to a real-time alerting system
| Title | An evaluation metric for object detection algorithms in autonomous navigation systems and its application to a real-time alerting system |
| Authors | Harshitha Machiraju, Sumohana S. Channappayya |
| Abstract | An autonomous navigation system relies on a number of sensors including radar, LIDAR and a visible light camera for its operation. We focus our attention on the visible light camera in this work. Object detection is the key first step to processing the video input from the camera. Specifically, we address the problem of assessing the performance of object detection algorithms in hazardous driving conditions that an autonomous navigation system is expected to encounter in a realistic scenario. To this end, we propose a novel metric for quantifying the degradation in performance of an object detection algorithm under different weather conditions. Additionally’ we introduce a real-time method to detect extreme variations in performance of the algorithm that can be used to issue an alert. We evaluate the performance of our metric and alerting system and demonstrate its utility using the YOLOv2 object detection algorithm trained on the KITTI and virtual KITTI dataset. |
| Tasks | Autonomous Navigation, Object Detection |
| Published | 2018-10-10 |
| URL | https://ieeexplore.ieee.org/abstract/document/8451718 |
| https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8451718 | |
| PWC | https://paperswithcode.com/paper/an-evaluation-metric-for-object-detection |
| Repo | https://github.com/code-Assasin/Realtime-Alerting-System-for-Autonomous-Vehicles |
| Framework | none |
Better Conversations by Modeling, Filtering, and Optimizing for Coherence and Diversity
| Title | Better Conversations by Modeling, Filtering, and Optimizing for Coherence and Diversity |
| Authors | Xinnuo Xu, Ond{\v{r}}ej Du{\v{s}}ek, Ioannis Konstas, Verena Rieser |
| Abstract | We present three enhancements to existing encoder-decoder models for open-domain conversational agents, aimed at effectively modeling coherence and promoting output diversity: (1) We introduce a measure of coherence as the GloVe embedding similarity between the dialogue context and the generated response, (2) we filter our training corpora based on the measure of coherence to obtain topically coherent and lexically diverse context-response pairs, (3) we then train a response generator using a conditional variational autoencoder model that incorporates the measure of coherence as a latent variable and uses a context gate to guarantee topical consistency with the context and promote lexical diversity. Experiments on the OpenSubtitles corpus show a substantial improvement over competitive neural models in terms of BLEU score as well as metrics of coherence and diversity. |
| Tasks | Dialogue Generation |
| Published | 2018-10-01 |
| URL | https://www.aclweb.org/anthology/D18-1432/ |
| https://www.aclweb.org/anthology/D18-1432 | |
| PWC | https://paperswithcode.com/paper/better-conversations-by-modeling-filtering |
| Repo | https://github.com/XinnuoXu/CVAE_Dial |
| Framework | pytorch |
