October 20, 2019

2661 words 13 mins read

Paper Group AWR 212

Deep Learning for Classification Tasks on Geospatial Vector Polygons. Generating Distractors for Reading Comprehension Questions from Real Examinations. Neural Likelihoods via Cumulative Distribution Functions. LemmaTag: Jointly Tagging and Lemmatizing for Morphologically-Rich Languages with BRNNs. Likelihood-free inference with emulator networks. …

Deep Learning for Classification Tasks on Geospatial Vector Polygons


Title	Deep Learning for Classification Tasks on Geospatial Vector Polygons
Authors	Rein van ‘t Veer, Peter Bloem, Erwin Folmer
Abstract	In this paper, we evaluate the accuracy of deep learning approaches on geospatial vector geometry classification tasks. The purpose of this evaluation is to investigate the ability of deep learning models to learn from geometry coordinates directly. Previous machine learning research applied to geospatial polygon data did not use geometries directly, but derived properties thereof. These are produced by way of extracting geometry properties such as Fourier descriptors. Instead, our introduced deep neural net architectures are able to learn on sequences of coordinates mapped directly from polygons. In three classification tasks we show that the deep learning architectures are competitive with common learning algorithms that require extracted features.
Tasks
Published	2018-06-11
URL	https://arxiv.org/abs/1806.03857v2
PDF	https://arxiv.org/pdf/1806.03857v2.pdf
PWC	https://paperswithcode.com/paper/deep-learning-for-classification-tasks-on
Repo	https://github.com/SPINlab/geometry-learning
Framework	tf

Generating Distractors for Reading Comprehension Questions from Real Examinations


Title	Generating Distractors for Reading Comprehension Questions from Real Examinations
Authors	Yifan Gao, Lidong Bing, Piji Li, Irwin King, Michael R. Lyu
Abstract	We investigate the task of distractor generation for multiple choice reading comprehension questions from examinations. In contrast to all previous works, we do not aim at preparing words or short phrases distractors, instead, we endeavor to generate longer and semantic-rich distractors which are closer to distractors in real reading comprehension from examinations. Taking a reading comprehension article, a pair of question and its correct option as input, our goal is to generate several distractors which are somehow related to the answer, consistent with the semantic context of the question and have some trace in the article. We propose a hierarchical encoder-decoder framework with static and dynamic attention mechanisms to tackle this task. Specifically, the dynamic attention can combine sentence-level and word-level attention varying at each recurrent time step to generate a more readable sequence. The static attention is to modulate the dynamic attention not to focus on question irrelevant sentences or sentences which contribute to the correct option. Our proposed framework outperforms several strong baselines on the first prepared distractor generation dataset of real reading comprehension questions. For human evaluation, compared with those distractors generated by baselines, our generated distractors are more functional to confuse the annotators.
Tasks	Reading Comprehension
Published	2018-09-08
URL	http://arxiv.org/abs/1809.02768v2
PDF	http://arxiv.org/pdf/1809.02768v2.pdf
PWC	https://paperswithcode.com/paper/generating-distractors-for-reading
Repo	https://github.com/Evan-Gao/Distractor-Generation-RACE
Framework	none

Neural Likelihoods via Cumulative Distribution Functions


Title	Neural Likelihoods via Cumulative Distribution Functions
Authors	Pawel Chilinski, Ricardo Silva
Abstract	We leverage neural networks as universal approximators of monotonic functions to build a parameterization of conditional cumulative distribution functions. By a modification of backpropagation as applied both to parameters and outputs, we show that we are able to build black box density estimators which are competitive against recently proposed models, while avoiding assumptions concerning the base distribution in a mixture model. That is, it makes no use of parametric models as building blocks. This approach removes some undesirable degrees of freedom on the design on neural networks for flexible conditional density estimation, while implementation can be easily accomplished by standard algorithms readily available in popular neural network toolboxes.
Tasks	Density Estimation
Published	2018-11-02
URL	http://arxiv.org/abs/1811.00974v1
PDF	http://arxiv.org/pdf/1811.00974v1.pdf
PWC	https://paperswithcode.com/paper/neural-likelihoods-via-cumulative
Repo	https://github.com/pawelc/NeuralLikelihoods0
Framework	tf

LemmaTag: Jointly Tagging and Lemmatizing for Morphologically-Rich Languages with BRNNs


Title	LemmaTag: Jointly Tagging and Lemmatizing for Morphologically-Rich Languages with BRNNs
Authors	Daniel Kondratyuk, Tomáš Gavenčiak, Milan Straka, Jan Hajič
Abstract	We present LemmaTag, a featureless neural network architecture that jointly generates part-of-speech tags and lemmas for sentences by using bidirectional RNNs with character-level and word-level embeddings. We demonstrate that both tasks benefit from sharing the encoding part of the network, predicting tag subcategories, and using the tagger output as an input to the lemmatizer. We evaluate our model across several languages with complex morphology, which surpasses state-of-the-art accuracy in both part-of-speech tagging and lemmatization in Czech, German, and Arabic.
Tasks	Lemmatization, Part-Of-Speech Tagging
Published	2018-08-10
URL	http://arxiv.org/abs/1808.03703v2
PDF	http://arxiv.org/pdf/1808.03703v2.pdf
PWC	https://paperswithcode.com/paper/lemmatag-jointly-tagging-and-lemmatizing-for
Repo	https://github.com/hyperparticle/LemmaTag
Framework	tf

Likelihood-free inference with emulator networks


Title	Likelihood-free inference with emulator networks
Authors	Jan-Matthis Lueckmann, Giacomo Bassetto, Theofanis Karaletsos, Jakob H. Macke
Abstract	Approximate Bayesian Computation (ABC) provides methods for Bayesian inference in simulation-based stochastic models which do not permit tractable likelihoods. We present a new ABC method which uses probabilistic neural emulator networks to learn synthetic likelihoods on simulated data – both local emulators which approximate the likelihood for specific observed data, as well as global ones which are applicable to a range of data. Simulations are chosen adaptively using an acquisition function which takes into account uncertainty about either the posterior distribution of interest, or the parameters of the emulator. Our approach does not rely on user-defined rejection thresholds or distance functions. We illustrate inference with emulator networks on synthetic examples and on a biophysical neuron model, and show that emulators allow accurate and efficient inference even on high-dimensional problems which are challenging for conventional ABC approaches.
Tasks	Bayesian Inference
Published	2018-05-23
URL	https://arxiv.org/abs/1805.09294v2
PDF	https://arxiv.org/pdf/1805.09294v2.pdf
PWC	https://paperswithcode.com/paper/likelihood-free-inference-with-emulator
Repo	https://github.com/justinalsing/pydelfi
Framework	tf

New Heuristics for Parallel and Scalable Bayesian Optimization


Title	New Heuristics for Parallel and Scalable Bayesian Optimization
Authors	Ran Rubin
Abstract	Bayesian optimization has emerged as a strong candidate tool for global optimization of functions with expensive evaluation costs. However, due to the dynamic nature of research in Bayesian approaches, and the evolution of computing technology, using Bayesian optimization in a parallel computing environment remains a challenge for the non-expert. In this report, I review the state-of-the-art in parallel and scalable Bayesian optimization methods. In addition, I propose practical ways to avoid a few of the pitfalls of Bayesian optimization, such as oversampling of edge parameters and over-exploitation of high performance parameters. Finally, I provide relatively simple, heuristic algorithms, along with their open source software implementations, that can be immediately and easily deployed in any computing environment.
Tasks
Published	2018-07-01
URL	http://arxiv.org/abs/1807.00373v2
PDF	http://arxiv.org/pdf/1807.00373v2.pdf
PWC	https://paperswithcode.com/paper/new-heuristics-for-parallel-and-scalable
Repo	https://github.com/ranr01/miniBOP
Framework	none

Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding


Title	Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
Authors	Kexin Yi, Jiajun Wu, Chuang Gan, Antonio Torralba, Pushmeet Kohli, Joshua B. Tenenbaum
Abstract	We marry two powerful ideas: deep representation learning for visual recognition and language understanding, and symbolic program execution for reasoning. Our neural-symbolic visual question answering (NS-VQA) system first recovers a structural scene representation from the image and a program trace from the question. It then executes the program on the scene representation to obtain an answer. Incorporating symbolic structure as prior knowledge offers three unique advantages. First, executing programs on a symbolic space is more robust to long program traces; our model can solve complex reasoning tasks better, achieving an accuracy of 99.8% on the CLEVR dataset. Second, the model is more data- and memory-efficient: it performs well after learning on a small number of training data; it can also encode an image into a compact representation, requiring less storage than existing methods for offline question answering. Third, symbolic program execution offers full transparency to the reasoning process; we are thus able to interpret and diagnose each execution step.
Tasks	Representation Learning, Visual Question Answering
Published	2018-10-04
URL	http://arxiv.org/abs/1810.02338v2
PDF	http://arxiv.org/pdf/1810.02338v2.pdf
PWC	https://paperswithcode.com/paper/neural-symbolic-vqa-disentangling-reasoning
Repo	https://github.com/kexinyi/ns-vqa
Framework	pytorch

A Unified Mammogram Analysis Method via Hybrid Deep Supervision


Title	A Unified Mammogram Analysis Method via Hybrid Deep Supervision
Authors	Rongzhao Zhang, Han Zhang, Albert C. S. Chung
Abstract	Automatic mammogram classification and mass segmentation play a critical role in a computer-aided mammogram screening system. In this work, we present a unified mammogram analysis framework for both whole-mammogram classification and segmentation. Our model is designed based on a deep U-Net with residual connections, and equipped with the novel hybrid deep supervision (HDS) scheme for end-to-end multi-task learning. As an extension of deep supervision (DS), HDS not only can force the model to learn more discriminative features like DS, but also seamlessly integrates segmentation and classification tasks into one model, thus the model can benefit from both pixel-wise and image-wise supervisions. We extensively validate the proposed method on the widely-used INbreast dataset. Ablation study corroborates that pixel-wise and image-wise supervisions are mutually beneficial, evidencing the efficacy of HDS. The results of 5-fold cross validation indicate that our unified model matches state-of-the-art performance on both mammogram segmentation and classification tasks, which achieves an average segmentation Dice similarity coefficient (DSC) of 0.85 and a classification accuracy of 0.89. The code is available at https://github.com/angrypudding/hybrid-ds.
Tasks	Multi-Task Learning, Whole Mammogram Classification
Published	2018-08-31
URL	http://arxiv.org/abs/1808.10646v1
PDF	http://arxiv.org/pdf/1808.10646v1.pdf
PWC	https://paperswithcode.com/paper/a-unified-mammogram-analysis-method-via
Repo	https://github.com/angrypudding/hybrid-ds
Framework	pytorch

GANimation: Anatomically-aware Facial Animation from a Single Image


Title	GANimation: Anatomically-aware Facial Animation from a Single Image
Authors	Albert Pumarola, Antonio Agudo, Aleix M. Martinez, Alberto Sanfeliu, Francesc Moreno-Noguer
Abstract	Recent advances in Generative Adversarial Networks (GANs) have shown impressive results for task of facial expression synthesis. The most successful architecture is StarGAN, that conditions GANs generation process with images of a specific domain, namely a set of images of persons sharing the same expression. While effective, this approach can only generate a discrete number of expressions, determined by the content of the dataset. To address this limitation, in this paper, we introduce a novel GAN conditioning scheme based on Action Units (AU) annotations, which describes in a continuous manifold the anatomical facial movements defining a human expression. Our approach allows controlling the magnitude of activation of each AU and combine several of them. Additionally, we propose a fully unsupervised strategy to train the model, that only requires images annotated with their activated AUs, and exploit attention mechanisms that make our network robust to changing backgrounds and lighting conditions. Extensive evaluation show that our approach goes beyond competing conditional generators both in the capability to synthesize a much wider range of expressions ruled by anatomically feasible muscle movements, as in the capacity of dealing with images in the wild.
Tasks	Conditional Image Generation, Face Generation, Image-to-Image Translation
Published	2018-07-24
URL	http://arxiv.org/abs/1807.09251v2
PDF	http://arxiv.org/pdf/1807.09251v2.pdf
PWC	https://paperswithcode.com/paper/ganimation-anatomically-aware-facial
Repo	https://github.com/albertpumarola/GANimation
Framework	pytorch

Face Synthesis for Eyeglass-Robust Face Recognition


Title	Face Synthesis for Eyeglass-Robust Face Recognition
Authors	Jianzhu Guo, Xiangyu Zhu, Zhen Lei, Stan Z. Li
Abstract	In the application of face recognition, eyeglasses could significantly degrade the recognition accuracy. A feasible method is to collect large-scale face images with eyeglasses for training deep learning methods. However, it is difficult to collect the images with and without glasses of the same identity, so that it is difficult to optimize the intra-variations caused by eyeglasses. In this paper, we propose to address this problem in a virtual synthesis manner. The high-fidelity face images with eyeglasses are synthesized based on 3D face model and 3D eyeglasses. Models based on deep learning methods are then trained on the synthesized eyeglass face dataset, achieving better performance than previous ones. Experiments on the real face database validate the effectiveness of our synthesized data for improving eyeglass face recognition performance.
Tasks	Face Generation, Face Recognition, Robust Face Recognition
Published	2018-06-04
URL	http://arxiv.org/abs/1806.01196v1
PDF	http://arxiv.org/pdf/1806.01196v1.pdf
PWC	https://paperswithcode.com/paper/face-synthesis-for-eyeglass-robust-face
Repo	https://github.com/cleardusk/MeGlass
Framework	none

Sanity Checks for Saliency Maps


Title	Sanity Checks for Saliency Maps
Authors	Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim
Abstract	Saliency methods have emerged as a popular tool to highlight features in an input deemed relevant for the prediction of a learned model. Several saliency methods have been proposed, often guided by visual appeal on image data. In this work, we propose an actionable methodology to evaluate what kinds of explanations a given method can and cannot provide. We find that reliance, solely, on visual assessment can be misleading. Through extensive experiments we show that some existing saliency methods are independent both of the model and of the data generating process. Consequently, methods that fail the proposed tests are inadequate for tasks that are sensitive to either data or model, such as, finding outliers in the data, explaining the relationship between inputs and outputs that the model learned, and debugging the model. We interpret our findings through an analogy with edge detection in images, a technique that requires neither training data nor model. Theory in the case of a linear model and a single-layer convolutional neural network supports our experimental findings.
Tasks	Edge Detection
Published	2018-10-08
URL	http://arxiv.org/abs/1810.03292v2
PDF	http://arxiv.org/pdf/1810.03292v2.pdf
PWC	https://paperswithcode.com/paper/sanity-checks-for-saliency-maps
Repo	https://github.com/jendawkins/saliencySanity
Framework	pytorch

EffNet: An Efficient Structure for Convolutional Neural Networks


Title	EffNet: An Efficient Structure for Convolutional Neural Networks
Authors	Ido Freeman, Lutz Roese-Koerner, Anton Kummert
Abstract	With the ever increasing application of Convolutional Neural Networks to customer products the need emerges for models to efficiently run on embedded, mobile hardware. Slimmer models have therefore become a hot research topic with various approaches which vary from binary networks to revised convolution layers. We offer our contribution to the latter and propose a novel convolution block which significantly reduces the computational burden while surpassing the current state-of-the-art. Our model, dubbed EffNet, is optimised for models which are slim to begin with and is created to tackle issues in existing models such as MobileNet and ShuffleNet.
Tasks
Published	2018-01-19
URL	http://arxiv.org/abs/1801.06434v6
PDF	http://arxiv.org/pdf/1801.06434v6.pdf
PWC	https://paperswithcode.com/paper/effnet-an-efficient-structure-for
Repo	https://github.com/andrijdavid/EffNet
Framework	pytorch

Word Sense Induction with Neural biLM and Symmetric Patterns


Title	Word Sense Induction with Neural biLM and Symmetric Patterns
Authors	Asaf Amrami, Yoav Goldberg
Abstract	An established method for Word Sense Induction (WSI) uses a language model to predict probable substitutes for target words, and induces senses by clustering these resulting substitute vectors. We replace the ngram-based language model (LM) with a recurrent one. Beyond being more accurate, the use of the recurrent LM allows us to effectively query it in a creative way, using what we call dynamic symmetric patterns. The combination of the RNN-LM and the dynamic symmetric patterns results in strong substitute vectors for WSI, allowing to surpass the current state-of-the-art on the SemEval 2013 WSI shared task by a large margin.
Tasks	Word Sense Induction
Published	2018-08-26
URL	http://arxiv.org/abs/1808.08518v2
PDF	http://arxiv.org/pdf/1808.08518v2.pdf
PWC	https://paperswithcode.com/paper/word-sense-induction-with-neural-bilm-and
Repo	https://github.com/asafamr/SymPatternWSI
Framework	pytorch

Applying Domain Randomization to Synthetic Data for Object Category Detection


Title	Applying Domain Randomization to Synthetic Data for Object Category Detection
Authors	João Borrego, Atabak Dehban, Rui Figueiredo, Plinio Moreno, Alexandre Bernardino, José Santos-Victor
Abstract	Recent advances in deep learning-based object detection techniques have revolutionized their applicability in several fields. However, since these methods rely on unwieldy and large amounts of data, a common practice is to download models pre-trained on standard datasets and fine-tune them for specific application domains with a small set of domain relevant images. In this work, we show that using synthetic datasets that are not necessarily photo-realistic can be a better alternative to simply fine-tune pre-trained networks. Specifically, our results show an impressive 25% improvement in the mAP metric over a fine-tuning baseline when only about 200 labelled images are available to train. Finally, an ablation study of our results is presented to delineate the individual contribution of different components in the randomization pipeline.
Tasks	Object Detection
Published	2018-07-16
URL	http://arxiv.org/abs/1807.09834v1
PDF	http://arxiv.org/pdf/1807.09834v1.pdf
PWC	https://paperswithcode.com/paper/applying-domain-randomization-to-synthetic
Repo	https://github.com/jsbruglie/tf-shape-detection
Framework	tf

Simpler but More Accurate Semantic Dependency Parsing


Title	Simpler but More Accurate Semantic Dependency Parsing
Authors	Timothy Dozat, Christopher D. Manning
Abstract	While syntactic dependency annotations concentrate on the surface or functional structure of a sentence, semantic dependency annotations aim to capture between-word relationships that are more closely related to the meaning of a sentence, using graph-structured representations. We extend the LSTM-based syntactic parser of Dozat and Manning (2017) to train on and generate these graph structures. The resulting system on its own achieves state-of-the-art performance, beating the previous, substantially more complex state-of-the-art system by 0.6% labeled F1. Adding linguistically richer input representations pushes the margin even higher, allowing us to beat it by 1.9% labeled F1.
Tasks	Dependency Parsing, Semantic Dependency Parsing
Published	2018-07-03
URL	http://arxiv.org/abs/1807.01396v1
PDF	http://arxiv.org/pdf/1807.01396v1.pdf
PWC	https://paperswithcode.com/paper/simpler-but-more-accurate-semantic-dependency
Repo	https://github.com/tdozat/Parser-v3
Framework	tf