October 19, 2019

2980 words 14 mins read

Paper Group ANR 130

Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation. PAC Learning Guarantees Under Covariate Shift. Measuring the Robustness of Graph Properties. An optimal approximation of discrete random variables with respect to the Kolmogorov distance. Recurrently Controlled Recurrent Net …

Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation


Title	Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation
Authors	Rakshith Shetty, Bernt Schiele, Mario Fritz
Abstract	Importance of visual context in scene understanding tasks is well recognized in the computer vision community. However, to what extent the computer vision models for image classification and semantic segmentation are dependent on the context to make their predictions is unclear. A model overly relying on context will fail when encountering objects in context distributions different from training data and hence it is important to identify these dependencies before we can deploy the models in the real-world. We propose a method to quantify the sensitivity of black-box vision models to visual context by editing images to remove selected objects and measuring the response of the target models. We apply this methodology on two tasks, image classification and semantic segmentation, and discover undesirable dependency between objects and context, for example that “sidewalk” segmentation relies heavily on “cars” being present in the image. We propose an object removal based data augmentation solution to mitigate this dependency and increase the robustness of classification and segmentation models to contextual variations. Our experiments show that the proposed data augmentation helps these models improve the performance in out-of-context scenarios, while preserving the performance on regular data.
Tasks	Data Augmentation, Image Classification, Scene Understanding, Semantic Segmentation
Published	2018-12-17
URL	http://arxiv.org/abs/1812.06707v1
PDF	http://arxiv.org/pdf/1812.06707v1.pdf
PWC	https://paperswithcode.com/paper/not-using-the-car-to-see-the-sidewalk
Repo
Framework

PAC Learning Guarantees Under Covariate Shift


Title	PAC Learning Guarantees Under Covariate Shift
Authors	Artidoro Pagnoni, Stefan Gramatovici, Samuel Liu
Abstract	We consider the Domain Adaptation problem, also known as the covariate shift problem, where the distributions that generate the training and test data differ while retaining the same labeling function. This problem occurs across a large range of practical applications, and is related to the more general challenge of transfer learning. Most recent work on the topic focuses on optimization techniques that are specific to an algorithm or practical use case rather than a more general approach. The sparse literature attempting to provide general bounds seems to suggest that efficient learning even under strong assumptions is not possible for covariate shift. Our main contribution is to recontextualize these results by showing that any Probably Approximately Correct (PAC) learnable concept class is still PAC learnable under covariate shift conditions with only a polynomial increase in the number of training samples. This approach essentially demonstrates that the Domain Adaptation learning problem is as hard as the underlying PAC learning problem, provided some conditions over the training and test distributions. We also present bounds for the rejection sampling algorithm, justifying it as a solution to the Domain Adaptation problem in certain scenarios.
Tasks	Domain Adaptation, Transfer Learning
Published	2018-12-16
URL	http://arxiv.org/abs/1812.06393v1
PDF	http://arxiv.org/pdf/1812.06393v1.pdf
PWC	https://paperswithcode.com/paper/pac-learning-guarantees-under-covariate-shift
Repo
Framework

Measuring the Robustness of Graph Properties


Title	Measuring the Robustness of Graph Properties
Authors	Yali Wan, Marina Meila
Abstract	In this paper, we propose a perturbation framework to measure the robustness of graph properties. Although there are already perturbation methods proposed to tackle this problem, they are limited by the fact that the strength of the perturbation cannot be well controlled. We firstly provide a perturbation framework on graphs by introducing weights on the nodes, of which the magnitude of perturbation can be easily controlled through the variance of the weights. Meanwhile, the topology of the graphs are also preserved to avoid uncontrollable strength in the perturbation. We then extend the measure of robustness in the robust statistics literature to the graph properties.
Tasks
Published	2018-12-03
URL	http://arxiv.org/abs/1901.09661v1
PDF	http://arxiv.org/pdf/1901.09661v1.pdf
PWC	https://paperswithcode.com/paper/measuring-the-robustness-of-graph-properties
Repo
Framework

An optimal approximation of discrete random variables with respect to the Kolmogorov distance


Title	An optimal approximation of discrete random variables with respect to the Kolmogorov distance
Authors	Liat Cohen, Dror Fried, Gera Weiss
Abstract	We present an algorithm that takes a discrete random variable $X$ and a number $m$ and computes a random variable whose support (set of possible outcomes) is of size at most $m$ and whose Kolmogorov distance from $X$ is minimal. In addition to a formal theoretical analysis of the correctness and of the computational complexity of the algorithm, we present a detailed empirical evaluation that shows how the proposed approach performs in practice in different applications and domains.
Tasks
Published	2018-05-19
URL	http://arxiv.org/abs/1805.07535v1
PDF	http://arxiv.org/pdf/1805.07535v1.pdf
PWC	https://paperswithcode.com/paper/an-optimal-approximation-of-discrete-random
Repo
Framework

Recurrently Controlled Recurrent Networks


Title	Recurrently Controlled Recurrent Networks
Authors	Yi Tay, Luu Anh Tuan, Siu Cheung Hui
Abstract	Recurrent neural networks (RNNs) such as long short-term memory and gated recurrent units are pivotal building blocks across a broad spectrum of sequence modeling problems. This paper proposes a recurrently controlled recurrent network (RCRN) for expressive and powerful sequence encoding. More concretely, the key idea behind our approach is to learn the recurrent gating functions using recurrent networks. Our architecture is split into two components - a controller cell and a listener cell whereby the recurrent controller actively influences the compositionality of the listener cell. We conduct extensive experiments on a myriad of tasks in the NLP domain such as sentiment analysis (SST, IMDb, Amazon reviews, etc.), question classification (TREC), entailment classification (SNLI, SciTail), answer selection (WikiQA, TrecQA) and reading comprehension (NarrativeQA). Across all 26 datasets, our results demonstrate that RCRN not only consistently outperforms BiLSTMs but also stacked BiLSTMs, suggesting that our controller architecture might be a suitable replacement for the widely adopted stacked architecture.
Tasks	Answer Selection, Reading Comprehension, Sentiment Analysis
Published	2018-11-24
URL	http://arxiv.org/abs/1811.09786v1
PDF	http://arxiv.org/pdf/1811.09786v1.pdf
PWC	https://paperswithcode.com/paper/recurrently-controlled-recurrent-networks
Repo
Framework

Structured Parallel Programming Language Based on True Concurrency


Title	Structured Parallel Programming Language Based on True Concurrency
Authors	Yong Wang
Abstract	Based on our previous work on algebraic laws for true concurrency, we design a skeleton of structured parallel programming language for true concurrency called SPPLTC. Different to most programming languages, SPPLTC has an explicit parallel operator as an essential operator. SPPLTC can structure a truly concurrent graph to a normal form. This means that it is possible to implement a compiler for SPPLTC.
Tasks
Published	2018-10-31
URL	http://arxiv.org/abs/1810.13446v1
PDF	http://arxiv.org/pdf/1810.13446v1.pdf
PWC	https://paperswithcode.com/paper/structured-parallel-programming-language
Repo
Framework

Conditional Transfer with Dense Residual Attention: Synthesizing traffic signs from street-view imagery


Title	Conditional Transfer with Dense Residual Attention: Synthesizing traffic signs from street-view imagery
Authors	Clint Sebastian, Ries Uittenbogaard, Julien Vijverberg, Bas Boom, Peter H. N. de With
Abstract	Object detection and classification of traffic signs in street-view imagery is an essential element for asset management, map making and autonomous driving. However, some traffic signs occur rarely and consequently, they are difficult to recognize automatically. To improve the detection and classification rates, we propose to generate images of traffic signs, which are then used to train a detector/classifier. In this research, we present an end-to-end framework that generates a realistic image of a traffic sign from a given image of a traffic sign and a pictogram of the target class. We propose a residual attention mechanism with dense concatenation called Dense Residual Attention, that preserves the background information while transferring the object information. We also propose to utilize multi-scale discriminators, so that the smaller scales of the output guide the higher resolution output. We have performed detection and classification tests across a large number of traffic sign classes, by training the detector using the combination of real and generated data. The newly trained model reduces the number of false positives by 1.2 - 1.5% at 99% recall in the detection tests and an absolute improvement of 4.65% (top-1 accuracy) in the classification tests.
Tasks	Autonomous Driving, Object Detection
Published	2018-09-05
URL	http://arxiv.org/abs/1809.01444v1
PDF	http://arxiv.org/pdf/1809.01444v1.pdf
PWC	https://paperswithcode.com/paper/conditional-transfer-with-dense-residual
Repo
Framework

Multi-Task Generative Adversarial Nets with Shared Memory for Cross-Domain Coordination Control


Title	Multi-Task Generative Adversarial Nets with Shared Memory for Cross-Domain Coordination Control
Authors	JunPing Wang, WenSheng Zhang, Ian Thomas, ShiHui Duan, YouKang Shi
Abstract	Generating sequential decision process from huge amounts of measured process data is a future research direction for collaborative factory automation, making full use of those online or offline process data to directly design flexible make decisions policy, and evaluate performance. The key challenges for the sequential decision process is to online generate sequential decision-making policy directly, and transferring knowledge across tasks domain. Most multi-task policy generating algorithms often suffer from insufficient generating cross-task sharing structure at discrete-time nonlinear systems with applications. This paper proposes the multi-task generative adversarial nets with shared memory for cross-domain coordination control, which can generate sequential decision policy directly from raw sensory input of all of tasks, and online evaluate performance of system actions in discrete-time nonlinear systems. Experiments have been undertaken using a professional flexible manufacturing testbed deployed within a smart factory of Weichai Power in China. Results on three groups of discrete-time nonlinear control tasks show that our proposed model can availably improve the performance of task with the help of other related tasks.
Tasks	Decision Making
Published	2018-07-01
URL	http://arxiv.org/abs/1807.00298v1
PDF	http://arxiv.org/pdf/1807.00298v1.pdf
PWC	https://paperswithcode.com/paper/multi-task-generative-adversarial-nets-with
Repo
Framework

Zero-Shot Question Generation from Knowledge Graphs for Unseen Predicates and Entity Types


Title	Zero-Shot Question Generation from Knowledge Graphs for Unseen Predicates and Entity Types
Authors	Hady Elsahar, Christophe Gravier, Frederique Laforest
Abstract	We present a neural model for question generation from knowledge base triples in a “Zero-Shot” setup, that is generating questions for triples containing predicates, subject types or object types that were not seen at training time. Our model leverages triples occurrences in the natural language corpus in an encoder-decoder architecture, paired with an original part-of-speech copy action mechanism to generate questions. Benchmark and human evaluation show that our model sets a new state-of-the-art for zero-shot QG.
Tasks	Knowledge Graphs, Question Generation
Published	2018-02-19
URL	http://arxiv.org/abs/1802.06842v1
PDF	http://arxiv.org/pdf/1802.06842v1.pdf
PWC	https://paperswithcode.com/paper/zero-shot-question-generation-from-knowledge
Repo
Framework

Improving Chemical Autoencoder Latent Space and Molecular De novo Generation Diversity with Heteroencoders


Title	Improving Chemical Autoencoder Latent Space and Molecular De novo Generation Diversity with Heteroencoders
Authors	Esben Jannik Bjerrum, Boris Sattarov
Abstract	Chemical autoencoders are attractive models as they combine chemical space navigation with possibilities for de-novo molecule generation in areas of interest. This enables them to produce focused chemical libraries around a single lead compound for employment early in a drug discovery project. Here it is shown that the choice of chemical representation, such as SMILES strings, has a large influence on the properties of the latent space. It is further explored to what extent translating between different chemical representations influences the latent space similarity to the SMILES strings or circular fingerprints. By employing SMILES enumeration for either the encoder or decoder, it is found that the decoder has the largest influence on the properties of the latent space. Training a sequence to sequence heteroencoder based on recurrent neural networks(RNNs) with long short-term memory cells (LSTM) to predict different enumerated SMILES strings from the same canonical SMILES string gives the largest similarity between latent space distance and molecular similarity measured as circular fingerprints similarity. Using the output from the bottleneck in QSAR modelling of five molecular datasets shows that heteroencoder derived vectors markedly outperforms autoencoder derived vectors as well as models built using ECFP4 fingerprints, underlining the increased chemical relevance of the latent space. However, the use of enumeration during training of the decoder leads to a markedly increase in the rate of decoding to a different molecules than encoded, a tendency that can be counteracted with more complex network architectures.
Tasks	Drug Discovery
Published	2018-06-25
URL	http://arxiv.org/abs/1806.09300v2
PDF	http://arxiv.org/pdf/1806.09300v2.pdf
PWC	https://paperswithcode.com/paper/improving-chemical-autoencoder-latent-space
Repo
Framework

Stellar Cluster Detection using GMM with Deep Variational Autoencoder


Title	Stellar Cluster Detection using GMM with Deep Variational Autoencoder
Authors	Arnab Karmakar, Deepak Mishra, Anandmayee Tej
Abstract	Detecting stellar clusters have always been an important research problem in Astronomy. Although images do not convey very detailed information in detecting stellar density enhancements, we attempt to understand if new machine learning techniques can reveal patterns that would assist in drawing better inferences from the available image data. This paper describes an unsupervised approach in detecting star clusters using Deep Variational Autoencoder combined with a Gaussian Mixture Model. We show that our method works significantly well in comparison with state-of-the-art detection algorithm in recognizing a variety of star clusters even in the presence of noise and distortion.
Tasks
Published	2018-09-05
URL	http://arxiv.org/abs/1809.01434v1
PDF	http://arxiv.org/pdf/1809.01434v1.pdf
PWC	https://paperswithcode.com/paper/stellar-cluster-detection-using-gmm-with-deep
Repo
Framework

Learning a Single Tucker Decomposition Network for Lossy Image Compression with Multiple Bits-Per-Pixel Rates


Title	Learning a Single Tucker Decomposition Network for Lossy Image Compression with Multiple Bits-Per-Pixel Rates
Authors	Jianrui Cai, Zisheng Cao, Lei Zhang
Abstract	Lossy image compression (LIC), which aims to utilize inexact approximations to represent an image more compactly, is a classical problem in image processing. Recently, deep convolutional neural networks (CNNs) have achieved interesting results in LIC by learning an encoder-quantizer-decoder network from a large amount of data. However, existing CNN-based LIC methods usually can only train a network for a specific bits-per-pixel (bpp). Such a “one network per bpp” problem limits the generality and flexibility of CNNs to practical LIC applications. In this paper, we propose to learn a single CNN which can perform LIC at multiple bpp rates. A simple yet effective Tucker Decomposition Network (TDNet) is developed, where there is a novel tucker decomposition layer (TDL) to decompose a latent image representation into a set of projection matrices and a core tensor. By changing the rank of the core tensor and its quantization, we can easily adjust the bpp rate of latent image representation within a single CNN. Furthermore, an iterative non-uniform quantization scheme is presented to optimize the quantizer, and a coarse-to-fine training strategy is introduced to reconstruct the decompressed images. Extensive experiments demonstrate the state-of-the-art compression performance of TDNet in terms of both PSNR and MS-SSIM indices.
Tasks	Image Compression, Quantization
Published	2018-07-10
URL	http://arxiv.org/abs/1807.03470v1
PDF	http://arxiv.org/pdf/1807.03470v1.pdf
PWC	https://paperswithcode.com/paper/learning-a-single-tucker-decomposition
Repo
Framework

Understanding the Role of Two-Sided Argumentation in Online Consumer Reviews: A Language-Based Perspective


Title	Understanding the Role of Two-Sided Argumentation in Online Consumer Reviews: A Language-Based Perspective
Authors	Bernhard Lutz, Nicolas Pröllochs, Dirk Neumann
Abstract	This paper examines the effect of two-sided argumentation on the perceived helpfulness of online consumer reviews. In contrast to previous works, our analysis thereby sheds light on the reception of reviews from a language-based perspective. For this purpose, we propose an intriguing text analysis approach based on distributed text representations and multi-instance learning to operationalize the two-sidedness of argumentation in review texts. A subsequent empirical analysis using a large corpus of Amazon reviews suggests that two-sided argumentation in reviews significantly increases their helpfulness. We find this effect to be stronger for positive reviews than for negative reviews, whereas a higher degree of emotional language weakens the effect. Our findings have immediate implications for retailer platforms, which can utilize our results to optimize their customer feedback system and to present more useful product reviews.
Tasks
Published	2018-10-25
URL	http://arxiv.org/abs/1810.10942v2
PDF	http://arxiv.org/pdf/1810.10942v2.pdf
PWC	https://paperswithcode.com/paper/understanding-the-role-of-two-sided
Repo
Framework

An Unsupervised Approach for Aspect Category Detection Using Soft Cosine Similarity Measure


Title	An Unsupervised Approach for Aspect Category Detection Using Soft Cosine Similarity Measure
Authors	Erfan Ghadery, Sajad Movahedi, Heshaam Faili, Azadeh Shakery
Abstract	Aspect category detection is one of the important and challenging subtasks of aspect-based sentiment analysis. Given a set of pre-defined categories, this task aims to detect categories which are indicated implicitly or explicitly in a given review sentence. Supervised machine learning approaches perform well to accomplish this subtask. Note that, the performance of these methods depends on the availability of labeled train data, which is often difficult and costly to obtain. Besides, most of these supervised methods require feature engineering to perform well. In this paper, we propose an unsupervised method to address aspect category detection task without the need for any feature engineering. Our method utilizes clusters of unlabeled reviews and soft cosine similarity measure to accomplish aspect category detection task. Experimental results on SemEval-2014 restaurant dataset shows that proposed unsupervised approach outperforms several baselines by a substantial margin.
Tasks	Aspect-Based Sentiment Analysis, Feature Engineering, Sentiment Analysis
Published	2018-12-08
URL	https://arxiv.org/abs/1812.03361v2
PDF	https://arxiv.org/pdf/1812.03361v2.pdf
PWC	https://paperswithcode.com/paper/an-unsupervised-approach-for-aspect-category
Repo
Framework

Universal Approximation with Quadratic Deep Networks


Title	Universal Approximation with Quadratic Deep Networks
Authors	Fenglei Fan, Jinjun Xiong, Ge Wang
Abstract	Recently, deep learning has achieved huge successes in many important applications. In our previous studies, we proposed quadratic/second-order neurons and deep quadratic neural networks. In a quadratic neuron, the inner product of a vector of data and the corresponding weights in a conventional neuron is replaced with a quadratic function. The resultant quadratic neuron enjoys an enhanced expressive capability over the conventional neuron. However, how quadratic neurons improve the expressing capability of a deep quadratic network has not been studied up to now, preferably in relation to that of a conventional neural network. Regarding this, we ask four basic questions in this paper: (1) for the one-hidden-layer network structure, is there any function that a quadratic network can approximate much more efficiently than a conventional network? (2) for the same multi-layer network structure, is there any function that can be expressed by a quadratic network but cannot be expressed with conventional neurons in the same structure? (3) Does a quadratic network give a new insight into universal approximation? (4) To approximate the same class of functions with the same error bound, is a quantized quadratic network able to enjoy a lower number of weights than a quantized conventional network? Our main contributions are the four interconnected theorems shedding light upon these four questions and demonstrating the merits of a quadratic network in terms of expressive efficiency, unique capability, compact architecture and computational capacity respectively.
Tasks	Speech Recognition
Published	2018-07-31
URL	https://arxiv.org/abs/1808.00098v3
PDF	https://arxiv.org/pdf/1808.00098v3.pdf
PWC	https://paperswithcode.com/paper/universal-approximation-with-quadratic-deep
Repo
Framework