January 30, 2020

3012 words 15 mins read

Paper Group ANR 474

UGAN: Untraceable GAN for Multi-Domain Face Translation. Information Maximizing Visual Question Generation. A Deep Learning Model for Chilean Bills Classification. Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites. Genetic Algorithm based Multi-Objective Optimization of Solidification in Die Casting using Deep Neural Network a …

UGAN: Untraceable GAN for Multi-Domain Face Translation


Title	UGAN: Untraceable GAN for Multi-Domain Face Translation
Authors	Defa Zhu, Si Liu, Wentao Jiang, Chen Gao, Tianyi Wu, Qaingchang Wang, Guodong Guo
Abstract	The multi-domain image-to-image translation is a challenging task where the goal is to translate an image into multiple different domains. The target-only characteristics are desired for translated images, while the source-only characteristics should be erased. However, recent methods often suffer from retaining the characteristics of the source domain, which are incompatible with the target domain. To address this issue, we propose a method called Untraceable GAN, which has a novel source classifier to differentiate which domain an image is translated from, and determines whether the translated image still retains the characteristics of the source domain. Furthermore, we take the prototype of the target domain as the guidance for the translator to effectively synthesize the target-only characteristics. The translator is learned to synthesize the target-only characteristics and make the source domain untraceable for the discriminator, so that the source-only characteristics are erased. Finally, extensive experiments on three face editing tasks, including face aging, makeup, and expression editing, show that the proposed UGAN can produce superior results over the state-of-the-art models. The source code will be released.
Tasks	Image-to-Image Translation
Published	2019-07-26
URL	https://arxiv.org/abs/1907.11418v2
PDF	https://arxiv.org/pdf/1907.11418v2.pdf
PWC	https://paperswithcode.com/paper/ugan-untraceable-gan-for-multi-domain-face
Repo
Framework

Information Maximizing Visual Question Generation


Title	Information Maximizing Visual Question Generation
Authors	Ranjay Krishna, Michael Bernstein, Li Fei-Fei
Abstract	Though image-to-sequence generation models have become overwhelmingly popular in human-computer communications, they suffer from strongly favoring safe generic questions (“What is in this picture?"). Generating uninformative but relevant questions is not sufficient or useful. We argue that a good question is one that has a tightly focused purpose — one that is aimed at expecting a specific type of response. We build a model that maximizes mutual information between the image, the expected answer and the generated question. To overcome the non-differentiability of discrete natural language tokens, we introduce a variational continuous latent space onto which the expected answers project. We regularize this latent space with a second latent space that ensures clustering of similar answers. Even when we don’t know the expected answer, this second latent space can generate goal-driven questions specifically aimed at extracting objects (“what is the person throwing”), attributes, (“What kind of shirt is the person wearing?"), color (“what color is the frisbee?"), material (“What material is the frisbee?"), etc. We quantitatively show that our model is able to retain information about an expected answer category, resulting in more diverse, goal-driven questions. We launch our model on a set of real world images and extract previously unseen visual concepts.
Tasks	Question Generation
Published	2019-03-27
URL	http://arxiv.org/abs/1903.11207v1
PDF	http://arxiv.org/pdf/1903.11207v1.pdf
PWC	https://paperswithcode.com/paper/information-maximizing-visual-question
Repo
Framework

A Deep Learning Model for Chilean Bills Classification


Title	A Deep Learning Model for Chilean Bills Classification
Authors	Daniel San Martin, Daniel Manzano
Abstract	Automatic bill classification is an attractive task with many potential applications such as automated detection and counting in images or videos. To address this purpose we present a Deep Learning Model to classify Chilean Banknotes, because of its successful results in image processing applications. For optimal performance of the proposed model, data augmentation techniques are introduced due to the limited number of image samples. Positive results were achieved in this work, verifying that it could be a stating point to be extended to more complex applications.
Tasks	Data Augmentation
Published	2019-12-21
URL	https://arxiv.org/abs/1912.12120v1
PDF	https://arxiv.org/pdf/1912.12120v1.pdf
PWC	https://paperswithcode.com/paper/a-deep-learning-model-for-chilean-bills
Repo
Framework

Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites


Title	Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites
Authors	Alexey Tikhonov, Viacheslav Shibaev, Aleksander Nagaev, Aigul Nugmanova, Ivan P. Yamshchikov
Abstract	This paper shows that standard assessment methodology for style transfer has several significant problems. First, the standard metrics for style accuracy and semantics preservation vary significantly on different re-runs. Therefore one has to report error margins for the obtained results. Second, starting with certain values of bilingual evaluation understudy (BLEU) between input and output and accuracy of the sentiment transfer the optimization of these two standard metrics diverge from the intuitive goal of the style transfer task. Finally, due to the nature of the task itself, there is a specific dependence between these two metrics that could be easily manipulated. Under these circumstances, we suggest taking BLEU between input and human-written reformulations into consideration for benchmarks. We also propose three new architectures that outperform state of the art in terms of this metric.
Tasks	Style Transfer
Published	2019-08-19
URL	https://arxiv.org/abs/1908.06809v2
PDF	https://arxiv.org/pdf/1908.06809v2.pdf
PWC	https://paperswithcode.com/paper/style-transfer-for-texts-to-err-is-human-but
Repo
Framework

Genetic Algorithm based Multi-Objective Optimization of Solidification in Die Casting using Deep Neural Network as Surrogate Model


Title	Genetic Algorithm based Multi-Objective Optimization of Solidification in Die Casting using Deep Neural Network as Surrogate Model
Authors	Shantanu Shahane, Narayana Aluru, Placid Ferreira, Shiv G Kapoor, Surya Pratap Vanka
Abstract	In this paper, a novel strategy of multi-objective optimization of die casting is presented. The cooling of molten metal inside the mold is achieved by passing a coolant, typically water through the cooling lines in the die. Depending on the cooling line location, coolant flow rate and die geometry, nonuniform temperatures are imposed on the molten metal at the mold wall. This boundary condition along with the initial molten metal temperature affect the product quality quantified in terms of micro-structure parameters and yield strength. A finite volume based numerical solver is used to correlate the inputs to outputs. The objective of this research is to estimate the initial and wall temperatures so as to optimize the product quality. The non-dominated sorting genetic algorithm (NSGA–II) which is popular for solving multi-objective optimization problems is used. The number of function evaluations required for NSGA–II can be of the order of millions and hence, the finite volume solver cannot be used directly for optimization. Thus, a neural network trained using the results from the numerical solver is used as a surrogate model. Simplified versions of the actual problem are designed to verify results of the genetic algorithm. An innovative local sensitivity based approach is used to rank the final Pareto optimal solutions and choose a single best design.
Tasks
Published	2019-01-08
URL	http://arxiv.org/abs/1901.02364v1
PDF	http://arxiv.org/pdf/1901.02364v1.pdf
PWC	https://paperswithcode.com/paper/genetic-algorithm-based-multi-objective
Repo
Framework

Topological based classification of paper domains using graph convolutional networks


Title	Topological based classification of paper domains using graph convolutional networks
Authors	Idan Benami, Keren Cohen, Oved Nagar, Yoram Louzoun
Abstract	The main approaches for node classification in graphs are information propagation and the association of the class of the node with external information. State of the art methods merge these approaches through Graph Convolutional Networks. We here use the association of topological features of the nodes with their class to predict this class. Moreover, combining topological information with information propagation improves classification accuracy on the standard CiteSeer and Cora paper classification task. Topological features and information propagation produce results almost as good as text-based classification, without no textual or content information. We propose to represent the topology and information propagation through a GCN with the neighboring training node classification as an input and the current node classification as output. Such a formalism outperforms state of the art methods.
Tasks	Node Classification
Published	2019-04-10
URL	http://arxiv.org/abs/1904.07787v1
PDF	http://arxiv.org/pdf/1904.07787v1.pdf
PWC	https://paperswithcode.com/paper/190407787
Repo
Framework

Accuracy Booster: Performance Boosting using Feature Map Re-calibration


Title	Accuracy Booster: Performance Boosting using Feature Map Re-calibration
Authors	Pravendra Singh, Pratik Mazumder, Vinay P. Namboodiri
Abstract	Convolution Neural Networks (CNN) have been extremely successful in solving intensive computer vision tasks. The convolutional filters used in CNNs have played a major role in this success, by extracting useful features from the inputs. Recently researchers have tried to boost the performance of CNNs by re-calibrating the feature maps produced by these filters, e.g., Squeeze-and-Excitation Networks (SENets). These approaches have achieved better performance by Exciting up the important channels or feature maps while diminishing the rest. However, in the process, architectural complexity has increased. We propose an architectural block that introduces much lower complexity than the existing methods of CNN performance boosting while performing significantly better than them. We carry out experiments on the CIFAR, ImageNet and MS-COCO datasets, and show that the proposed block can challenge the state-of-the-art results. Our method boosts the ResNet-50 architecture to perform comparably to the ResNet-152 architecture, which is a three times deeper network, on classification. We also show experimentally that our method is not limited to classification but also generalizes well to other tasks such as object detection.
Tasks	Calibration, Object Detection
Published	2019-03-11
URL	https://arxiv.org/abs/1903.04407v2
PDF	https://arxiv.org/pdf/1903.04407v2.pdf
PWC	https://paperswithcode.com/paper/accuracy-booster-performance-boosting-using
Repo
Framework

Assessing Transferability from Simulation to Reality for Reinforcement Learning


Title	Assessing Transferability from Simulation to Reality for Reinforcement Learning
Authors	Fabio Muratore, Michael Gienger, Jan Peters
Abstract	Learning robot control policies from physics simulations is of great interest to the robotics community as it may render the learning process faster, cheaper, and safer by alleviating the need for expensive real-world experiments. However, the direct transfer of learned behavior from simulation to reality is a major challenge. Optimizing a policy on a slightly faulty simulator can easily lead to the maximization of the `Simulation Optimization Bias` (SOB). In this case, the optimizer exploits modeling errors of the simulator such that the resulting behavior can potentially damage the robot. We tackle this challenge by applying domain randomization, i.e., randomizing the parameters of the physics simulations during learning. We propose an algorithm called Simulation-based Policy Optimization with Transferability Assessment (SPOTA) which uses an estimator of the SOB to formulate a stopping criterion for training. The introduced estimator quantifies the over-fitting to the set of domains experienced while training. Our experimental results on two different second order nonlinear systems show that the new simulation-based policy search algorithm is able to learn a control policy exclusively from a randomized simulator, which can be applied directly to real systems without any additional training.
Tasks
Published	2019-07-10
URL	https://arxiv.org/abs/1907.04685v2
PDF	https://arxiv.org/pdf/1907.04685v2.pdf
PWC	https://paperswithcode.com/paper/assessing-transferability-from-simulation-to
Repo
Framework

A Multivariate Model for Representing Semantic Non-compositionality


Title	A Multivariate Model for Representing Semantic Non-compositionality
Authors	Meghdad Farahmand
Abstract	Semantically non-compositional phrases constitute an intriguing research topic in Natural Language Processing. Semantic non-compositionality –the situation when the meaning of a phrase cannot be derived from the meaning of its components, is the main characteristic of such phrases, however, they bear other characteristics such as high statistical association and non-substitutability. In this work, we present a model for identifying non-compositional phrases that takes into account all of these characteristics. We show that the presented model remarkably outperforms the existing models of identifying non-compositional phrases that mostly focus only on one of these characteristics.
Tasks
Published	2019-08-15
URL	https://arxiv.org/abs/1908.05490v1
PDF	https://arxiv.org/pdf/1908.05490v1.pdf
PWC	https://paperswithcode.com/paper/a-multivariate-model-for-representing
Repo
Framework

Parameterized Structured Pruning for Deep Neural Networks


Title	Parameterized Structured Pruning for Deep Neural Networks
Authors	Guenther Schindler, Wolfgang Roth, Franz Pernkopf, Holger Froening
Abstract	As a result of the growing size of Deep Neural Networks (DNNs), the gap to hardware capabilities in terms of memory and compute increases. To effectively compress DNNs, quantization and connection pruning are usually considered. However, unconstrained pruning usually leads to unstructured parallelism, which maps poorly to massively parallel processors, and substantially reduces the efficiency of general-purpose processors. Similar applies to quantization, which often requires dedicated hardware. We propose Parameterized Structured Pruning (PSP), a novel method to dynamically learn the shape of DNNs through structured sparsity. PSP parameterizes structures (e.g. channel- or layer-wise) in a weight tensor and leverages weight decay to learn a clear distinction between important and unimportant structures. As a result, PSP maintains prediction performance, creates a substantial amount of sparsity that is structured and, thus, easy and efficient to map to a variety of massively parallel processors, which are mandatory for utmost compute power and energy efficiency. PSP is experimentally validated on the popular CIFAR10/100 and ILSVRC2012 datasets using ResNet and DenseNet architectures, respectively.
Tasks	Quantization
Published	2019-06-12
URL	https://arxiv.org/abs/1906.05180v1
PDF	https://arxiv.org/pdf/1906.05180v1.pdf
PWC	https://paperswithcode.com/paper/parameterized-structured-pruning-for-deep
Repo
Framework

Total stochastic gradient algorithms and applications in reinforcement learning


Title	Total stochastic gradient algorithms and applications in reinforcement learning
Authors	Paavo Parmas
Abstract	Backpropagation and the chain rule of derivatives have been prominent; however, the total derivative rule has not enjoyed the same amount of attention. In this work we show how the total derivative rule leads to an intuitive visual framework for creating gradient estimators on graphical models. In particular, previous “policy gradient theorems” are easily derived. We derive new gradient estimators based on density estimation, as well as a likelihood ratio gradient, which “jumps” to an intermediate node, not directly to the objective function. We evaluate our methods on model-based policy gradient algorithms, achieve good performance, and present evidence towards demystifying the success of the popular PILCO algorithm.
Tasks	Density Estimation
Published	2019-02-05
URL	http://arxiv.org/abs/1902.01722v1
PDF	http://arxiv.org/pdf/1902.01722v1.pdf
PWC	https://paperswithcode.com/paper/total-stochastic-gradient-algorithms-and
Repo
Framework

Theoretical Guarantees for Model Auditing with Finite Adversaries


Title	Theoretical Guarantees for Model Auditing with Finite Adversaries
Authors	Mario Diaz, Peter Kairouz, Jiachun Liao, Lalitha Sankar
Abstract	Privacy concerns have led to the development of privacy-preserving approaches for learning models from sensitive data. Yet, in practice, even models learned with privacy guarantees can inadvertently memorize unique training examples or leak sensitive features. To identify such privacy violations, existing model auditing techniques use finite adversaries defined as machine learning models with (a) access to some finite side information (e.g., a small auditing dataset), and (b) finite capacity (e.g., a fixed neural network architecture). Our work investigates the requirements under which an unsuccessful attempt to identify privacy violations by a finite adversary implies that no stronger adversary can succeed at such a task. We do so via parameters that quantify the capabilities of the finite adversary, including the size of the neural network employed by such an adversary and the amount of side information it has access to as well as the regularity of the (perhaps privacy-guaranteeing) audited model.
Tasks
Published	2019-11-08
URL	https://arxiv.org/abs/1911.03405v1
PDF	https://arxiv.org/pdf/1911.03405v1.pdf
PWC	https://paperswithcode.com/paper/theoretical-guarantees-for-model-auditing
Repo
Framework

A Comparative Study of Filtering Approaches Applied to Color Archival Document Images


Title	A Comparative Study of Filtering Approaches Applied to Color Archival Document Images
Authors	Walid Elhedda, Maroua Mehri, Mohamed Ali Mahjoub
Abstract	Current systems used by the Tunisian national archives for the automatic transcription of archival documents are hindered by many issues related to the performance of the optical character recognition (OCR) tools. Indeed, using a classical OCR system to transcribe and index ancient Arabic documents is not a straightforward task due to the idiosyncrasies of this category of documents, such as noise and degradation. Thus, applying an enhancement method or a denoising technique remains an essential prerequisite step to ease the archival document image analysis task. The state-of-the-art methods addressing the use of degraded document image enhancement and denoising are mainly based on applying filters. The most common filtering techniques applied to color images in the literature may be categorized into four approaches: scalar, marginal, vector and hybrid. To provide a set of comprehensive guidelines on the strengths and weaknesses of these filtering approaches, a thorough comparative study is proposed in this article. Numerical experiments are carried out in this study on color archival document images to show and quantify the performance of each assessed filtering approach.
Tasks	Denoising, Image Enhancement, Optical Character Recognition
Published	2019-08-16
URL	https://arxiv.org/abs/1908.09007v1
PDF	https://arxiv.org/pdf/1908.09007v1.pdf
PWC	https://paperswithcode.com/paper/a-comparative-study-of-filtering-approaches
Repo
Framework

An Annotation Saved is an Annotation Earned: Using Fully Synthetic Training for Object Instance Detection


Title	An Annotation Saved is an Annotation Earned: Using Fully Synthetic Training for Object Instance Detection
Authors	Stefan Hinterstoisser, Olivier Pauly, Hauke Heibel, Martina Marek, Martin Bokeloh
Abstract	Deep learning methods typically require vast amounts of training data to reach their full potential. While some publicly available datasets exists, domain specific data always needs to be collected and manually labeled, an expensive, time consuming and error prone process. Training with synthetic data is therefore very lucrative, as dataset creation and labeling comes for free. We propose a novel method for creating purely synthetic training data for object detection. We leverage a large dataset of 3D background models and densely render them using full domain randomization. This yields background images with realistic shapes and texture on top of which we render the objects of interest. During training, the data generation process follows a curriculum strategy guaranteeing that all foreground models are presented to the network equally under all possible poses and conditions with increasing complexity. As a result, we entirely control the underlying statistics and we create optimal training samples at every stage of training. Using a set of 64 retail objects, we demonstrate that our simple approach enables the training of detectors that outperform models trained with real data on a challenging evaluation dataset.
Tasks	Object Detection
Published	2019-02-26
URL	http://arxiv.org/abs/1902.09967v1
PDF	http://arxiv.org/pdf/1902.09967v1.pdf
PWC	https://paperswithcode.com/paper/an-annotation-saved-is-an-annotation-earned
Repo
Framework

Bounded rational decision-making from elementary computations that reduce uncertainty


Title	Bounded rational decision-making from elementary computations that reduce uncertainty
Authors	Sebastian Gottwald, Daniel A. Braun
Abstract	In its most basic form, decision-making can be viewed as a computational process that progressively eliminates alternatives, thereby reducing uncertainty. Such processes are generally costly, meaning that the amount of uncertainty that can be reduced is limited by the amount of available computational resources. Here, we introduce the notion of elementary computation based on a fundamental principle for probability transfers that reduce uncertainty. Elementary computations can be considered as the inverse of Pigou-Dalton transfers applied to probability distributions, closely related to the concepts of majorization, T-transforms, and generalized entropies that induce a preorder on the space of probability distributions. As a consequence we can define resource cost functions that are order-preserving and therefore monotonic with respect to the uncertainty reduction. This leads to a comprehensive notion of decision-making processes with limited resources. Along the way, we prove several new results on majorization theory, as well as on entropy and divergence measures.
Tasks	Decision Making
Published	2019-04-08
URL	http://arxiv.org/abs/1904.03964v1
PDF	http://arxiv.org/pdf/1904.03964v1.pdf
PWC	https://paperswithcode.com/paper/bounded-rational-decision-making-from
Repo
Framework