July 29, 2019

3608 words 17 mins read

Paper Group AWR 107

Learning Explanatory Rules from Noisy Data. Generation of High Dynamic Range Illumination from a Single Image for the Enhancement of Undesirably Illuminated Images. GeneGAN: Learning Object Transfiguration and Attribute Subspace from Unpaired Data. Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork. Learning Deep Nearest Neighbo …

Learning Explanatory Rules from Noisy Data


Title	Learning Explanatory Rules from Noisy Data
Authors	Richard Evans, Edward Grefenstette
Abstract	Artificial Neural Networks are powerful function approximators capable of modelling solutions to a wide variety of problems, both supervised and unsupervised. As their size and expressivity increases, so too does the variance of the model, yielding a nearly ubiquitous overfitting problem. Although mitigated by a variety of model regularisation methods, the common cure is to seek large amounts of training data—which is not necessarily easily obtained—that sufficiently approximates the data distribution of the domain we wish to test on. In contrast, logic programming methods such as Inductive Logic Programming offer an extremely data-efficient process by which models can be trained to reason on symbolic domains. However, these methods are unable to deal with the variety of domains neural networks can be applied to: they are not robust to noise in or mislabelling of inputs, and perhaps more importantly, cannot be applied to non-symbolic domains where the data is ambiguous, such as operating on raw pixels. In this paper, we propose a Differentiable Inductive Logic framework, which can not only solve tasks which traditional ILP systems are suited for, but shows a robustness to noise and error in the training data which ILP cannot cope with. Furthermore, as it is trained by backpropagation against a likelihood objective, it can be hybridised by connecting it with neural networks over ambiguous data in order to be applied to domains which ILP cannot address, while providing data efficiency and generalisation beyond what neural networks on their own can achieve.
Tasks
Published	2017-11-13
URL	http://arxiv.org/abs/1711.04574v2
PDF	http://arxiv.org/pdf/1711.04574v2.pdf
PWC	https://paperswithcode.com/paper/learning-explanatory-rules-from-noisy-data
Repo	https://github.com/ai-systems/DILP-Core
Framework	tf

Generation of High Dynamic Range Illumination from a Single Image for the Enhancement of Undesirably Illuminated Images


Title	Generation of High Dynamic Range Illumination from a Single Image for the Enhancement of Undesirably Illuminated Images
Authors	Jae Sung Park, Nam Ik Cho
Abstract	This paper presents an algorithm that enhances undesirably illuminated images by generating and fusing multi-level illuminations from a single image.The input image is first decomposed into illumination and reflectance components by using an edge-preserving smoothing filter. Then the reflectance component is scaled up to improve the image details in bright areas. The illumination component is scaled up and down to generate several illumination images that correspond to certain camera exposure values different from the original. The virtual multi-exposure illuminations are blended into an enhanced illumination, where we also propose a method to generate appropriate weight maps for the tone fusion. Finally, an enhanced image is obtained by multiplying the equalized illumination and enhanced reflectance. Experiments show that the proposed algorithm produces visually pleasing output and also yields comparable objective results to the conventional enhancement methods, while requiring modest computational loads.
Tasks
Published	2017-08-02
URL	http://arxiv.org/abs/1708.00636v1
PDF	http://arxiv.org/pdf/1708.00636v1.pdf
PWC	https://paperswithcode.com/paper/generation-of-high-dynamic-range-illumination
Repo	https://github.com/JasonBournePark/EnhanceHDR
Framework	none

GeneGAN: Learning Object Transfiguration and Attribute Subspace from Unpaired Data


Title	GeneGAN: Learning Object Transfiguration and Attribute Subspace from Unpaired Data
Authors	Shuchang Zhou, Taihong Xiao, Yi Yang, Dieqiao Feng, Qinyao He, Weiran He
Abstract	Object Transfiguration replaces an object in an image with another object from a second image. For example it can perform tasks like “putting exactly those eyeglasses from image A on the nose of the person in image B”. Usage of exemplar images allows more precise specification of desired modifications and improves the diversity of conditional image generation. However, previous methods that rely on feature space operations, require paired data and/or appearance models for training or disentangling objects from background. In this work, we propose a model that can learn object transfiguration from two unpaired sets of images: one set containing images that “have” that kind of object, and the other set being the opposite, with the mild constraint that the objects be located approximately at the same place. For example, the training data can be one set of reference face images that have eyeglasses, and another set of images that have not, both of which spatially aligned by face landmarks. Despite the weak 0/1 labels, our model can learn an “eyeglasses” subspace that contain multiple representatives of different types of glasses. Consequently, we can perform fine-grained control of generated images, like swapping the glasses in two images by swapping the projected components in the “eyeglasses” subspace, to create novel images of people wearing eyeglasses. Overall, our deterministic generative model learns disentangled attribute subspaces from weakly labeled data by adversarial training. Experiments on CelebA and Multi-PIE datasets validate the effectiveness of the proposed model on real world data, in generating images with specified eyeglasses, smiling, hair styles, and lighting conditions etc. The code is available online.
Tasks	Conditional Image Generation, Image Generation
Published	2017-05-14
URL	http://arxiv.org/abs/1705.04932v1
PDF	http://arxiv.org/pdf/1705.04932v1.pdf
PWC	https://paperswithcode.com/paper/genegan-learning-object-transfiguration-and
Repo	https://github.com/Prinsphield/GeneGAN
Framework	tf

Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork


Title	Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork
Authors	Wei Ren Tan, Chee Seng Chan, Hernan Aguirre, Kiyoshi Tanaka
Abstract	This paper proposes a series of new approaches to improve Generative Adversarial Network (GAN) for conditional image synthesis and we name the proposed model as ArtGAN. One of the key innovation of ArtGAN is that, the gradient of the loss function w.r.t. the label (randomly assigned to each generated image) is back-propagated from the categorical discriminator to the generator. With the feedback from the label information, the generator is able to learn more efficiently and generate image with better quality. Inspired by recent works, an autoencoder is incorporated into the categorical discriminator for additional complementary information. Last but not least, we introduce a novel strategy to improve the image quality. In the experiments, we evaluate ArtGAN on CIFAR-10 and STL-10 via ablation studies. The empirical results showed that our proposed model outperforms the state-of-the-art results on CIFAR-10 in terms of Inception score. Qualitatively, we demonstrate that ArtGAN is able to generate plausible-looking images on Oxford-102 and CUB-200, as well as able to draw realistic artworks based on style, artist, and genre. The source code and models are available at: https://github.com/cs-chan/ArtGAN
Tasks	Conditional Image Generation, Image Generation
Published	2017-08-31
URL	http://arxiv.org/abs/1708.09533v2
PDF	http://arxiv.org/pdf/1708.09533v2.pdf
PWC	https://paperswithcode.com/paper/improved-artgan-for-conditional-synthesis-of
Repo	https://github.com/cs-chan/Artwork-Synthesis
Framework	tf

Learning Deep Nearest Neighbor Representations Using Differentiable Boundary Trees


Title	Learning Deep Nearest Neighbor Representations Using Differentiable Boundary Trees
Authors	Daniel Zoran, Balaji Lakshminarayanan, Charles Blundell
Abstract	Nearest neighbor (kNN) methods have been gaining popularity in recent years in light of advances in hardware and efficiency of algorithms. There is a plethora of methods to choose from today, each with their own advantages and disadvantages. One requirement shared between all kNN based methods is the need for a good representation and distance measure between samples. We introduce a new method called differentiable boundary tree which allows for learning deep kNN representations. We build on the recently proposed boundary tree algorithm which allows for efficient nearest neighbor classification, regression and retrieval. By modelling traversals in the tree as stochastic events, we are able to form a differentiable cost function which is associated with the tree’s predictions. Using a deep neural network to transform the data and back-propagating through the tree allows us to learn good representations for kNN methods. We demonstrate that our method is able to learn suitable representations allowing for very efficient trees with a clearly interpretable structure.
Tasks
Published	2017-02-28
URL	http://arxiv.org/abs/1702.08833v1
PDF	http://arxiv.org/pdf/1702.08833v1.pdf
PWC	https://paperswithcode.com/paper/learning-deep-nearest-neighbor
Repo	https://github.com/thadikari/differentiable-boundary-trees
Framework	tf

sgmcmc: An R Package for Stochastic Gradient Markov Chain Monte Carlo


Title	sgmcmc: An R Package for Stochastic Gradient Markov Chain Monte Carlo
Authors	Jack Baker, Paul Fearnhead, Emily B. Fox, Christopher Nemeth
Abstract	This paper introduces the R package sgmcmc; which can be used for Bayesian inference on problems with large datasets using stochastic gradient Markov chain Monte Carlo (SGMCMC). Traditional Markov chain Monte Carlo (MCMC) methods, such as Metropolis-Hastings, are known to run prohibitively slowly as the dataset size increases. SGMCMC solves this issue by only using a subset of data at each iteration. SGMCMC requires calculating gradients of the log likelihood and log priors, which can be time consuming and error prone to perform by hand. The sgmcmc package calculates these gradients itself using automatic differentiation, making the implementation of these methods much easier. To do this, the package uses the software library TensorFlow, which has a variety of statistical distributions and mathematical operations as standard, meaning a wide class of models can be built using this framework. SGMCMC has become widely adopted in the machine learning literature, but less so in the statistics community. We believe this may be partly due to lack of software; this package aims to bridge this gap.
Tasks	Bayesian Inference
Published	2017-10-02
URL	http://arxiv.org/abs/1710.00578v3
PDF	http://arxiv.org/pdf/1710.00578v3.pdf
PWC	https://paperswithcode.com/paper/sgmcmc-an-r-package-for-stochastic-gradient
Repo	https://github.com/chris-nemeth/sgmcmc-review-paper
Framework	tf

An Iterative Closest Points Approach to Neural Generative Models


Title	An Iterative Closest Points Approach to Neural Generative Models
Authors	Joose Rajamäki, Perttu Hämäläinen
Abstract	We present a simple way to learn a transformation that maps samples of one distribution to the samples of another distribution. Our algorithm comprises an iteration of 1) drawing samples from some simple distribution and transforming them using a neural network, 2) determining pairwise correspondences between the transformed samples and training data (or a minibatch), and 3) optimizing the weights of the neural network being trained to minimize the distances between the corresponding vectors. This can be considered as a variant of the Iterative Closest Points (ICP) algorithm, common in geometric computer vision, although ICP typically operates on sensor point clouds and linear transforms instead of random sample sets and neural nonlinear transforms. We demonstrate the algorithm on simple synthetic data and MNIST data. We furthermore demonstrate that the algorithm is capable of handling distributions with both continuous and discrete variables.
Tasks
Published	2017-11-16
URL	http://arxiv.org/abs/1711.06562v4
PDF	http://arxiv.org/pdf/1711.06562v4.pdf
PWC	https://paperswithcode.com/paper/an-iterative-closest-points-approach-to
Repo	https://github.com/JooseRajamaeki/ICP
Framework	none

The Atari Grand Challenge Dataset


Title	The Atari Grand Challenge Dataset
Authors	Vitaly Kurin, Sebastian Nowozin, Katja Hofmann, Lucas Beyer, Bastian Leibe
Abstract	Recent progress in Reinforcement Learning (RL), fueled by its combination, with Deep Learning has enabled impressive results in learning to interact with complex virtual environments, yet real-world applications of RL are still scarce. A key limitation is data efficiency, with current state-of-the-art approaches requiring millions of training samples. A promising way to tackle this problem is to augment RL with learning from human demonstrations. However, human demonstration data is not yet readily available. This hinders progress in this direction. The present work addresses this problem as follows. We (i) collect and describe a large dataset of human Atari 2600 replays – the largest and most diverse such data set publicly released to date, (ii) illustrate an example use of this dataset by analyzing the relation between demonstration quality and imitation learning performance, and (iii) outline possible research directions that are opened up by our work.
Tasks	Imitation Learning
Published	2017-05-31
URL	http://arxiv.org/abs/1705.10998v1
PDF	http://arxiv.org/pdf/1705.10998v1.pdf
PWC	https://paperswithcode.com/paper/the-atari-grand-challenge-dataset
Repo	https://github.com/yobibyte/atarigrandchallenge
Framework	none

Deep Semi-Random Features for Nonlinear Function Approximation


Title	Deep Semi-Random Features for Nonlinear Function Approximation
Authors	Kenji Kawaguchi, Bo Xie, Vikas Verma, Le Song
Abstract	We propose semi-random features for nonlinear function approximation. The flexibility of semi-random feature lies between the fully adjustable units in deep learning and the random features used in kernel methods. For one hidden layer models with semi-random features, we prove with no unrealistic assumptions that the model classes contain an arbitrarily good function as the width increases (universality), and despite non-convexity, we can find such a good function (optimization theory) that generalizes to unseen new data (generalization bound). For deep models, with no unrealistic assumptions, we prove universal approximation ability, a lower bound on approximation error, a partial optimization guarantee, and a generalization bound. Depending on the problems, the generalization bound of deep semi-random features can be exponentially better than the known bounds of deep ReLU nets; our generalization error bound can be independent of the depth, the number of trainable weights as well as the input dimensionality. In experiments, we show that semi-random features can match the performance of neural networks by using slightly more units, and it outperforms random features by using significantly fewer units. Moreover, we introduce a new implicit ensemble method by using semi-random features.
Tasks
Published	2017-02-28
URL	http://arxiv.org/abs/1702.08882v7
PDF	http://arxiv.org/pdf/1702.08882v7.pdf
PWC	https://paperswithcode.com/paper/deep-semi-random-features-for-nonlinear
Repo	https://github.com/zixu1986/semi-random
Framework	tf

Strengths and Weaknesses of Deep Learning Models for Face Recognition Against Image Degradations


Title	Strengths and Weaknesses of Deep Learning Models for Face Recognition Against Image Degradations
Authors	Klemen Grm, Vitomir Štruc, Anais Artiges, Matthieu Caron, Hazim Kemal Ekenel
Abstract	Deep convolutional neural networks (CNNs) based approaches are the state-of-the-art in various computer vision tasks, including face recognition. Considerable research effort is currently being directed towards further improving deep CNNs by focusing on more powerful model architectures and better learning techniques. However, studies systematically exploring the strengths and weaknesses of existing deep models for face recognition are still relatively scarce in the literature. In this paper, we try to fill this gap and study the effects of different covariates on the verification performance of four recent deep CNN models using the Labeled Faces in the Wild (LFW) dataset. Specifically, we investigate the influence of covariates related to: image quality – blur, JPEG compression, occlusion, noise, image brightness, contrast, missing pixels; and model characteristics – CNN architecture, color information, descriptor computation; and analyze their impact on the face verification performance of AlexNet, VGG-Face, GoogLeNet, and SqueezeNet. Based on comprehensive and rigorous experimentation, we identify the strengths and weaknesses of the deep learning models, and present key areas for potential future research. Our results indicate that high levels of noise, blur, missing pixels, and brightness have a detrimental effect on the verification performance of all models, whereas the impact of contrast changes and compression artifacts is limited. It has been found that the descriptor computation strategy and color information does not have a significant influence on performance.
Tasks	Face Recognition, Face Verification
Published	2017-10-04
URL	http://arxiv.org/abs/1710.01494v1
PDF	http://arxiv.org/pdf/1710.01494v1.pdf
PWC	https://paperswithcode.com/paper/strengths-and-weaknesses-of-deep-learning
Repo	https://github.com/kgrm/face-recog-eval
Framework	none

Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Models


Title	Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Models
Authors	Wieland Brendel, Jonas Rauber, Matthias Bethge
Abstract	Many machine learning algorithms are vulnerable to almost imperceptible perturbations of their inputs. So far it was unclear how much risk adversarial perturbations carry for the safety of real-world machine learning applications because most methods used to generate such perturbations rely either on detailed model information (gradient-based attacks) or on confidence scores such as class probabilities (score-based attacks), neither of which are available in most real-world scenarios. In many such cases one currently needs to retreat to transfer-based attacks which rely on cumbersome substitute models, need access to the training data and can be defended against. Here we emphasise the importance of attacks which solely rely on the final model decision. Such decision-based attacks are (1) applicable to real-world black-box models such as autonomous cars, (2) need less knowledge and are easier to apply than transfer-based attacks and (3) are more robust to simple defences than gradient- or score-based attacks. Previous attacks in this category were limited to simple models or simple datasets. Here we introduce the Boundary Attack, a decision-based attack that starts from a large adversarial perturbation and then seeks to reduce the perturbation while staying adversarial. The attack is conceptually simple, requires close to no hyperparameter tuning, does not rely on substitute models and is competitive with the best gradient-based attacks in standard computer vision tasks like ImageNet. We apply the attack on two black-box algorithms from Clarifai.com. The Boundary Attack in particular and the class of decision-based attacks in general open new avenues to study the robustness of machine learning models and raise new questions regarding the safety of deployed machine learning systems. An implementation of the attack is available as part of Foolbox at https://github.com/bethgelab/foolbox .
Tasks
Published	2017-12-12
URL	http://arxiv.org/abs/1712.04248v2
PDF	http://arxiv.org/pdf/1712.04248v2.pdf
PWC	https://paperswithcode.com/paper/decision-based-adversarial-attacks-reliable
Repo	https://github.com/Cold-Winter/boundary_attack_foolbox
Framework	tf

Looking at Outfit to Parse Clothing


Title	Looking at Outfit to Parse Clothing
Authors	Pongsate Tangseng, Zhipeng Wu, Kota Yamaguchi
Abstract	This paper extends fully-convolutional neural networks (FCN) for the clothing parsing problem. Clothing parsing requires higher-level knowledge on clothing semantics and contextual cues to disambiguate fine-grained categories. We extend FCN architecture with a side-branch network which we refer outfit encoder to predict a consistent set of clothing labels to encourage combinatorial preference, and with conditional random field (CRF) to explicitly consider coherent label assignment to the given image. The empirical results using Fashionista and CFPD datasets show that our model achieves state-of-the-art performance in clothing parsing, without additional supervision during training. We also study the qualitative influence of annotation on the current clothing parsing benchmarks, with our Web-based tool for multi-scale pixel-wise annotation and manual refinement effort to the Fashionista dataset. Finally, we show that the image representation of the outfit encoder is useful for dress-up image retrieval application.
Tasks	Image Retrieval
Published	2017-03-04
URL	http://arxiv.org/abs/1703.01386v1
PDF	http://arxiv.org/pdf/1703.01386v1.pdf
PWC	https://paperswithcode.com/paper/looking-at-outfit-to-parse-clothing
Repo	https://github.com/ReemHal/Browser-Based-Annotator
Framework	none

Autoencoding Variational Inference For Topic Models


Title	Autoencoding Variational Inference For Topic Models
Authors	Akash Srivastava, Charles Sutton
Abstract	Topic models are one of the most popular methods for learning representations of text, but a major challenge is that any change to the topic model requires mathematically deriving a new inference algorithm. A promising approach to address this problem is autoencoding variational Bayes (AEVB), but it has proven diffi- cult to apply to topic models in practice. We present what is to our knowledge the first effective AEVB based inference method for latent Dirichlet allocation (LDA), which we call Autoencoded Variational Inference For Topic Model (AVITM). This model tackles the problems caused for AEVB by the Dirichlet prior and by component collapsing. We find that AVITM matches traditional methods in accuracy with much better inference time. Indeed, because of the inference network, we find that it is unnecessary to pay the computational cost of running variational optimization on test data. Because AVITM is black box, it is readily applied to new topic models. As a dramatic illustration of this, we present a new topic model called ProdLDA, that replaces the mixture model in LDA with a product of experts. By changing only one line of code from LDA, we find that ProdLDA yields much more interpretable topics, even if LDA is trained via collapsed Gibbs sampling.
Tasks	Topic Models
Published	2017-03-04
URL	http://arxiv.org/abs/1703.01488v1
PDF	http://arxiv.org/pdf/1703.01488v1.pdf
PWC	https://paperswithcode.com/paper/autoencoding-variational-inference-for-topic
Repo	https://github.com/vlukiyanov/pt-avitm
Framework	pytorch

Are we done with object recognition? The iCub robot’s perspective


Title	Are we done with object recognition? The iCub robot’s perspective
Authors	Giulia Pasquale, Carlo Ciliberto, Francesca Odone, Lorenzo Rosasco, Lorenzo Natale
Abstract	We report on an extensive study of the benefits and limitations of current deep learning approaches to object recognition in robot vision scenarios, introducing a novel dataset used for our investigation. To avoid the biases in currently available datasets, we consider a natural human-robot interaction setting to design a data-acquisition protocol for visual object recognition on the iCub humanoid robot. Analyzing the performance of off-the-shelf models trained off-line on large-scale image retrieval datasets, we show the necessity for knowledge transfer. We evaluate different ways in which this last step can be done, and identify the major bottlenecks affecting robotic scenarios. By studying both object categorization and identification problems, we highlight key differences between object recognition in robotics applications and in image retrieval tasks, for which the considered deep learning approaches have been originally designed. In a nutshell, our results confirm the remarkable improvements yield by deep learning in this setting, while pointing to specific open challenges that need be addressed for seamless deployment in robotics.
Tasks	Image Retrieval, Object Recognition, Transfer Learning
Published	2017-09-28
URL	http://arxiv.org/abs/1709.09882v2
PDF	http://arxiv.org/pdf/1709.09882v2.pdf
PWC	https://paperswithcode.com/paper/are-we-done-with-object-recognition-the-icub
Repo	https://github.com/VadymV/ICOS
Framework	tf

Towards a Common Implementation of Reinforcement Learning for Multiple Robotic Tasks


Title	Towards a Common Implementation of Reinforcement Learning for Multiple Robotic Tasks
Authors	Angel Martínez-Tenor, Juan Antonio Fernández-Madrigal, Ana Cruz-Martín, Javier González-Jiménez
Abstract	Mobile robots are increasingly being employed for performing complex tasks in dynamic environments. Reinforcement learning (RL) methods are recognized to be promising for specifying such tasks in a relatively simple manner. However, the strong dependency between the learning method and the task to learn is a well-known problem that restricts practical implementations of RL in robotics, often requiring major modifications of parameters and adding other techniques for each particular task. In this paper we present a practical core implementation of RL which enables the learning process for multiple robotic tasks with minimal per-task tuning or none. Based on value iteration methods, this implementation includes a novel approach for action selection, called Q-biased softmax regression (QBIASSR), which avoids poor performance of the learning process when the robot reaches new unexplored states. Our approach takes advantage of the structure of the state space by attending the physical variables involved (e.g., distances to obstacles, X,Y,{\theta} pose, etc.), thus experienced sets of states may favor the decision-making process of unexplored or rarely-explored states. This improvement has a relevant role in reducing the tuning of the algorithm for particular tasks. Experiments with real and simulated robots, performed with the software framework also introduced here, show that our implementation is effectively able to learn different robotic tasks without tuning the learning method. Results also suggest that the combination of true online SARSA({\lambda}) with QBIASSR can outperform the existing RL core algorithms in low-dimensional robotic tasks.
Tasks	Decision Making
Published	2017-02-21
URL	http://arxiv.org/abs/1702.06329v1
PDF	http://arxiv.org/pdf/1702.06329v1.pdf
PWC	https://paperswithcode.com/paper/towards-a-common-implementation-of
Repo	https://github.com/angelmtenor/RL-ROBOT
Framework	none