July 26, 2019

3021 words 15 mins read

Paper Group ANR 783

Deep Reinforcement Learning Attention Selection for Person Re-Identification. A unified view of entropy-regularized Markov decision processes. Context Based Visual Content Verification. Morph-fitting: Fine-Tuning Word Vector Spaces with Simple Language-Specific Rules. Residual Conv-Deconv Grid Network for Semantic Segmentation. A Localisation-Segme …

Deep Reinforcement Learning Attention Selection for Person Re-Identification


Title	Deep Reinforcement Learning Attention Selection for Person Re-Identification
Authors	Xu Lan, Hanxiao Wang, Shaogang Gong, Xiatian Zhu
Abstract	Existing person re-identification (re-id) methods assume the provision of accurately cropped person bounding boxes with minimum background noise, mostly by manually cropping. This is significantly breached in practice when person bounding boxes must be detected automatically given a very large number of images and/or videos processed. Compared to carefully cropped manually, auto-detected bounding boxes are far less accurate with random amount of background clutter which can degrade notably person re-id matching accuracy. In this work, we develop a joint learning deep model that optimises person re-id attention selection within any auto-detected person bounding boxes by reinforcement learning of background clutter minimisation subject to re-id label pairwise constraints. Specifically, we formulate a novel unified re-id architecture called Identity DiscriminativE Attention reinforcement Learning (IDEAL) to accurately select re-id attention in auto-detected bounding boxes for optimising re-id performance. Our model can improve re-id accuracy comparable to that from exhaustive human manual cropping of bounding boxes with additional advantages from identity discriminative attention selection that specially benefits re-id tasks beyond human knowledge. Extensive comparative evaluations demonstrate the re-id advantages of the proposed IDEAL model over a wide range of state-of-the-art re-id methods on two auto-detected re-id benchmarks CUHK03 and Market-1501.
Tasks	Person Re-Identification
Published	2017-07-10
URL	http://arxiv.org/abs/1707.02785v4
PDF	http://arxiv.org/pdf/1707.02785v4.pdf
PWC	https://paperswithcode.com/paper/deep-reinforcement-learning-attention
Repo
Framework

A unified view of entropy-regularized Markov decision processes


Title	A unified view of entropy-regularized Markov decision processes
Authors	Gergely Neu, Anders Jonsson, Vicenç Gómez
Abstract	We propose a general framework for entropy-regularized average-reward reinforcement learning in Markov decision processes (MDPs). Our approach is based on extending the linear-programming formulation of policy optimization in MDPs to accommodate convex regularization functions. Our key result is showing that using the conditional entropy of the joint state-action distributions as regularization yields a dual optimization problem closely resembling the Bellman optimality equations. This result enables us to formalize a number of state-of-the-art entropy-regularized reinforcement learning algorithms as approximate variants of Mirror Descent or Dual Averaging, and thus to argue about the convergence properties of these methods. In particular, we show that the exact version of the TRPO algorithm of Schulman et al. (2015) actually converges to the optimal policy, while the entropy-regularized policy gradient methods of Mnih et al. (2016) may fail to converge to a fixed point. Finally, we illustrate empirically the effects of using various regularization techniques on learning performance in a simple reinforcement learning setup.
Tasks	Policy Gradient Methods
Published	2017-05-22
URL	http://arxiv.org/abs/1705.07798v1
PDF	http://arxiv.org/pdf/1705.07798v1.pdf
PWC	https://paperswithcode.com/paper/a-unified-view-of-entropy-regularized-markov
Repo
Framework

Context Based Visual Content Verification


Title	Context Based Visual Content Verification
Authors	Martin Lukac, Aigerim Bazarbayeva, Michitaka Kameyama
Abstract	In this paper the intermediary visual content verification method based on multi-level co-occurrences is studied. The co-occurrence statistics are in general used to determine relational properties between objects based on information collected from data. As such these measures are heavily subject to relative number of occurrences and give only limited amount of accuracy when predicting objects in real world. In order to improve the accuracy of this method in the verification task, we include the context information such as location, type of environment etc. In order to train our model we provide new annotated dataset the Advanced Attribute VOC (AAVOC) that contains additional properties of the image. We show that the usage of context greatly improve the accuracy of verification with up to 16% improvement.
Tasks
Published	2017-09-01
URL	http://arxiv.org/abs/1709.00141v1
PDF	http://arxiv.org/pdf/1709.00141v1.pdf
PWC	https://paperswithcode.com/paper/context-based-visual-content-verification
Repo
Framework

Morph-fitting: Fine-Tuning Word Vector Spaces with Simple Language-Specific Rules


Title	Morph-fitting: Fine-Tuning Word Vector Spaces with Simple Language-Specific Rules
Authors	Ivan Vulić, Nikola Mrkšić, Roi Reichart, Diarmuid Ó Séaghdha, Steve Young, Anna Korhonen
Abstract	Morphologically rich languages accentuate two properties of distributional vector space models: 1) the difficulty of inducing accurate representations for low-frequency word forms; and 2) insensitivity to distinct lexical relations that have similar distributional signatures. These effects are detrimental for language understanding systems, which may infer that ‘inexpensive’ is a rephrasing for ‘expensive’ or may not associate ‘acquire’ with ‘acquires’. In this work, we propose a novel morph-fitting procedure which moves past the use of curated semantic lexicons for improving distributional vector spaces. Instead, our method injects morphological constraints generated using simple language-specific rules, pulling inflectional forms of the same word close together and pushing derivational antonyms far apart. In intrinsic evaluation over four languages, we show that our approach: 1) improves low-frequency word estimates; and 2) boosts the semantic quality of the entire word vector collection. Finally, we show that morph-fitted vectors yield large gains in the downstream task of dialogue state tracking, highlighting the importance of morphology for tackling long-tail phenomena in language understanding tasks.
Tasks	Dialogue State Tracking
Published	2017-06-01
URL	http://arxiv.org/abs/1706.00377v1
PDF	http://arxiv.org/pdf/1706.00377v1.pdf
PWC	https://paperswithcode.com/paper/morph-fitting-fine-tuning-word-vector-spaces
Repo
Framework

Residual Conv-Deconv Grid Network for Semantic Segmentation


Title	Residual Conv-Deconv Grid Network for Semantic Segmentation
Authors	Damien Fourure, Rémi Emonet, Elisa Fromont, Damien Muselet, Alain Tremeau, Christian Wolf
Abstract	This paper presents GridNet, a new Convolutional Neural Network (CNN) architecture for semantic image segmentation (full scene labelling). Classical neural networks are implemented as one stream from the input to the output with subsampling operators applied in the stream in order to reduce the feature maps size and to increase the receptive field for the final prediction. However, for semantic image segmentation, where the task consists in providing a semantic class to each pixel of an image, feature maps reduction is harmful because it leads to a resolution loss in the output prediction. To tackle this problem, our GridNet follows a grid pattern allowing multiple interconnected streams to work at different resolutions. We show that our network generalizes many well known networks such as conv-deconv, residual or U-Net networks. GridNet is trained from scratch and achieves competitive results on the Cityscapes dataset.
Tasks	Semantic Segmentation
Published	2017-07-25
URL	http://arxiv.org/abs/1707.07958v2
PDF	http://arxiv.org/pdf/1707.07958v2.pdf
PWC	https://paperswithcode.com/paper/residual-conv-deconv-grid-network-for
Repo
Framework

A Localisation-Segmentation Approach for Multi-label Annotation of Lumbar Vertebrae using Deep Nets


Title	A Localisation-Segmentation Approach for Multi-label Annotation of Lumbar Vertebrae using Deep Nets
Authors	Anjany Sekuboyina, Alexander Valentinitsch, Jan S. Kirschke, Bjoern H. Menze
Abstract	Multi-class segmentation of vertebrae is a non-trivial task mainly due to the high correlation in the appearance of adjacent vertebrae. Hence, such a task calls for the consideration of both global and local context. Based on this motivation, we propose a two-staged approach that, given a computed tomography dataset of the spine, segments the five lumbar vertebrae and simultaneously labels them. The first stage employs a multi-layered perceptron performing non-linear regression for locating the lumbar region using the global context. The second stage, comprised of a fully-convolutional deep network, exploits the local context in the localised lumbar region to segment and label the lumbar vertebrae in one go. Aided with practical data augmentation for training, our approach is highly generalisable, capable of successfully segmenting both healthy and abnormal vertebrae (fractured and scoliotic spines). We consistently achieve an average Dice coefficient of over 90 percent on a publicly available dataset of the xVertSeg segmentation challenge of MICCAI 2016. This is particularly noteworthy because the xVertSeg dataset is beset with severe deformities in the form of vertebral fractures and scoliosis.
Tasks	Data Augmentation
Published	2017-03-13
URL	http://arxiv.org/abs/1703.04347v1
PDF	http://arxiv.org/pdf/1703.04347v1.pdf
PWC	https://paperswithcode.com/paper/a-localisation-segmentation-approach-for
Repo
Framework

Deep learning bank distress from news and numerical financial data


Title	Deep learning bank distress from news and numerical financial data
Authors	Paola Cerchiello, Giancarlo Nicola, Samuel Ronnqvist, Peter Sarlin
Abstract	In this paper we focus our attention on the exploitation of the information contained in financial news to enhance the performance of a classifier of bank distress. Such information should be analyzed and inserted into the predictive model in the most efficient way and this task deals with all the issues related to text analysis and specifically analysis of news media. Among the different models proposed for such purpose, we investigate one of the possible deep learning approaches, based on a doc2vec representation of the textual data, a kind of neural network able to map the sequential and symbolic text input onto a reduced latent semantic space. Afterwards, a second supervised neural network is trained combining news data with standard financial figures to classify banks whether in distressed or tranquil states, based on a small set of known distress events. Then the final aim is not only the improvement of the predictive performance of the classifier but also to assess the importance of news data in the classification process. Does news data really bring more useful information not contained in standard financial variables? Our results seem to confirm such hypothesis.
Tasks
Published	2017-06-29
URL	http://arxiv.org/abs/1706.09627v3
PDF	http://arxiv.org/pdf/1706.09627v3.pdf
PWC	https://paperswithcode.com/paper/deep-learning-bank-distress-from-news-and
Repo
Framework

Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning


Title	Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning
Authors	Shixiang Gu, Timothy Lillicrap, Zoubin Ghahramani, Richard E. Turner, Bernhard Schölkopf, Sergey Levine
Abstract	Off-policy model-free deep reinforcement learning methods using previously collected data can improve sample efficiency over on-policy policy gradient techniques. On the other hand, on-policy algorithms are often more stable and easier to use. This paper examines, both theoretically and empirically, approaches to merging on- and off-policy updates for deep reinforcement learning. Theoretical results show that off-policy updates with a value function estimator can be interpolated with on-policy policy gradient updates whilst still satisfying performance bounds. Our analysis uses control variate methods to produce a family of policy gradient algorithms, with several recently proposed algorithms being special cases of this family. We then provide an empirical comparison of these techniques with the remaining algorithmic details fixed, and show how different mixing of off-policy gradient estimates with on-policy samples contribute to improvements in empirical performance. The final algorithm provides a generalization and unification of existing deep policy gradient techniques, has theoretical guarantees on the bias introduced by off-policy updates, and improves on the state-of-the-art model-free deep RL methods on a number of OpenAI Gym continuous control benchmarks.
Tasks	Continuous Control
Published	2017-06-01
URL	http://arxiv.org/abs/1706.00387v1
PDF	http://arxiv.org/pdf/1706.00387v1.pdf
PWC	https://paperswithcode.com/paper/interpolated-policy-gradient-merging-on
Repo
Framework

Weakly supervised training of deep convolutional neural networks for overhead pedestrian localization in depth fields


Title	Weakly supervised training of deep convolutional neural networks for overhead pedestrian localization in depth fields
Authors	Alessandro Corbetta, Vlado Menkovski, Federico Toschi
Abstract	Overhead depth map measurements capture sufficient amount of information to enable human experts to track pedestrians accurately. However, fully automating this process using image analysis algorithms can be challenging. Even though hand-crafted image analysis algorithms are successful in many common cases, they fail frequently when there are complex interactions of multiple objects in the image. Many of the assumptions underpinning the hand-crafted solutions do not hold in these cases and the multitude of exceptions are hard to model precisely. Deep Learning (DL) algorithms, on the other hand, do not require hand crafted solutions and are the current state-of-the-art in object localization in images. However, they require exceeding amount of annotations to produce successful models. In the case of object localization these annotations are difficult and time consuming to produce. In this work we present an approach for developing pedestrian localization models using DL algorithms with efficient weak supervision from an expert. We circumvent the need for annotation of large corpus of data by annotating only small amount of patches and relying on synthetic data augmentation as a vehicle for injecting expert knowledge in the model training. This approach of weak supervision through expert selection of representative patches, suitable transformations and synthetic data augmentations enables us to successfully develop DL models for pedestrian localization efficiently.
Tasks	Data Augmentation, Object Localization
Published	2017-06-09
URL	http://arxiv.org/abs/1706.02850v1
PDF	http://arxiv.org/pdf/1706.02850v1.pdf
PWC	https://paperswithcode.com/paper/weakly-supervised-training-of-deep
Repo
Framework

Solving a Path Planning Problem in a Partially Known Environment using a Swarm Algorithm


Title	Solving a Path Planning Problem in a Partially Known Environment using a Swarm Algorithm
Authors	Esh Vckay, Mansimar Aneja, Dipti Deodhare
Abstract	This paper proposes a path planning strategy for an Autonomous Ground Vehicle (AGV) navigating in a partially known environment. Global path planning is performed by first using a spatial database of the region to be traversed containing selected attributes such as height data and soil information from a suitable spatial database. The database is processed using a biomimetic swarm algorithm that is inspired by the nest building strategies followed by termites. Local path planning is performed online utilizing information regarding contingencies that affect the safe navigation of the AGV from various sensors. The simulation discussed has been implemented on the open source Player-Stage-Gazebo platform.
Tasks
Published	2017-05-09
URL	https://arxiv.org/abs/1705.03176v2
PDF	https://arxiv.org/pdf/1705.03176v2.pdf
PWC	https://paperswithcode.com/paper/solving-a-path-planning-problem-in-a
Repo
Framework

Use Privacy in Data-Driven Systems: Theory and Experiments with Machine Learnt Programs


Title	Use Privacy in Data-Driven Systems: Theory and Experiments with Machine Learnt Programs
Authors	Anupam Datta, Matthew Fredrikson, Gihyuk Ko, Piotr Mardziel, Shayak Sen
Abstract	This paper presents an approach to formalizing and enforcing a class of use privacy properties in data-driven systems. In contrast to prior work, we focus on use restrictions on proxies (i.e. strong predictors) of protected information types. Our definition relates proxy use to intermediate computations that occur in a program, and identify two essential properties that characterize this behavior: 1) its result is strongly associated with the protected information type in question, and 2) it is likely to causally affect the final output of the program. For a specific instantiation of this definition, we present a program analysis technique that detects instances of proxy use in a model, and provides a witness that identifies which parts of the corresponding program exhibit the behavior. Recognizing that not all instances of proxy use of a protected information type are inappropriate, we make use of a normative judgment oracle that makes this inappropriateness determination for a given witness. Our repair algorithm uses the witness of an inappropriate proxy use to transform the model into one that provably does not exhibit proxy use, while avoiding changes that unduly affect classification accuracy. Using a corpus of social datasets, our evaluation shows that these algorithms are able to detect proxy use instances that would be difficult to find using existing techniques, and subsequently remove them while maintaining acceptable classification performance.
Tasks
Published	2017-05-22
URL	http://arxiv.org/abs/1705.07807v3
PDF	http://arxiv.org/pdf/1705.07807v3.pdf
PWC	https://paperswithcode.com/paper/use-privacy-in-data-driven-systems-theory-and
Repo
Framework

On Unifying Deep Generative Models


Title	On Unifying Deep Generative Models
Authors	Zhiting Hu, Zichao Yang, Ruslan Salakhutdinov, Eric P. Xing
Abstract	Deep generative models have achieved impressive success in recent years. Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), as emerging families for generative model learning, have largely been considered as two distinct paradigms and received extensive independent studies respectively. This paper aims to establish formal connections between GANs and VAEs through a new formulation of them. We interpret sample generation in GANs as performing posterior inference, and show that GANs and VAEs involve minimizing KL divergences of respective posterior and inference distributions with opposite directions, extending the two learning phases of classic wake-sleep algorithm, respectively. The unified view provides a powerful tool to analyze a diverse set of existing model variants, and enables to transfer techniques across research lines in a principled way. For example, we apply the importance weighting method in VAE literatures for improved GAN learning, and enhance VAEs with an adversarial mechanism that leverages generated samples. Experiments show generality and effectiveness of the transferred techniques.
Tasks
Published	2017-06-02
URL	http://arxiv.org/abs/1706.00550v5
PDF	http://arxiv.org/pdf/1706.00550v5.pdf
PWC	https://paperswithcode.com/paper/on-unifying-deep-generative-models
Repo
Framework

Neural Stain-Style Transfer Learning using GAN for Histopathological Images


Title	Neural Stain-Style Transfer Learning using GAN for Histopathological Images
Authors	Hyungjoo Cho, Sungbin Lim, Gunho Choi, Hyunseok Min
Abstract	Performance of data-driven network for tumor classification varies with stain-style of histopathological images. This article proposes the stain-style transfer (SST) model based on conditional generative adversarial networks (GANs) which is to learn not only the certain color distribution but also the corresponding histopathological pattern. Our model considers feature-preserving loss in addition to well-known GAN loss. Consequently our model does not only transfers initial stain-styles to the desired one but also prevent the degradation of tumor classifier on transferred images. The model is examined using the CAMELYON16 dataset.
Tasks	Style Transfer, Transfer Learning
Published	2017-10-23
URL	http://arxiv.org/abs/1710.08543v2
PDF	http://arxiv.org/pdf/1710.08543v2.pdf
PWC	https://paperswithcode.com/paper/neural-stain-style-transfer-learning-using
Repo
Framework

A Generative Parser with a Discriminative Recognition Algorithm


Title	A Generative Parser with a Discriminative Recognition Algorithm
Authors	Jianpeng Cheng, Adam Lopez, Mirella Lapata
Abstract	Generative models defining joint distributions over parse trees and sentences are useful for parsing and language modeling, but impose restrictions on the scope of features and are often outperformed by discriminative models. We propose a framework for parsing and language modeling which marries a generative model with a discriminative recognition model in an encoder-decoder setting. We provide interpretations of the framework based on expectation maximization and variational inference, and show that it enables parsing and language modeling within a single implementation. On the English Penn Treen-bank, our framework obtains competitive performance on constituency parsing while matching the state-of-the-art single-model language modeling score.
Tasks	Constituency Parsing, Language Modelling
Published	2017-08-01
URL	http://arxiv.org/abs/1708.00415v2
PDF	http://arxiv.org/pdf/1708.00415v2.pdf
PWC	https://paperswithcode.com/paper/a-generative-parser-with-a-discriminative
Repo
Framework

Controllable Abstractive Summarization


Title	Controllable Abstractive Summarization
Authors	Angela Fan, David Grangier, Michael Auli
Abstract	Current models for document summarization disregard user preferences such as the desired length, style, the entities that the user might be interested in, or how much of the document the user has already read. We present a neural summarization model with a simple but effective mechanism to enable users to specify these high level attributes in order to control the shape of the final summaries to better suit their needs. With user input, our system can produce high quality summaries that follow user preferences. Without user input, we set the control variables automatically. On the full text CNN-Dailymail dataset, we outperform state of the art abstractive systems (both in terms of F1-ROUGE1 40.38 vs. 39.53 and human evaluation).
Tasks	Abstractive Text Summarization, Document Summarization
Published	2017-11-14
URL	http://arxiv.org/abs/1711.05217v2
PDF	http://arxiv.org/pdf/1711.05217v2.pdf
PWC	https://paperswithcode.com/paper/controllable-abstractive-summarization
Repo
Framework