Paper Group ANR 518
Human Iris Recognition in Post-mortem Subjects: Study and Database. Entropy-Constrained Training of Deep Neural Networks. Deep Neural Network Compression for Aircraft Collision Avoidance Systems. QUENN: QUantization Engine for low-power Neural Networks. Towards increased trustworthiness of deep learning segmentation methods on cardiac MRI. Filter D …
Human Iris Recognition in Post-mortem Subjects: Study and Database
Title | Human Iris Recognition in Post-mortem Subjects: Study and Database |
Authors | Mateusz Trokielewicz, Adam Czajka, Piotr Maciejewicz |
Abstract | This paper presents a unique study of post-mortem human iris recognition and the first known to us database of near-infrared and visible-light iris images of deceased humans collected up to almost 17 days after death. We used four different iris recognition methods to analyze the dynamics of iris quality decay in short-term comparisons (samples collected up to 60 hours after death) and long-term comparisons (for samples acquired up to 407 hours after demise). This study shows that post-mortem iris recognition is possible and occasionally works even 17 days after death. These conclusions contradict a promulgated rumor that iris is unusable shortly after decease. We make this dataset publicly available to let others verify our findings and to research new aspects of this important and unfamiliar topic. We are not aware of any earlier papers offering post-mortem human iris images and such comprehensive analysis employing four different matchers. |
Tasks | Iris Recognition |
Published | 2018-09-01 |
URL | http://arxiv.org/abs/1809.00174v1 |
http://arxiv.org/pdf/1809.00174v1.pdf | |
PWC | https://paperswithcode.com/paper/human-iris-recognition-in-post-mortem |
Repo | |
Framework | |
Entropy-Constrained Training of Deep Neural Networks
Title | Entropy-Constrained Training of Deep Neural Networks |
Authors | Simon Wiedemann, Arturo Marban, Klaus-Robert Müller, Wojciech Samek |
Abstract | We propose a general framework for neural network compression that is motivated by the Minimum Description Length (MDL) principle. For that we first derive an expression for the entropy of a neural network, which measures its complexity explicitly in terms of its bit-size. Then, we formalize the problem of neural network compression as an entropy-constrained optimization objective. This objective generalizes many of the compression techniques proposed in the literature, in that pruning or reducing the cardinality of the weight elements of the network can be seen special cases of entropy-minimization techniques. Furthermore, we derive a continuous relaxation of the objective, which allows us to minimize it using gradient based optimization techniques. Finally, we show that we can reach state-of-the-art compression results on different network architectures and data sets, e.g. achieving x71 compression gains on a VGG-like architecture. |
Tasks | Neural Network Compression |
Published | 2018-12-18 |
URL | http://arxiv.org/abs/1812.07520v2 |
http://arxiv.org/pdf/1812.07520v2.pdf | |
PWC | https://paperswithcode.com/paper/entropy-constrained-training-of-deep-neural |
Repo | |
Framework | |
Deep Neural Network Compression for Aircraft Collision Avoidance Systems
Title | Deep Neural Network Compression for Aircraft Collision Avoidance Systems |
Authors | Kyle D. Julian, Mykel J. Kochenderfer, Michael P. Owen |
Abstract | One approach to designing decision making logic for an aircraft collision avoidance system frames the problem as a Markov decision process and optimizes the system using dynamic programming. The resulting collision avoidance strategy can be represented as a numeric table. This methodology has been used in the development of the Airborne Collision Avoidance System X (ACAS X) family of collision avoidance systems for manned and unmanned aircraft, but the high dimensionality of the state space leads to very large tables. To improve storage efficiency, a deep neural network is used to approximate the table. With the use of an asymmetric loss function and a gradient descent algorithm, the parameters for this network can be trained to provide accurate estimates of table values while preserving the relative preferences of the possible advisories for each state. By training multiple networks to represent subtables, the network also decreases the required runtime for computing the collision avoidance advisory. Simulation studies show that the network improves the safety and efficiency of the collision avoidance system. Because only the network parameters need to be stored, the required storage space is reduced by a factor of 1000, enabling the collision avoidance system to operate using current avionics systems. |
Tasks | Decision Making, Neural Network Compression |
Published | 2018-10-09 |
URL | http://arxiv.org/abs/1810.04240v1 |
http://arxiv.org/pdf/1810.04240v1.pdf | |
PWC | https://paperswithcode.com/paper/deep-neural-network-compression-for-aircraft |
Repo | |
Framework | |
QUENN: QUantization Engine for low-power Neural Networks
Title | QUENN: QUantization Engine for low-power Neural Networks |
Authors | Miguel de Prado, Maurizio Denna, Luca Benini, Nuria Pazos |
Abstract | Deep Learning is moving to edge devices, ushering in a new age of distributed Artificial Intelligence (AI). The high demand of computational resources required by deep neural networks may be alleviated by approximate computing techniques, and most notably reduced-precision arithmetic with coarsely quantized numerical representations. In this context, Bonseyes comes in as an initiative to enable stakeholders to bring AI to low-power and autonomous environments such as: Automotive, Medical Healthcare and Consumer Electronics. To achieve this, we introduce LPDNN, a framework for optimized deployment of Deep Neural Networks on heterogeneous embedded devices. In this work, we detail the quantization engine that is integrated in LPDNN. The engine depends on a fine-grained workflow which enables a Neural Network Design Exploration and a sensitivity analysis of each layer for quantization. We demonstrate the engine with a case study on Alexnet and VGG16 for three different techniques for direct quantization: standard fixed-point, dynamic fixed-point and k-means clustering, and demonstrate the potential of the latter. We argue that using a Gaussian quantizer with k-means clustering can achieve better performance than linear quantizers. Without retraining, we achieve over 55.64% saving for weights’ storage and 69.17% for run-time memory accesses with less than 1% drop in top5 accuracy in Imagenet. |
Tasks | Quantization |
Published | 2018-11-14 |
URL | http://arxiv.org/abs/1811.05896v1 |
http://arxiv.org/pdf/1811.05896v1.pdf | |
PWC | https://paperswithcode.com/paper/quenn-quantization-engine-for-low-power |
Repo | |
Framework | |
Towards increased trustworthiness of deep learning segmentation methods on cardiac MRI
Title | Towards increased trustworthiness of deep learning segmentation methods on cardiac MRI |
Authors | Jörg Sander, Bob D. de Vos, Jelmer M. Wolterink, Ivana Išgum |
Abstract | Current state-of-the-art deep learning segmentation methods have not yet made a broad entrance into the clinical setting in spite of high demand for such automatic methods. One important reason is the lack of reliability caused by models that fail unnoticed and often locally produce anatomically implausible results that medical experts would not make. This paper presents an automatic image segmentation method based on (Bayesian) dilated convolutional networks (DCNN) that generate segmentation masks and spatial uncertainty maps for the input image at hand. The method was trained and evaluated using segmentation of the left ventricle (LV) cavity, right ventricle (RV) endocardium and myocardium (Myo) at end-diastole (ED) and end-systole (ES) in 100 cardiac 2D MR scans from the MICCAI 2017 Challenge (ACDC). Combining segmentations and uncertainty maps and employing a human-in-the-loop setting, we provide evidence that image areas indicated as highly uncertain regarding the obtained segmentation almost entirely cover regions of incorrect segmentations. The fused information can be harnessed to increase segmentation performance. Our results reveal that we can obtain valuable spatial uncertainty maps with low computational effort using DCNNs. |
Tasks | Semantic Segmentation |
Published | 2018-09-27 |
URL | http://arxiv.org/abs/1809.10430v3 |
http://arxiv.org/pdf/1809.10430v3.pdf | |
PWC | https://paperswithcode.com/paper/towards-increased-trustworthiness-of-deep |
Repo | |
Framework | |
Filter Distillation for Network Compression
Title | Filter Distillation for Network Compression |
Authors | Xavier Suau, Luca Zappella, Nicholas Apostoloff |
Abstract | In this paper we introduce Principal Filter Analysis (PFA), an easy to use and effective method for neural network compression. PFA exploits the correlation between filter responses within network layers to recommend a smaller network that maintain as much as possible the accuracy of the full model. We propose two algorithms: the first allows users to target compression to specific network property, such as number of trainable variable (footprint), and produces a compressed model that satisfies the requested property while preserving the maximum amount of spectral energy in the responses of each layer, while the second is a parameter-free heuristic that selects the compression used at each layer by trying to mimic an ideal set of uncorrelated responses. Since PFA compresses networks based on the correlation of their responses we show in our experiments that it gains the additional flexibility of adapting each architecture to a specific domain while compressing. PFA is evaluated against several architectures and datasets, and shows considerable compression rates without compromising accuracy, e.g., for VGG-16 on CIFAR-10, CIFAR-100 and ImageNet, PFA achieves a compression rate of 8x, 3x, and 1.4x with an accuracy gain of 0.4%, 1.4% points, and 2.4% respectively. Our tests show that PFA is competitive with state-of-the-art approaches while removing adoption barriers thanks to its practical implementation, intuitive philosophy and ease of use. |
Tasks | Domain Adaptation, Neural Network Compression |
Published | 2018-07-20 |
URL | https://arxiv.org/abs/1807.10585v4 |
https://arxiv.org/pdf/1807.10585v4.pdf | |
PWC | https://paperswithcode.com/paper/network-compression-using-correlation |
Repo | |
Framework | |
Unsupervised Textual Grounding: Linking Words to Image Concepts
Title | Unsupervised Textual Grounding: Linking Words to Image Concepts |
Authors | Raymond A. Yeh, Minh N. Do, Alexander G. Schwing |
Abstract | Textual grounding, i.e., linking words to objects in images, is a challenging but important task for robotics and human-computer interaction. Existing techniques benefit from recent progress in deep learning and generally formulate the task as a supervised learning problem, selecting a bounding box from a set of possible options. To train these deep net based approaches, access to a large-scale datasets is required, however, constructing such a dataset is time-consuming and expensive. Therefore, we develop a completely unsupervised mechanism for textual grounding using hypothesis testing as a mechanism to link words to detected image concepts. We demonstrate our approach on the ReferIt Game dataset and the Flickr30k data, outperforming baselines by 7.98% and 6.96% respectively. |
Tasks | |
Published | 2018-03-29 |
URL | http://arxiv.org/abs/1803.11185v1 |
http://arxiv.org/pdf/1803.11185v1.pdf | |
PWC | https://paperswithcode.com/paper/unsupervised-textual-grounding-linking-words |
Repo | |
Framework | |
Exploiting Images for Video Recognition with Hierarchical Generative Adversarial Networks
Title | Exploiting Images for Video Recognition with Hierarchical Generative Adversarial Networks |
Authors | Feiwu Yu, Xinxiao Wu, Yuchao Sun, Lixin Duan |
Abstract | Existing deep learning methods of video recognition usually require a large number of labeled videos for training. But for a new task, videos are often unlabeled and it is also time-consuming and labor-intensive to annotate them. Instead of human annotation, we try to make use of existing fully labeled images to help recognize those videos. However, due to the problem of domain shifts and heterogeneous feature representations, the performance of classifiers trained on images may be dramatically degraded for video recognition tasks. In this paper, we propose a novel method, called Hierarchical Generative Adversarial Networks (HiGAN), to enhance recognition in videos (i.e., target domain) by transferring knowledge from images (i.e., source domain). The HiGAN model consists of a \emph{low-level} conditional GAN and a \emph{high-level} conditional GAN. By taking advantage of these two-level adversarial learning, our method is capable of learning a domain-invariant feature representation of source images and target videos. Comprehensive experiments on two challenging video recognition datasets (i.e. UCF101 and HMDB51) demonstrate the effectiveness of the proposed method when compared with the existing state-of-the-art domain adaptation methods. |
Tasks | Domain Adaptation, Video Recognition |
Published | 2018-05-11 |
URL | http://arxiv.org/abs/1805.04384v1 |
http://arxiv.org/pdf/1805.04384v1.pdf | |
PWC | https://paperswithcode.com/paper/exploiting-images-for-video-recognition-with |
Repo | |
Framework | |
Max-Diversity Distributed Learning: Theory and Algorithms
Title | Max-Diversity Distributed Learning: Theory and Algorithms |
Authors | Yong Liu, Jian Li, Weiping Wang |
Abstract | We study the risk performance of distributed learning for the regularization empirical risk minimization with fast convergence rate, substantially improving the error analysis of the existing divide-and-conquer based distributed learning. An interesting theoretical finding is that the larger the diversity of each local estimate is, the tighter the risk bound is. This theoretical analysis motivates us to devise an effective maxdiversity distributed learning algorithm (MDD). Experimental results show that MDD can outperform the existing divide-andconquer methods but with a bit more time. Theoretical analysis and empirical results demonstrate that our proposed MDD is sound and effective. |
Tasks | |
Published | 2018-12-19 |
URL | http://arxiv.org/abs/1812.07738v2 |
http://arxiv.org/pdf/1812.07738v2.pdf | |
PWC | https://paperswithcode.com/paper/max-diversity-distributed-learning-theory-and |
Repo | |
Framework | |
Learning Style Compatibility for Furniture
Title | Learning Style Compatibility for Furniture |
Authors | Divyansh Aggarwal, Elchin Valiyev, Fadime Sener, Angela Yao |
Abstract | When judging style, a key question that often arises is whether or not a pair of objects are compatible with each other. In this paper we investigate how Siamese networks can be used efficiently for assessing the style compatibility between images of furniture items. We show that the middle layers of pretrained CNNs can capture essential information about furniture style, which allows for efficient applications of such networks for this task. We also use a joint image-text embedding method that allows for the querying of stylistically compatible furniture items, along with additional attribute constraints based on text. To evaluate our methods, we collect and present a large scale dataset of images of furniture of different style categories accompanied by text attributes. |
Tasks | |
Published | 2018-12-09 |
URL | http://arxiv.org/abs/1812.03570v1 |
http://arxiv.org/pdf/1812.03570v1.pdf | |
PWC | https://paperswithcode.com/paper/learning-style-compatibility-for-furniture |
Repo | |
Framework | |
Social Network based Short-Term Stock Trading System
Title | Social Network based Short-Term Stock Trading System |
Authors | Paolo Cremonesi, Chiara Francalanci, Alessandro Poli, Roberto Pagano, Luca Mazzoni, Alberto Maggioni, Mehdi Elahi |
Abstract | This paper proposes a novel adaptive algorithm for the automated short-term trading of financial instrument. The algorithm adopts a semantic sentiment analysis technique to inspect the Twitter posts and to use them to predict the behaviour of the stock market. Indeed, the algorithm is specifically developed to take advantage of both the sentiment and the past values of a certain financial instrument in order to choose the best investment decision. This allows the algorithm to ensure the maximization of the obtainable profits by trading on the stock market. We have conducted an investment simulation and compared the performance of our proposed with a well-known benchmark (DJTATO index) and the optimal results, in which an investor knows in advance the future price of a product. The result shows that our approach outperforms the benchmark and achieves the performance score close to the optimal result. |
Tasks | Sentiment Analysis |
Published | 2018-01-16 |
URL | http://arxiv.org/abs/1801.05295v1 |
http://arxiv.org/pdf/1801.05295v1.pdf | |
PWC | https://paperswithcode.com/paper/social-network-based-short-term-stock-trading |
Repo | |
Framework | |
DOOM Level Generation using Generative Adversarial Networks
Title | DOOM Level Generation using Generative Adversarial Networks |
Authors | Edoardo Giacomello, Pier Luca Lanzi, Daniele Loiacono |
Abstract | We applied Generative Adversarial Networks (GANs) to learn a model of DOOM levels from human-designed content. Initially, we analysed the levels and extracted several topological features. Then, for each level, we extracted a set of images identifying the occupied area, the height map, the walls, and the position of game objects. We trained two GANs: one using plain level images, one using both the images and some of the features extracted during the preliminary analysis. We used the two networks to generate new levels and compared the results to assess whether the network trained using also the topological features could generate levels more similar to human-designed ones. Our results show that GANs can capture intrinsic structure of DOOM levels and appears to be a promising approach to level generation in first person shooter games. |
Tasks | |
Published | 2018-04-24 |
URL | http://arxiv.org/abs/1804.09154v1 |
http://arxiv.org/pdf/1804.09154v1.pdf | |
PWC | https://paperswithcode.com/paper/doom-level-generation-using-generative |
Repo | |
Framework | |
CUNI System for the WMT18 Multimodal Translation Task
Title | CUNI System for the WMT18 Multimodal Translation Task |
Authors | Jindřich Helcl, Jindřich Libovický, Dušan Variš |
Abstract | We present our submission to the WMT18 Multimodal Translation Task. The main feature of our submission is applying a self-attentive network instead of a recurrent neural network. We evaluate two methods of incorporating the visual features in the model: first, we include the image representation as another input to the network; second, we train the model to predict the visual features and use it as an auxiliary objective. For our submission, we acquired both textual and multimodal additional data. Both of the proposed methods yield significant improvements over recurrent networks and self-attentive textual baselines. |
Tasks | |
Published | 2018-11-12 |
URL | http://arxiv.org/abs/1811.04697v1 |
http://arxiv.org/pdf/1811.04697v1.pdf | |
PWC | https://paperswithcode.com/paper/cuni-system-for-the-wmt18-multimodal |
Repo | |
Framework | |
Generating Rescheduling Knowledge using Reinforcement Learning in a Cognitive Architecture
Title | Generating Rescheduling Knowledge using Reinforcement Learning in a Cognitive Architecture |
Authors | Jorge A. Palombarini, Juan Cruz Barsce, Ernesto C. Martínez |
Abstract | In order to reach higher degrees of flexibility, adaptability and autonomy in manufacturing systems, it is essential to develop new rescheduling methodologies which resort to cognitive capabilities, similar to those found in human beings. Artificial cognition is important for designing planning and control systems that generate and represent knowledge about heuristics for repair-based scheduling. Rescheduling knowledge in the form of decision rules is used to deal with unforeseen events and disturbances reactively in real time, and take advantage of the ability to act interactively with the user to counteract the effects of disruptions. In this work, to achieve the aforementioned goals, a novel approach to generate rescheduling knowledge in the form of dynamic first-order logical rules is proposed. The proposed approach is based on the integration of reinforcement learning with artificial cognitive capabilities involving perception and reasoning/learning skills embedded in the Soar cognitive architecture. An industrial example is discussed showing that the approach enables the scheduling system to assess its operational range in an autonomic way, and to acquire experience through intensive simulation while performing repair tasks. |
Tasks | |
Published | 2018-05-12 |
URL | http://arxiv.org/abs/1805.04752v1 |
http://arxiv.org/pdf/1805.04752v1.pdf | |
PWC | https://paperswithcode.com/paper/generating-rescheduling-knowledge-using |
Repo | |
Framework | |
Multiobjective Optimization Differential Evolution Enhanced with Principle Component Analysis for Constrained Optimization
Title | Multiobjective Optimization Differential Evolution Enhanced with Principle Component Analysis for Constrained Optimization |
Authors | Wei Huang, Tao Xu, Kangshun Li, Jun He |
Abstract | Multiobjective evolutionary algorithms (MOEAs) have been successfully applied to a number of constrained optimization problems. Many of them adopt mutation and crossover operators from differential evolution. However, these operators do not explicitly utilise features of fitness landscapes. To improve the performance of algorithms, this paper aims at designing a search operator adapting to fitness landscapes. Through an observation, we find that principle component analysis (PCA) can be used to characterise fitness landscapes. Based on this finding, a new search operator, called PCA-projection, is proposed. In order to verify the effectiveness of PCA-projection, we design two algorithms enhanced with PCA-projection for solving constrained optimization problems, called PMODE and HECO-PDE, respectively. Experiments have been conducted on the IEEE CEC 2017 competition benchmark suite in constrained optimisation. PMODE and HECO-PDE are compared with the algorithms from the IEEE CEC 2018 competition and another recent MOEA for constrained optimisation. Experimental results show that an algorithm enhanced with PCA-projection performs better than its corresponding opponent without this operator. Furthermore, HECO-PDE is ranked first on all dimensions according to the competition rules. This study reveals that decomposition-based MOEAs, such as HECO-PDE, are competitive with best single-objective and multiobjective evolutionary algorithms for constrained optimisation, but MOEAs based on non-dominance, such as PMODE, may not. |
Tasks | Multiobjective Optimization |
Published | 2018-05-01 |
URL | https://arxiv.org/abs/1805.00272v2 |
https://arxiv.org/pdf/1805.00272v2.pdf | |
PWC | https://paperswithcode.com/paper/multiobjective-optimization-differential |
Repo | |
Framework | |