July 28, 2019

3177 words 15 mins read

Paper Group ANR 226

LCDet: Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems. Aggregating incoherent agents who disagree. Can CNN Construct Highly Accurate Models Efficiently for High-Dimensional Problems in Complex Product Designs?. Machine Learning and Social Robotics for Detecting Early Signs of Dementia. Une mesure d’exper …

LCDet: Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems


Title	LCDet: Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems
Authors	Subarna Tripathi, Gokce Dane, Byeongkeun Kang, Vasudev Bhaskaran, Truong Nguyen
Abstract	Deep convolutional Neural Networks (CNN) are the state-of-the-art performers for object detection task. It is well known that object detection requires more computation and memory than image classification. Thus the consolidation of a CNN-based object detection for an embedded system is more challenging. In this work, we propose LCDet, a fully-convolutional neural network for generic object detection that aims to work in embedded systems. We design and develop an end-to-end TensorFlow(TF)-based model. Additionally, we employ 8-bit quantization on the learned weights. We use face detection as a use case. Our TF-Slim based network can predict different faces of different shapes and sizes in a single forward pass. Our experimental results show that the proposed method achieves comparative accuracy comparing with state-of-the-art CNN-based face detection methods, while reducing the model size by 3x and memory-BW by ~4x comparing with one of the best real-time CNN-based object detector such as YOLO. TF 8-bit quantized model provides additional 4x memory reduction while keeping the accuracy as good as the floating point model. The proposed model thus becomes amenable for embedded implementations.
Tasks	Face Detection, Image Classification, Object Detection, Quantization
Published	2017-05-16
URL	http://arxiv.org/abs/1705.05922v1
PDF	http://arxiv.org/pdf/1705.05922v1.pdf
PWC	https://paperswithcode.com/paper/lcdet-low-complexity-fully-convolutional
Repo
Framework

Aggregating incoherent agents who disagree


Title	Aggregating incoherent agents who disagree
Authors	Richard Pettigrew
Abstract	In this paper, we explore how we should aggregate the degrees of belief of of a group of agents to give a single coherent set of degrees of belief, when at least some of those agents might be probabilistically incoherent. There are a number of way of aggregating degrees of belief, and there are a number of ways of fixing incoherent degrees of belief. When we have picked one of each, should we aggregate first and then fix, or fix first and then aggregate? Or should we try to do both at once? And when do these different procedures agree with one another? In this paper, we focus particularly on the final question.
Tasks
Published	2017-09-12
URL	http://arxiv.org/abs/1709.03981v1
PDF	http://arxiv.org/pdf/1709.03981v1.pdf
PWC	https://paperswithcode.com/paper/aggregating-incoherent-agents-who-disagree
Repo
Framework

Can CNN Construct Highly Accurate Models Efficiently for High-Dimensional Problems in Complex Product Designs?


Title	Can CNN Construct Highly Accurate Models Efficiently for High-Dimensional Problems in Complex Product Designs?
Authors	Yu Li, Hu Wang, Juanjuan Liu
Abstract	With the increase of the nonlinearity and dimension, it is difficult for the present popular metamodeling techniques to construct reliable metamodels. To address this problem, Convolutional Neural Network (CNN) is introduced to construct a highly accurate metamodel efficiently. Considering the inherent characteristics of the CNN, it is a potential modeling tool to handle highly nonlinear and dimensional problems (hundreds-dimensional problems) with the limited training samples. In order to evaluate the proposed CNN metamodel for hundreds-dimensional and strong nonlinear problems, CNN is compared with other metamodeling techniques. Furthermore, several high-dimensional analytical functions are also employed to test the CNN metamodel. Testing and comparisons confirm the efficiency and capability of the CNN metamodel for hundreds-dimensional and strong nonlinear problems. Moreover, the proposed CNN metamodel is also applied to IsoGeometric Analysis (IGA)-based optimization successfully.
Tasks
Published	2017-11-15
URL	http://arxiv.org/abs/1712.01639v3
PDF	http://arxiv.org/pdf/1712.01639v3.pdf
PWC	https://paperswithcode.com/paper/can-cnn-construct-highly-accurate-models
Repo
Framework


Title	Machine Learning and Social Robotics for Detecting Early Signs of Dementia
Authors	Patrik Jonell, Joseph Mendelson, Thomas Storskog, Goran Hagman, Per Ostberg, Iolanda Leite, Taras Kucherenko, Olga Mikheeva, Ulrika Akenine, Vesna Jelic, Alina Solomon, Jonas Beskow, Joakim Gustafson, Miia Kivipelto, Hedvig Kjellstrom
Abstract	This paper presents the EACare project, an ambitious multi-disciplinary collaboration with the aim to develop an embodied system, capable of carrying out neuropsychological tests to detect early signs of dementia, e.g., due to Alzheimer’s disease. The system will use methods from Machine Learning and Social Robotics, and be trained with examples of recorded clinician-patient interactions. The interaction will be developed using a participatory design approach. We describe the scope and method of the project, and report on a first Wizard of Oz prototype.
Tasks
Published	2017-09-05
URL	http://arxiv.org/abs/1709.01613v1
PDF	http://arxiv.org/pdf/1709.01613v1.pdf
PWC	https://paperswithcode.com/paper/machine-learning-and-social-robotics-for
Repo
Framework

Une mesure d’expertise pour le crowdsourcing


Title	Une mesure d’expertise pour le crowdsourcing
Authors	Hosna Ouni, Arnaud Martin, Laetitia Gros, Mouloud Kharoune, Zoltan Miklos
Abstract	Crowdsourcing, a major economic issue, is the fact that the firm outsources internal task to the crowd. It is a form of digital subcontracting for the general public. The evaluation of the participants work quality is a major issue in crowdsourcing. Indeed, contributions must be controlled to ensure the effectiveness and relevance of the campaign. We are particularly interested in small, fast and not automatable tasks. Several methods have been proposed to solve this problem, but they are applicable when the “golden truth” is not always known. This work has the particularity to propose a method for calculating the degree of expertise in the presence of gold data in crowdsourcing. This method is based on the belief function theory and proposes a structuring of data using graphs. The proposed approach will be assessed and applied to the data.
Tasks
Published	2017-01-17
URL	http://arxiv.org/abs/1701.04645v1
PDF	http://arxiv.org/pdf/1701.04645v1.pdf
PWC	https://paperswithcode.com/paper/une-mesure-dexpertise-pour-le-crowdsourcing
Repo
Framework

PathNet: Evolution Channels Gradient Descent in Super Neural Networks


Title	PathNet: Evolution Channels Gradient Descent in Super Neural Networks
Authors	Chrisantha Fernando, Dylan Banarse, Charles Blundell, Yori Zwols, David Ha, Andrei A. Rusu, Alexander Pritzel, Daan Wierstra
Abstract	For artificial general intelligence (AGI) it would be efficient if multiple users trained the same giant neural network, permitting parameter reuse, without catastrophic forgetting. PathNet is a first step in this direction. It is a neural network algorithm that uses agents embedded in the neural network whose task is to discover which parts of the network to re-use for new tasks. Agents are pathways (views) through the network which determine the subset of parameters that are used and updated by the forwards and backwards passes of the backpropogation algorithm. During learning, a tournament selection genetic algorithm is used to select pathways through the neural network for replication and mutation. Pathway fitness is the performance of that pathway measured according to a cost function. We demonstrate successful transfer learning; fixing the parameters along a path learned on task A and re-evolving a new population of paths for task B, allows task B to be learned faster than it could be learned from scratch or after fine-tuning. Paths evolved on task B re-use parts of the optimal path evolved on task A. Positive transfer was demonstrated for binary MNIST, CIFAR, and SVHN supervised learning classification tasks, and a set of Atari and Labyrinth reinforcement learning tasks, suggesting PathNets have general applicability for neural network training. Finally, PathNet also significantly improves the robustness to hyperparameter choices of a parallel asynchronous reinforcement learning algorithm (A3C).
Tasks	Transfer Learning
Published	2017-01-30
URL	http://arxiv.org/abs/1701.08734v1
PDF	http://arxiv.org/pdf/1701.08734v1.pdf
PWC	https://paperswithcode.com/paper/pathnet-evolution-channels-gradient-descent
Repo
Framework

Deep Feature Consistent Deep Image Transformations: Downscaling, Decolorization and HDR Tone Mapping


Title	Deep Feature Consistent Deep Image Transformations: Downscaling, Decolorization and HDR Tone Mapping
Authors	Xianxu Hou, Jiang Duan, Guoping Qiu
Abstract	Building on crucial insights into the determining factors of the visual integrity of an image and the property of deep convolutional neural network (CNN), we have developed the Deep Feature Consistent Deep Image Transformation (DFC-DIT) framework which unifies challenging one-to-many mapping image processing problems such as image downscaling, decolorization (colour to grayscale conversion) and high dynamic range (HDR) image tone mapping. We train one CNN as a non-linear mapper to transform an input image to an output image following what we term the deep feature consistency principle which is enforced through another pretrained and fixed deep CNN. This is the first work that uses deep learning to solve and unify these three common image processing tasks. We present experimental results to demonstrate the effectiveness of the DFC-DIT technique and its state of the art performances.
Tasks
Published	2017-07-29
URL	http://arxiv.org/abs/1707.09482v2
PDF	http://arxiv.org/pdf/1707.09482v2.pdf
PWC	https://paperswithcode.com/paper/deep-feature-consistent-deep-image
Repo
Framework

Correcting Nuisance Variation using Wasserstein Distance


Title	Correcting Nuisance Variation using Wasserstein Distance
Authors	Gil Tabak, Minjie Fan, Samuel J. Yang, Stephan Hoyer, Geoff Davis
Abstract	Profiling cellular phenotypes from microscopic imaging can provide meaningful biological information resulting from various factors affecting the cells. One motivating application is drug development: morphological cell features can be captured from images, from which similarities between different drug compounds applied at different doses can be quantified. The general approach is to find a function mapping the images to an embedding space of manageable dimensionality whose geometry captures relevant features of the input images. An important known issue for such methods is separating relevant biological signal from nuisance variation. For example, the embedding vectors tend to be more correlated for cells that were cultured and imaged during the same week than for those from different weeks, despite having identical drug compounds applied in both cases. In this case, the particular batch in which a set of experiments were conducted constitutes the domain of the data; an ideal set of image embeddings should contain only the relevant biological information (e.g. drug effects). We develop a general framework for adjusting the image embeddings in order to `forget’ domain-specific information while preserving relevant biological information. To achieve this, we minimize a loss function based on distances between marginal distributions (such as the Wasserstein distance) of embeddings across domains for each replicated treatment. For the dataset we present results with, the only replicated treatment happens to be the negative control treatment, for which we do not expect any treatment-induced cell morphology changes. We find that for our transformed embeddings (i) the underlying geometric structure is not only preserved but the embeddings also carry improved biological signal; and (ii) less domain-specific information is present. \|
Tasks
Published	2017-11-02
URL	https://arxiv.org/abs/1711.00882v2
PDF	https://arxiv.org/pdf/1711.00882v2.pdf
PWC	https://paperswithcode.com/paper/correcting-nuisance-variation-using
Repo
Framework

Method and apparatus for automatic text input insertion in digital devices with a restricted number of keys


Title	Method and apparatus for automatic text input insertion in digital devices with a restricted number of keys
Authors	Nikolaos Tselios, Manolis Maragoudakis
Abstract	A device which contains number of symbol input keys, where the number of available keys is less than the number of symbols of an alphabet of any given language, screen, and dynamic reordering table of the symbols which are mapped onto those keys, according to a disambiguation method based on the previously entered symbols. The device incorporates a previously entered keystrokes tracking mechanism, and the key selected by the user detector, as well as a mechanism to select the dynamic symbol reordering mapped onto this key according to the information contained to the reordering table. The reordering table occurs from a disambiguation method which reorders the symbol appearance. The reordering information occurs from Bayesian Belief network construction and training from text corpora of the specific language.
Tasks
Published	2017-07-29
URL	http://arxiv.org/abs/1707.09487v1
PDF	http://arxiv.org/pdf/1707.09487v1.pdf
PWC	https://paperswithcode.com/paper/method-and-apparatus-for-automatic-text-input
Repo
Framework

A correlational analysis of multiagent sensorimotor interactions: clustering autonomous and controllable entities


Title	A correlational analysis of multiagent sensorimotor interactions: clustering autonomous and controllable entities
Authors	M. Sánchez-Fibla, C. Moulin-Frier, X. Arsiwalla, P. Verschure
Abstract	A first step to reach Theory of Mind (ToM) abilities (attribution of beliefs to others) in synthetic agents through sensorimotor interactions, would be to tag sensory data with agent typology and action intentions: autonomous agent X moved an object under the box. We propose a dual arm robotic setup in which ToM could be probed. We then discuss what measures can be extracted from sensorimotor interaction data (based on a correlation analysis) in the proposed setup that allow to distinguish self than other and other/inanimate from other/active with intentions. We finally discuss what elements are missing in current cognitive architectures to be able to acquire ToM abilities in synthetic agents from sensorimotor interactions, bottom-up from reactive agent interaction behaviors and top-down from the optimization of social behaviour and cooperation.
Tasks
Published	2017-11-22
URL	http://arxiv.org/abs/1711.08333v1
PDF	http://arxiv.org/pdf/1711.08333v1.pdf
PWC	https://paperswithcode.com/paper/a-correlational-analysis-of-multiagent
Repo
Framework

Kinship Verification from Videos using Spatio-Temporal Texture Features and Deep Learning


Title	Kinship Verification from Videos using Spatio-Temporal Texture Features and Deep Learning
Authors	Elhocine Boutellaa, Miguel Bordallo López, Samy Ait-Aoudia, Xiaoyi Feng, Abdenour Hadid
Abstract	Automatic kinship verification using facial images is a relatively new and challenging research problem in computer vision. It consists in automatically predicting whether two persons have a biological kin relation by examining their facial attributes. While most of the existing works extract shallow handcrafted features from still face images, we approach this problem from spatio-temporal point of view and explore the use of both shallow texture features and deep features for characterizing faces. Promising results, especially those of deep features, are obtained on the benchmark UvA-NEMO Smile database. Our extensive experiments also show the superiority of using videos over still images, hence pointing out the important role of facial dynamics in kinship verification. Furthermore, the fusion of the two types of features (i.e. shallow spatio-temporal texture features and deep features) shows significant performance improvements compared to state-of-the-art methods.
Tasks
Published	2017-08-14
URL	http://arxiv.org/abs/1708.04069v1
PDF	http://arxiv.org/pdf/1708.04069v1.pdf
PWC	https://paperswithcode.com/paper/kinship-verification-from-videos-using-spatio
Repo
Framework

Rapid and Robust Automated Macroscopic Wood Identification System using Smartphone with Macro-lens


Title	Rapid and Robust Automated Macroscopic Wood Identification System using Smartphone with Macro-lens
Authors	Xin Jie Tang, Yong Haur Tay, Nordahlia Abdullah Siam, Seng Choon Lim
Abstract	Wood Identification has never been more important to serve the purpose of global forest species protection and timber regulation. Macroscopic level wood identification practiced by wood anatomists can identify wood up to genus level. This is sufficient to serve as a frontline identification to fight against illegal wood logging and timber trade for law enforcement authority. However, frontline enforcement official may lack of the accuracy and confidence of a well trained wood anatomist. Hence, computer assisted method such as machine vision methods are developed to do rapid field identification for law enforcement official. In this paper, we proposed a rapid and robust macroscopic wood identification system using machine vision method with off-the-shelf smartphone and retrofitted macro-lens. Our system is cost effective, easily accessible, fast and scalable at the same time provides human-level accuracy on identification. Camera-enabled smartphone with Internet connectivity coupled with a macro-lens provides a simple and effective digital acquisition of macroscopic wood images which are essential for macroscopic wood identification. The images are immediately streamed to a cloud server via Internet connection for identification which are done within seconds.
Tasks
Published	2017-09-24
URL	http://arxiv.org/abs/1709.08154v1
PDF	http://arxiv.org/pdf/1709.08154v1.pdf
PWC	https://paperswithcode.com/paper/rapid-and-robust-automated-macroscopic-wood
Repo
Framework

Adversarially Tuned Scene Generation


Title	Adversarially Tuned Scene Generation
Authors	V S R Veeravasarapu, Constantin Rothkopf, Ramesh Visvanathan
Abstract	Generalization performance of trained computer vision systems that use computer graphics (CG) generated data is not yet effective due to the concept of ‘domain-shift’ between virtual and real data. Although simulated data augmented with a few real world samples has been shown to mitigate domain shift and improve transferability of trained models, guiding or bootstrapping the virtual data generation with the distributions learnt from target real world domain is desired, especially in the fields where annotating even few real images is laborious (such as semantic labeling, and intrinsic images etc.). In order to address this problem in an unsupervised manner, our work combines recent advances in CG (which aims to generate stochastic scene layouts coupled with large collections of 3D object models) and generative adversarial training (which aims train generative models by measuring discrepancy between generated and real data in terms of their separability in the space of a deep discriminatively-trained classifier). Our method uses iterative estimation of the posterior density of prior distributions for a generative graphical model. This is done within a rejection sampling framework. Initially, we assume uniform distributions as priors on the parameters of a scene described by a generative graphical model. As iterations proceed the prior distributions get updated to distributions that are closer to the (unknown) distributions of target data. We demonstrate the utility of adversarially tuned scene generation on two real-world benchmark datasets (CityScapes and CamVid) for traffic scene semantic labeling with a deep convolutional net (DeepLab). We realized performance improvements by 2.28 and 3.14 points (using the IoU metric) between the DeepLab models trained on simulated sets prepared from the scene generation models before and after tuning to CityScapes and CamVid respectively.
Tasks	Scene Generation
Published	2017-01-02
URL	http://arxiv.org/abs/1701.00405v2
PDF	http://arxiv.org/pdf/1701.00405v2.pdf
PWC	https://paperswithcode.com/paper/adversarially-tuned-scene-generation
Repo
Framework

Neuro Fuzzy Modelling for Prediction of Consumer Price Index


Title	Neuro Fuzzy Modelling for Prediction of Consumer Price Index
Authors	Godwin Ambukege, Godfrey Justo, Joseph Mushi
Abstract	Economic indicators such as Consumer Price Index (CPI) have frequently used in predicting future economic wealth for financial policy makers of respective country. Most central banks, on guidelines of research studies, have recently adopted an inflation targeting monetary policy regime, which accounts for high requirement for effective prediction model of consumer price index. However, prediction accuracy by numerous studies is still low, which raises a need for improvement. This manuscript presents findings of study that use neuro fuzzy technique to design a machine-learning model that train and test data to predict a univariate time series CPI. The study establishes a matrix of monthly CPI data from secondary data source of Tanzania National Bureau of Statistics from January 2000 to December 2015 as case study and thereafter conducted simulation experiments on MATLAB whereby ninety five percent (95%) of data used to train the model and five percent (5%) for testing. Furthermore, the study use root mean square error (RMSE) and mean absolute percentage error (MAPE) as error metrics for model evaluation. The results show that the neuro fuzzy model have an architecture of 5:74:1 with Gaussian membership functions (2, 2, 2, 2, 2), provides RMSE of 0.44886 and MAPE 0.23384, which is far better compared to existing research studies.
Tasks	Time Series
Published	2017-10-09
URL	http://arxiv.org/abs/1710.05944v1
PDF	http://arxiv.org/pdf/1710.05944v1.pdf
PWC	https://paperswithcode.com/paper/neuro-fuzzy-modelling-for-prediction-of
Repo
Framework

Machine Learning Topological Invariants with Neural Networks


Title	Machine Learning Topological Invariants with Neural Networks
Authors	Pengfei Zhang, Huitao Shen, Hui Zhai
Abstract	In this Letter we supervisedly train neural networks to distinguish different topological phases in the context of topological band insulators. After training with Hamiltonians of one-dimensional insulators with chiral symmetry, the neural network can predict their topological winding numbers with nearly 100% accuracy, even for Hamiltonians with larger winding numbers that are not included in the training data. These results show a remarkable success that the neural network can capture the global and nonlinear topological features of quantum phases from local inputs. By opening up the neural network, we confirm that the network does learn the discrete version of the winding number formula. We also make a couple of remarks regarding the role of the symmetry and the opposite effect of regularization techniques when applying machine learning to physical systems.
Tasks
Published	2017-08-30
URL	http://arxiv.org/abs/1708.09401v3
PDF	http://arxiv.org/pdf/1708.09401v3.pdf
PWC	https://paperswithcode.com/paper/machine-learning-topological-invariants-with
Repo
Framework