October 18, 2019

3219 words 16 mins read

Paper Group ANR 476

Learning Scene Gist with Convolutional Neural Networks to Improve Object Recognition. Learning in POMDPs with Monte Carlo Tree Search. 3D CNN-based classification using sMRI and MD-DTI images for Alzheimer disease studies. The State of the Art in Integrating Machine Learning into Visual Analytics. Adaptive User-Oriented Direct Load-Control of Resid …

Learning Scene Gist with Convolutional Neural Networks to Improve Object Recognition


Title	Learning Scene Gist with Convolutional Neural Networks to Improve Object Recognition
Authors	Kevin Wu, Eric Wu, Gabriel Kreiman
Abstract	Advancements in convolutional neural networks (CNNs) have made significant strides toward achieving high performance levels on multiple object recognition tasks. While some approaches utilize information from the entire scene to propose regions of interest, the task of interpreting a particular region or object is still performed independently of other objects and features in the image. Here we demonstrate that a scene’s ‘gist’ can significantly contribute to how well humans can recognize objects. These findings are consistent with the notion that humans foveate on an object and incorporate information from the periphery to aid in recognition. We use a biologically inspired two-part convolutional neural network (‘GistNet’) that models the fovea and periphery to provide a proof-of-principle demonstration that computational object recognition can significantly benefit from the gist of the scene as contextual information. Our model yields accuracy improvements of up to 50% in certain object categories when incorporating contextual gist, while only increasing the original model size by 5%. This proposed model mirrors our intuition about how the human visual system recognizes objects, suggesting specific biologically plausible constraints to improve machine vision and building initial steps towards the challenge of scene understanding.
Tasks	Object Recognition, Scene Understanding
Published	2018-03-06
URL	http://arxiv.org/abs/1803.01967v2
PDF	http://arxiv.org/pdf/1803.01967v2.pdf
PWC	https://paperswithcode.com/paper/learning-scene-gist-with-convolutional-neural
Repo
Framework

Learning in POMDPs with Monte Carlo Tree Search


Title	Learning in POMDPs with Monte Carlo Tree Search
Authors	Sammie Katt, Frans A. Oliehoek, Christopher Amato
Abstract	The POMDP is a powerful framework for reasoning under outcome and information uncertainty, but constructing an accurate POMDP model is difficult. Bayes-Adaptive Partially Observable Markov Decision Processes (BA-POMDPs) extend POMDPs to allow the model to be learned during execution. BA-POMDPs are a Bayesian RL approach that, in principle, allows for an optimal trade-off between exploitation and exploration. Unfortunately, BA-POMDPs are currently impractical to solve for any non-trivial domain. In this paper, we extend the Monte-Carlo Tree Search method POMCP to BA-POMDPs and show that the resulting method, which we call BA-POMCP, is able to tackle problems that previous solution methods have been unable to solve. Additionally, we introduce several techniques that exploit the BA-POMDP structure to improve the efficiency of BA-POMCP along with proof of their convergence.
Tasks
Published	2018-06-14
URL	http://arxiv.org/abs/1806.05631v1
PDF	http://arxiv.org/pdf/1806.05631v1.pdf
PWC	https://paperswithcode.com/paper/learning-in-pomdps-with-monte-carlo-tree
Repo
Framework

3D CNN-based classification using sMRI and MD-DTI images for Alzheimer disease studies


Title	3D CNN-based classification using sMRI and MD-DTI images for Alzheimer disease studies
Authors	Alexander Khvostikov, Karim Aderghal, Jenny Benois-Pineau, Andrey Krylov, Gwenaelle Catheline
Abstract	Computer-aided early diagnosis of Alzheimers Disease (AD) and its prodromal form, Mild Cognitive Impairment (MCI), has been the subject of extensive research in recent years. Some recent studies have shown promising results in the AD and MCI determination using structural and functional Magnetic Resonance Imaging (sMRI, fMRI), Positron Emission Tomography (PET) and Diffusion Tensor Imaging (DTI) modalities. Furthermore, fusion of imaging modalities in a supervised machine learning framework has shown promising direction of research. In this paper we first review major trends in automatic classification methods such as feature extraction based methods as well as deep learning approaches in medical image analysis applied to the field of Alzheimer’s Disease diagnostics. Then we propose our own algorithm for Alzheimer’s Disease diagnostics based on a convolutional neural network and sMRI and DTI modalities fusion on hippocampal ROI using data from the Alzheimers Disease Neuroimaging Initiative (ADNI) database (http://adni.loni.usc.edu). Comparison with a single modality approach shows promising results. We also propose our own method of data augmentation for balancing classes of different size and analyze the impact of the ROI size on the classification results as well.
Tasks	Data Augmentation
Published	2018-01-18
URL	http://arxiv.org/abs/1801.05968v1
PDF	http://arxiv.org/pdf/1801.05968v1.pdf
PWC	https://paperswithcode.com/paper/3d-cnn-based-classification-using-smri-and-md
Repo
Framework

The State of the Art in Integrating Machine Learning into Visual Analytics


Title	The State of the Art in Integrating Machine Learning into Visual Analytics
Authors	A. Endert, W. Ribarsky, C. Turkay, W Wong, I. Nabney, I Díaz Blanco, Fabrice Rossi
Abstract	Visual analytics systems combine machine learning or other analytic techniques with interactive data visualization to promote sensemaking and analytical reasoning. It is through such techniques that people can make sense of large, complex data. While progress has been made, the tactful combination of machine learning and data visualization is still under-explored. This state-of-the-art report presents a summary of the progress that has been made by highlighting and synthesizing select research advances. Further, it presents opportunities and challenges to enhance the synergy between machine learning and visual analytics for impactful future research directions.
Tasks
Published	2018-02-22
URL	http://arxiv.org/abs/1802.07954v1
PDF	http://arxiv.org/pdf/1802.07954v1.pdf
PWC	https://paperswithcode.com/paper/the-state-of-the-art-in-integrating-machine
Repo
Framework

Adaptive User-Oriented Direct Load-Control of Residential Flexible Devices


Title	Adaptive User-Oriented Direct Load-Control of Residential Flexible Devices
Authors	Davide Frazzetto, Bijay Neupane, Torben Bach Pedersen, Thomas Dyhre Nielsen
Abstract	Demand Response (DR) schemes are effective tools to maintain a dynamic balance in energy markets with higher integration of fluctuating renewable energy sources. DR schemes can be used to harness residential devices’ flexibility and to utilize it to achieve social and financial objectives. However, existing DR schemes suffer from low user participation as they fail at taking into account the users’ requirements. First, DR schemes are highly demanding for the users, as users need to provide direct information, e.g. via surveys, on their energy consumption preferences. Second, the user utility models based on these surveys are hard-coded and do not adapt over time. Third, the existing scheduling techniques require the users to input their energy requirements on a daily basis. As an alternative, this paper proposes a DR scheme for user-oriented direct load-control of residential appliances operations. Instead of relying on user surveys to evaluate the user utility, we propose an online data-driven approach for estimating user utility functions, purely based on available load consumption data, that adaptively models the users’ preference over time. Our scheme is based on a day-ahead scheduling technique that transparently prescribes the users with optimal device operation schedules that take into account both financial benefits and user-perceived quality of service. To model day-ahead user energy demand and flexibility, we propose a probabilistic approach for generating flexibility models under uncertainty. Results on both real-world and simulated datasets show that our DR scheme can provide significant financial benefits while preserving the user-perceived quality of service.
Tasks
Published	2018-05-09
URL	http://arxiv.org/abs/1805.05470v1
PDF	http://arxiv.org/pdf/1805.05470v1.pdf
PWC	https://paperswithcode.com/paper/adaptive-user-oriented-direct-load-control-of
Repo
Framework

Non-Learning based Deep Parallel MRI Reconstruction (NLDpMRI)


Title	Non-Learning based Deep Parallel MRI Reconstruction (NLDpMRI)
Authors	Ali Pour Yazdanpanah, Onur Afacan, Simon K. Warfield
Abstract	Fast data acquisition in Magnetic Resonance Imaging (MRI) is vastly in demand and scan time directly depends on the number of acquired k-space samples. Recently, the deep learning-based MRI reconstruction techniques were suggested to accelerate MR image acquisition. The most common issues in any deep learning-based MRI reconstruction approaches are generalizability and transferability. For different MRI scanner configurations using these approaches, the network must be trained from scratch every time with new training dataset, acquired under new configurations, to be able to provide good reconstruction performance. Here, we propose a new generalized parallel imaging method based on deep neural networks called NLDpMRI to reduce any structured aliasing ambiguities related to the different k-space undersampling patterns for accelerated data acquisition. Two loss functions including non-regularized and regularized are proposed for parallel MRI reconstruction using deep network optimization and we reconstruct MR images by optimizing the proposed loss functions over the network parameters. Unlike any deep learning-based MRI reconstruction approaches, our method doesn’t include any training step that the network learns from a large number of training samples and it only needs the single undersampled multi-coil k-space data for reconstruction. Also, the proposed method can handle k-space data with different undersampling patterns, and the different number of coils. Experimental results show that the proposed method outperforms the current state-of-the-art GRAPPA method and the deep learning-based variational network method.
Tasks
Published	2018-08-06
URL	http://arxiv.org/abs/1808.02122v3
PDF	http://arxiv.org/pdf/1808.02122v3.pdf
PWC	https://paperswithcode.com/paper/non-learning-based-deep-parallel-mri
Repo
Framework

Inferring relevant features: from QFT to PCA


Title	Inferring relevant features: from QFT to PCA
Authors	Cédric Bény
Abstract	In many-body physics, renormalization techniques are used to extract aspects of a statistical or quantum state that are relevant at large scale, or for low energy experiments. Recent works have proposed that these features can be formally identified as those perturbations of the states whose distinguishability most resist coarse-graining. Here, we examine whether this same strategy can be used to identify important features of an unlabeled dataset. This approach indeed results in a technique very similar to kernel PCA (principal component analysis), but with a kernel function that is automatically adapted to the data, or “learned”. We test this approach on handwritten digits, and find that the most relevant features are significantly better for classification than those obtained from a simple gaussian kernel.
Tasks
Published	2018-02-16
URL	http://arxiv.org/abs/1802.05756v1
PDF	http://arxiv.org/pdf/1802.05756v1.pdf
PWC	https://paperswithcode.com/paper/inferring-relevant-features-from-qft-to-pca
Repo
Framework

Discriminative out-of-distribution detection for semantic segmentation


Title	Discriminative out-of-distribution detection for semantic segmentation
Authors	Petra Bevandić, Ivan Krešo, Marin Oršić, Siniša Šegvić
Abstract	Most classification and segmentation datasets assume a closed-world scenario in which predictions are expressed as distribution over a predetermined set of visual classes. However, such assumption implies unavoidable and often unnoticeable failures in presence of out-of-distribution (OOD) input. These failures are bound to happen in most real-life applications since current visual ontologies are far from being comprehensive. We propose to address this issue by discriminative detection of OOD pixels in input data. Different from recent approaches, we avoid to bring any decisions by only observing the training dataset of the primary model trained to solve the desired computer vision task. Instead, we train a dedicated OOD model which discriminates the primary training set from a much larger “background” dataset which approximates the variety of the visual world. We perform our experiments on high resolution natural images in a dense prediction setup. We use several road driving datasets as our training distribution, while we approximate the background distribution with the ILSVRC dataset. We evaluate our approach on WildDash test, which is currently the only public test dataset that includes out-of-distribution images. The obtained results show that the proposed approach succeeds to identify out-of-distribution pixels while outperforming previous work by a wide margin.
Tasks	Out-of-Distribution Detection, Semantic Segmentation
Published	2018-08-23
URL	http://arxiv.org/abs/1808.07703v2
PDF	http://arxiv.org/pdf/1808.07703v2.pdf
PWC	https://paperswithcode.com/paper/discriminative-out-of-distribution-detection
Repo
Framework


Title	SocialGCN: An Efficient Graph Convolutional Network based Model for Social Recommendation
Authors	Le Wu, Peijie Sun, Richang Hong, Yanjie Fu, Xiting Wang, Meng Wang
Abstract	Collaborative Filtering (CF) is one of the most successful approaches for recommender systems. With the emergence of online social networks, social recommendation has become a popular research direction. Most of these social recommendation models utilized each user’s local neighbors’ preferences to alleviate the data sparsity issue in CF. However, they only considered the local neighbors of each user and neglected the process that users’ preferences are influenced as information diffuses in the social network. Recently, Graph Convolutional Networks~(GCN) have shown promising results by modeling the information diffusion process in graphs that leverage both graph structure and node feature information. To this end, in this paper, we propose an effective graph convolutional neural network based model for social recommendation. Based on a classical CF model, the key idea of our proposed model is that we borrow the strengths of GCNs to capture how users’ preferences are influenced by the social diffusion process in social networks. The diffusion of users’ preferences is built on a layer-wise diffusion manner, with the initial user embedding as a function of the current user’s features and a free base user latent vector that is not contained in the user feature. Similarly, each item’s latent vector is also a combination of the item’s free latent vector, as well as its feature representation. Furthermore, we show that our proposed model is flexible when user and item features are not available. Finally, extensive experimental results on two real-world datasets clearly show the effectiveness of our proposed model.
Tasks	Recommendation Systems
Published	2018-11-07
URL	https://arxiv.org/abs/1811.02815v2
PDF	https://arxiv.org/pdf/1811.02815v2.pdf
PWC	https://paperswithcode.com/paper/socialgcn-an-efficient-graph-convolutional
Repo
Framework

Bootstrapped CNNs for Building Segmentation on RGB-D Aerial Imagery


Title	Bootstrapped CNNs for Building Segmentation on RGB-D Aerial Imagery
Authors	Clint Sebastian, Bas Boom, Thijs van Lankveld, Egor Bondarev, Peter H. N. De With
Abstract	Detection of buildings and other objects from aerial images has various applications in urban planning and map making. Automated building detection from aerial imagery is a challenging task, as it is prone to varying lighting conditions, shadows and occlusions. Convolutional Neural Networks (CNNs) are robust against some of these variations, although they fail to distinguish easy and difficult examples. We train a detection algorithm from RGB-D images to obtain a segmented mask by using the CNN architecture DenseNet.First, we improve the performance of the model by applying a statistical re-sampling technique called Bootstrapping and demonstrate that more informative examples are retained. Second, the proposed method outperforms the non-bootstrapped version by utilizing only one-sixth of the original training data and it obtains a precision-recall break-even of 95.10% on our aerial imagery dataset.
Tasks
Published	2018-10-08
URL	http://arxiv.org/abs/1810.03570v1
PDF	http://arxiv.org/pdf/1810.03570v1.pdf
PWC	https://paperswithcode.com/paper/bootstrapped-cnns-for-building-segmentation
Repo
Framework

Neural Tensor Factorization


Title	Neural Tensor Factorization
Authors	Xian Wu, Baoxu Shi, Yuxiao Dong, Chao Huang, Nitesh Chawla
Abstract	Neural collaborative filtering (NCF) and recurrent recommender systems (RRN) have been successful in modeling user-item relational data. However, they are also limited in their assumption of static or sequential modeling of relational data as they do not account for evolving users’ preference over time as well as changes in the underlying factors that drive the change in user-item relationship over time. We address these limitations by proposing a Neural Tensor Factorization (NTF) model for predictive tasks on dynamic relational data. The NTF model generalizes conventional tensor factorization from two perspectives: First, it leverages the long short-term memory architecture to characterize the multi-dimensional temporal interactions on relational data. Second, it incorporates the multi-layer perceptron structure for learning the non-linearities between different latent factors. Our extensive experiments demonstrate the significant improvement in rating prediction and link prediction on dynamic relational data by our NTF model over both neural network based factorization models and other traditional methods.
Tasks	Link Prediction, Recommendation Systems
Published	2018-02-13
URL	http://arxiv.org/abs/1802.04416v1
PDF	http://arxiv.org/pdf/1802.04416v1.pdf
PWC	https://paperswithcode.com/paper/neural-tensor-factorization
Repo
Framework

Behavior Analysis of NLI Models: Uncovering the Influence of Three Factors on Robustness


Title	Behavior Analysis of NLI Models: Uncovering the Influence of Three Factors on Robustness
Authors	Vicente Ivan Sanchez Carmona, Jeff Mitchell, Sebastian Riedel
Abstract	Natural Language Inference is a challenging task that has received substantial attention, and state-of-the-art models now achieve impressive test set performance in the form of accuracy scores. Here, we go beyond this single evaluation metric to examine robustness to semantically-valid alterations to the input data. We identify three factors - insensitivity, polarity and unseen pairs - and compare their impact on three SNLI models under a variety of conditions. Our results demonstrate a number of strengths and weaknesses in the models’ ability to generalise to new in-domain instances. In particular, while strong performance is possible on unseen hypernyms, unseen antonyms are more challenging for all the models. More generally, the models suffer from an insensitivity to certain small but semantically significant alterations, and are also often influenced by simple statistical correlations between words and training labels. Overall, we show that evaluations of NLI models can benefit from studying the influence of factors intrinsic to the models or found in the dataset used.
Tasks	Natural Language Inference
Published	2018-05-11
URL	http://arxiv.org/abs/1805.04212v1
PDF	http://arxiv.org/pdf/1805.04212v1.pdf
PWC	https://paperswithcode.com/paper/behavior-analysis-of-nli-models-uncovering
Repo
Framework

Neural Importance Sampling


Title	Neural Importance Sampling
Authors	Thomas Müller, Brian McWilliams, Fabrice Rousselle, Markus Gross, Jan Novák
Abstract	We propose to use deep neural networks for generating samples in Monte Carlo integration. Our work is based on non-linear independent components estimation (NICE), which we extend in numerous ways to improve performance and enable its application to integration problems. First, we introduce piecewise-polynomial coupling transforms that greatly increase the modeling power of individual coupling layers. Second, we propose to preprocess the inputs of neural networks using one-blob encoding, which stimulates localization of computation and improves inference. Third, we derive a gradient-descent-based optimization for the KL and the $\chi^2$ divergence for the specific application of Monte Carlo integration with unnormalized stochastic estimates of the target distribution. Our approach enables fast and accurate inference and efficient sample generation independently of the dimensionality of the integration domain. We show its benefits on generating natural images and in two applications to light-transport simulation: first, we demonstrate learning of joint path-sampling densities in the primary sample space and importance sampling of multi-dimensional path prefixes thereof. Second, we use our technique to extract conditional directional densities driven by the product of incident illumination and the BSDF in the rendering equation, and we leverage the densities for path guiding. In all applications, our approach yields on-par or higher performance than competing techniques at equal sample count.
Tasks
Published	2018-08-11
URL	https://arxiv.org/abs/1808.03856v5
PDF	https://arxiv.org/pdf/1808.03856v5.pdf
PWC	https://paperswithcode.com/paper/neural-importance-sampling
Repo
Framework

A Short Survey on Sense-Annotated Corpora


Title	A Short Survey on Sense-Annotated Corpora
Authors	Tommaso Pasini, Jose Camacho-Collados
Abstract	Large sense-annotated datasets are increasingly necessary for training deep supervised systems in Word Sense Disambiguation. However, gathering high-quality sense-annotated data for as many instances as possible is a laborious and expensive task. This has led to the proliferation of automatic and semi-automatic methods for overcoming the so-called knowledge-acquisition bottleneck. In this short survey we present an overview of sense-annotated corpora, annotated either manually- or (semi)automatically, that are currently available for different languages and featuring distinct lexical resources as inventory of senses, i.e. WordNet, Wikipedia, BabelNet. Furthermore, we provide the reader with general statistics of each dataset and an analysis of their specific features.
Tasks	Word Sense Disambiguation
Published	2018-02-13
URL	https://arxiv.org/abs/1802.04744v4
PDF	https://arxiv.org/pdf/1802.04744v4.pdf
PWC	https://paperswithcode.com/paper/a-short-survey-on-sense-annotated-corpora-for
Repo
Framework

Unsupervised word sense disambiguation in dynamic semantic spaces


Title	Unsupervised word sense disambiguation in dynamic semantic spaces
Authors	Jean-François Delpech
Abstract	In this paper, we are mainly concerned with the ability to quickly and automatically distinguish word senses in dynamic semantic spaces in which new terms and new senses appear frequently. Such spaces are built ‘“on the fly” from constantly evolving data sets such as Wikipedia, repositories of patent grants and applications, or large sets of legal documents for Technology Assisted Review and e-discovery. This immediacy rules out supervision as well as the use of a priori training sets. We show that the various senses of a term can be automatically made apparent with a simple clustering algorithm, each sense being a vector in the semantic space. While we only consider here semantic spaces built by using random vectors, this algorithm should work with any kind of embedding, provided meaningful similarities between terms can be computed and do fulfill at least the two basic conditions that terms which close meanings have high similarities and terms with unrelated meanings have near-zero similarities.
Tasks	Word Sense Disambiguation
Published	2018-02-07
URL	http://arxiv.org/abs/1802.02605v2
PDF	http://arxiv.org/pdf/1802.02605v2.pdf
PWC	https://paperswithcode.com/paper/unsupervised-word-sense-disambiguation-in
Repo
Framework