July 28, 2019

2837 words 14 mins read

Paper Group ANR 361

High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference. Detecting Off-topic Responses to Visual Prompts. Clustering is semidefinitely not that hard: Nonnegative SDP for manifold disentangling. How deep learning works –The geometry of deep learning. Supervised Typing of Big Graphs using Semanti …

High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference


Title	High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference
Authors	Xiaoguang Han, Zhen Li, Haibin Huang, Evangelos Kalogerakis, Yizhou Yu
Abstract	We propose a data-driven method for recovering miss-ing parts of 3D shapes. Our method is based on a new deep learning architecture consisting of two sub-networks: a global structure inference network and a local geometry refinement network. The global structure inference network incorporates a long short-term memorized context fusion module (LSTM-CF) that infers the global structure of the shape based on multi-view depth information provided as part of the input. It also includes a 3D fully convolutional (3DFCN) module that further enriches the global structure representation according to volumetric information in the input. Under the guidance of the global structure network, the local geometry refinement network takes as input lo-cal 3D patches around missing regions, and progressively produces a high-resolution, complete surface through a volumetric encoder-decoder architecture. Our method jointly trains the global structure inference and local geometry refinement networks in an end-to-end manner. We perform qualitative and quantitative evaluations on six object categories, demonstrating that our method outperforms existing state-of-the-art work on shape completion.
Tasks
Published	2017-09-22
URL	http://arxiv.org/abs/1709.07599v1
PDF	http://arxiv.org/pdf/1709.07599v1.pdf
PWC	https://paperswithcode.com/paper/high-resolution-shape-completion-using-deep
Repo
Framework

Detecting Off-topic Responses to Visual Prompts


Title	Detecting Off-topic Responses to Visual Prompts
Authors	Marek Rei
Abstract	Automated methods for essay scoring have made great progress in recent years, achieving accuracies very close to human annotators. However, a known weakness of such automated scorers is not taking into account the semantic relevance of the submitted text. While there is existing work on detecting answer relevance given a textual prompt, very little previous research has been done to incorporate visual writing prompts. We propose a neural architecture and several extensions for detecting off-topic responses to visual prompts and evaluate it on a dataset of texts written by language learners.
Tasks
Published	2017-07-17
URL	http://arxiv.org/abs/1707.05233v1
PDF	http://arxiv.org/pdf/1707.05233v1.pdf
PWC	https://paperswithcode.com/paper/detecting-off-topic-responses-to-visual
Repo
Framework

Clustering is semidefinitely not that hard: Nonnegative SDP for manifold disentangling


Title	Clustering is semidefinitely not that hard: Nonnegative SDP for manifold disentangling
Authors	Mariano Tepper, Anirvan M. Sengupta, Dmitri Chklovskii
Abstract	In solving hard computational problems, semidefinite program (SDP) relaxations often play an important role because they come with a guarantee of optimality. Here, we focus on a popular semidefinite relaxation of K-means clustering which yields the same solution as the non-convex original formulation for well segregated datasets. We report an unexpected finding: when data contains (greater than zero-dimensional) manifolds, the SDP solution captures such geometrical structures. Unlike traditional manifold embedding techniques, our approach does not rely on manually defining a kernel but rather enforces locality via a nonnegativity constraint. We thus call our approach NOnnegative MAnifold Disentangling, or NOMAD. To build an intuitive understanding of its manifold learning capabilities, we develop a theoretical analysis of NOMAD on idealized datasets. While NOMAD is convex and the globally optimal solution can be found by generic SDP solvers with polynomial time complexity, they are too slow for modern datasets. To address this problem, we analyze a non-convex heuristic and present a new, convex and yet efficient, algorithm, based on the conditional gradient method. Our results render NOMAD a versatile, understandable, and powerful tool for manifold learning.
Tasks
Published	2017-06-19
URL	http://arxiv.org/abs/1706.06028v4
PDF	http://arxiv.org/pdf/1706.06028v4.pdf
PWC	https://paperswithcode.com/paper/clustering-is-semidefinitely-not-that-hard
Repo
Framework

How deep learning works –The geometry of deep learning


Title	How deep learning works –The geometry of deep learning
Authors	Xiao Dong, Jiasong Wu, Ling Zhou
Abstract	Why and how that deep learning works well on different tasks remains a mystery from a theoretical perspective. In this paper we draw a geometric picture of the deep learning system by finding its analogies with two existing geometric structures, the geometry of quantum computations and the geometry of the diffeomorphic template matching. In this framework, we give the geometric structures of different deep learning systems including convolutional neural networks, residual networks, recursive neural networks, recurrent neural networks and the equilibrium prapagation framework. We can also analysis the relationship between the geometrical structures and their performance of different networks in an algorithmic level so that the geometric framework may guide the design of the structures and algorithms of deep learning systems.
Tasks
Published	2017-10-30
URL	http://arxiv.org/abs/1710.10784v1
PDF	http://arxiv.org/pdf/1710.10784v1.pdf
PWC	https://paperswithcode.com/paper/how-deep-learning-works-the-geometry-of-deep
Repo
Framework

Supervised Typing of Big Graphs using Semantic Embeddings


Title	Supervised Typing of Big Graphs using Semantic Embeddings
Authors	Mayank Kejriwal, Pedro Szekely
Abstract	We propose a supervised algorithm for generating type embeddings in the same semantic vector space as a given set of entity embeddings. The algorithm is agnostic to the derivation of the underlying entity embeddings. It does not require any manual feature engineering, generalizes well to hundreds of types and achieves near-linear scaling on Big Graphs containing many millions of triples and instances by virtue of an incremental execution. We demonstrate the utility of the embeddings on a type recommendation task, outperforming a non-parametric feature-agnostic baseline while achieving 15x speedup and near-constant memory usage on a full partition of DBpedia. Using state-of-the-art visualization, we illustrate the agreement of our extensionally derived DBpedia type embeddings with the manually curated domain ontology. Finally, we use the embeddings to probabilistically cluster about 4 million DBpedia instances into 415 types in the DBpedia ontology.
Tasks	Entity Embeddings, Feature Engineering
Published	2017-03-22
URL	http://arxiv.org/abs/1703.07805v1
PDF	http://arxiv.org/pdf/1703.07805v1.pdf
PWC	https://paperswithcode.com/paper/supervised-typing-of-big-graphs-using
Repo
Framework

Multi-View Image Generation from a Single-View


Title	Multi-View Image Generation from a Single-View
Authors	Bo Zhao, Xiao Wu, Zhi-Qi Cheng, Hao Liu, Zequn Jie, Jiashi Feng
Abstract	This paper addresses a challenging problem – how to generate multi-view cloth images from only a single view input. To generate realistic-looking images with different views from the input, we propose a new image generation model termed VariGANs that combines the strengths of the variational inference and the Generative Adversarial Networks (GANs). Our proposed VariGANs model generates the target image in a coarse-to-fine manner instead of a single pass which suffers from severe artifacts. It first performs variational inference to model global appearance of the object (e.g., shape and color) and produce a coarse image with a different view. Conditioned on the generated low resolution images, it then proceeds to perform adversarial learning to fill details and generate images of consistent details with the input. Extensive experiments conducted on two clothing datasets, MVC and DeepFashion, have demonstrated that images of a novel view generated by our model are more plausible than those generated by existing approaches, in terms of more consistent global appearance as well as richer and sharper details.
Tasks	Image Generation
Published	2017-04-17
URL	http://arxiv.org/abs/1704.04886v4
PDF	http://arxiv.org/pdf/1704.04886v4.pdf
PWC	https://paperswithcode.com/paper/multi-view-image-generation-from-a-single
Repo
Framework

Resilient Learning-Based Control for Synchronization of Passive Multi-Agent Systems under Attack


Title	Resilient Learning-Based Control for Synchronization of Passive Multi-Agent Systems under Attack
Authors	Arash Rahnama, Panos J. Antsaklis
Abstract	In this paper, we show synchronization for a group of output passive agents that communicate with each other according to an underlying communication graph to achieve a common goal. We propose a distributed event-triggered control framework that will guarantee synchronization and considerably decrease the required communication load on the band-limited network. We define a general Byzantine attack on the event-triggered multi-agent network system and characterize its negative effects on synchronization. The Byzantine agents are capable of intelligently falsifying their data and manipulating the underlying communication graph by altering their respective control feedback weights. We introduce a decentralized detection framework and analyze its steady-state and transient performances. We propose a way of identifying individual Byzantine neighbors and a learning-based method of estimating the attack parameters. Lastly, we propose learning-based control approaches to mitigate the negative effects of the adversarial attack.
Tasks	Adversarial Attack
Published	2017-09-28
URL	http://arxiv.org/abs/1709.10142v1
PDF	http://arxiv.org/pdf/1709.10142v1.pdf
PWC	https://paperswithcode.com/paper/resilient-learning-based-control-for
Repo
Framework

Does the Geometry of Word Embeddings Help Document Classification? A Case Study on Persistent Homology Based Representations


Title	Does the Geometry of Word Embeddings Help Document Classification? A Case Study on Persistent Homology Based Representations
Authors	Paul Michel, Abhilasha Ravichander, Shruti Rijhwani
Abstract	We investigate the pertinence of methods from algebraic topology for text data analysis. These methods enable the development of mathematically-principled isometric-invariant mappings from a set of vectors to a document embedding, which is stable with respect to the geometry of the document in the selected metric space. In this work, we evaluate the utility of these topology-based document representations in traditional NLP tasks, specifically document clustering and sentiment classification. We find that the embeddings do not benefit text analysis. In fact, performance is worse than simple techniques like $\textit{tf-idf}$, indicating that the geometry of the document does not provide enough variability for classification on the basis of topic or sentiment in the chosen datasets.
Tasks	Document Classification, Document Embedding, Sentiment Analysis, Word Embeddings
Published	2017-05-31
URL	http://arxiv.org/abs/1705.10900v1
PDF	http://arxiv.org/pdf/1705.10900v1.pdf
PWC	https://paperswithcode.com/paper/does-the-geometry-of-word-embeddings-help-1
Repo
Framework

Learning Sampling Distributions for Robot Motion Planning


Title	Learning Sampling Distributions for Robot Motion Planning
Authors	Brian Ichter, James Harrison, Marco Pavone
Abstract	A defining feature of sampling-based motion planning is the reliance on an implicit representation of the state space, which is enabled by a set of probing samples. Traditionally, these samples are drawn either probabilistically or deterministically to uniformly cover the state space. Yet, the motion of many robotic systems is often restricted to “small” regions of the state space, due to, for example, differential constraints or collision-avoidance constraints. To accelerate the planning process, it is thus desirable to devise non-uniform sampling strategies that favor sampling in those regions where an optimal solution might lie. This paper proposes a methodology for non-uniform sampling, whereby a sampling distribution is learned from demonstrations, and then used to bias sampling. The sampling distribution is computed through a conditional variational autoencoder, allowing sample generation from the latent space conditioned on the specific planning problem. This methodology is general, can be used in combination with any sampling-based planner, and can effectively exploit the underlying structure of a planning problem while maintaining the theoretical guarantees of sampling-based approaches. Specifically, on several planning problems, the proposed methodology is shown to effectively learn representations for the relevant regions of the state space, resulting in an order of magnitude improvement in terms of success rate and convergence to the optimal cost.
Tasks	Motion Planning
Published	2017-09-16
URL	http://arxiv.org/abs/1709.05448v3
PDF	http://arxiv.org/pdf/1709.05448v3.pdf
PWC	https://paperswithcode.com/paper/learning-sampling-distributions-for-robot
Repo
Framework

Direct and Real-Time Cardiovascular Risk Prediction


Title	Direct and Real-Time Cardiovascular Risk Prediction
Authors	Bob D. de Vos, Nikolas Lessmann, Pim A. de Jong, Max A. Viergever, Ivana Isgum
Abstract	Coronary artery calcium (CAC) burden quantified in low-dose chest CT is a predictor of cardiovascular events. We propose an automatic method for CAC quantification, circumventing intermediate segmentation of CAC. The method determines a bounding box around the heart using a ConvNet for localization. Subsequently, a dedicated ConvNet analyzes axial slices within the bounding boxes to determine CAC quantity by regression. A dataset of 1,546 baseline CT scans was used from the National Lung Screening Trial with manually identified CAC. The method achieved an ICC of 0.98 between manual reference and automatically obtained Agatston scores. Stratification of subjects into five cardiovascular risk categories resulted in an accuracy of 85% and Cohen’s linearly weighted $\kappa$ of 0.90. The results demonstrate that real-time quantification of CAC burden in chest CT without the need for segmentation of CAC is possible.
Tasks
Published	2017-12-08
URL	http://arxiv.org/abs/1712.02982v1
PDF	http://arxiv.org/pdf/1712.02982v1.pdf
PWC	https://paperswithcode.com/paper/direct-and-real-time-cardiovascular-risk
Repo
Framework

Good and safe uses of AI Oracles


Title	Good and safe uses of AI Oracles
Authors	Stuart Armstrong, Xavier O’Rorke
Abstract	It is possible that powerful and potentially dangerous artificial intelligence (AI) might be developed in the future. An Oracle is a design which aims to restrain the impact of a potentially dangerous AI by restricting the agent to no actions besides answering questions. Unfortunately, most Oracles will be motivated to gain more control over the world by manipulating users through the content of their answers, and Oracles of potentially high intelligence might be very successful at this \citep{DBLP:journals/corr/AlfonsecaCACAR16}. In this paper we present two designs for Oracles which, even under pessimistic assumptions, will not manipulate their users into releasing them and yet will still be incentivised to provide their users with helpful answers. The first design is the counterfactual Oracle – which choses its answer as if it expected nobody to ever read it. The second design is the low-bandwidth Oracle – which is limited by the quantity of information it can transmit.
Tasks
Published	2017-11-15
URL	http://arxiv.org/abs/1711.05541v5
PDF	http://arxiv.org/pdf/1711.05541v5.pdf
PWC	https://paperswithcode.com/paper/good-and-safe-uses-of-ai-oracles
Repo
Framework

Software-Defined Robotics – Idea & Approach


Title	Software-Defined Robotics – Idea & Approach
Authors	Ali Al-Bayaty
Abstract	The methodology of Software-Defined Robotics hierarchical-based and stand-alone framework can be designed and implemented to program and control different sets of robots, regardless of their manufacturers’ parameters and specifications, with unified commands and communications. This framework approach will increase the capability of (re)programming a specific group of robots during the runtime without affecting the others as desired in the critical missions and industrial operations, expand the shared bandwidth, enhance the reusability of code, leverage the computational processing power, decrease the unnecessary analyses of vast supplemental electrical components for each robot, as well as get advantages of the most state-of-the-art industrial trends in the cloud-based computing, Virtual Machines (VM), and Robot-as-a-Service (RaaS) technologies.
Tasks
Published	2017-08-20
URL	http://arxiv.org/abs/1708.05935v1
PDF	http://arxiv.org/pdf/1708.05935v1.pdf
PWC	https://paperswithcode.com/paper/software-defined-robotics-idea-approach
Repo
Framework

Online Deception Detection Refueled by Real World Data Collection


Title	Online Deception Detection Refueled by Real World Data Collection
Authors	Wenlin Yao, Zeyu Dai, Ruihong Huang, James Caverlee
Abstract	The lack of large realistic datasets presents a bottleneck in online deception detection studies. In this paper, we apply a data collection method based on social network analysis to quickly identify high-quality deceptive and truthful online reviews from Amazon. The dataset contains more than 10,000 deceptive reviews and is diverse in product domains and reviewers. Using this dataset, we explore effective general features for online deception detection that perform well across domains. We demonstrate that with generalized features - advertising speak and writing complexity scores - deception detection performance can be further improved by adding additional deceptive reviews from assorted domains in training. Finally, reviewer level evaluation gives an interesting insight into different deceptive reviewers’ writing styles.
Tasks	Deception Detection
Published	2017-07-28
URL	http://arxiv.org/abs/1707.09406v1
PDF	http://arxiv.org/pdf/1707.09406v1.pdf
PWC	https://paperswithcode.com/paper/online-deception-detection-refueled-by-real
Repo
Framework

Convergence of backpropagation with momentum for network architectures with skip connections


Title	Convergence of backpropagation with momentum for network architectures with skip connections
Authors	Chirag Agarwal, Joe Klobusicky, Dan Schonfeld
Abstract	We study a class of deep neural networks with networks that form a directed acyclic graph (DAG). For backpropagation defined by gradient descent with adaptive momentum, we show weights converge for a large class of nonlinear activation functions. The proof generalizes the results of Wu et al. (2008) who showed convergence for a feed forward network with one hidden layer. For an example of the effectiveness of DAG architectures, we describe an example of compression through an autoencoder, and compare against sequential feed forward networks under several metrics.
Tasks
Published	2017-05-21
URL	https://arxiv.org/abs/1705.07404v4
PDF	https://arxiv.org/pdf/1705.07404v4.pdf
PWC	https://paperswithcode.com/paper/convergence-of-backpropagation-with-momentum
Repo
Framework

Recognition of feature curves on 3D shapes using an algebraic approach to Hough transforms


Title	Recognition of feature curves on 3D shapes using an algebraic approach to Hough transforms
Authors	Maria-Laura Torrente, Silvia Biasotti, Bianca Falcidieno
Abstract	Feature curves are largely adopted to highlight shape features, such as sharp lines, or to divide surfaces into meaningful segments, like convex or concave regions. Extracting these curves is not sufficient to convey prominent and meaningful information about a shape. We have first to separate the curves belonging to features from those caused by noise and then to select the lines, which describe non-trivial portions of a surface. The automatic detection of such features is crucial for the identification and/or annotation of relevant parts of a given shape. To do this, the Hough transform (HT) is a feature extraction technique widely used in image analysis, computer vision and digital image processing, while, for 3D shapes, the extraction of salient feature curves is still an open problem. Thanks to algebraic geometry concepts, the HT technique has been recently extended to include a vast class of algebraic curves, thus proving to be a competitive tool for yielding an explicit representation of the diverse feature lines equations. In the paper, for the first time we apply this novel extension of the HT technique to the realm of 3D shapes in order to identify and localize semantic features like patterns, decorations or anatomical details on 3D objects (both complete and fragments), even in the case of features partially damaged or incomplete. The method recognizes various features, possibly compound, and it selects the most suitable feature profiles among families of algebraic curves.
Tasks
Published	2017-09-28
URL	http://arxiv.org/abs/1709.10177v1
PDF	http://arxiv.org/pdf/1709.10177v1.pdf
PWC	https://paperswithcode.com/paper/recognition-of-feature-curves-on-3d-shapes
Repo
Framework