Paper Group ANR 221
Spatial-Temporal Multi-Cue Network for Continuous Sign Language Recognition. A tutorial on the range variant of asymmetric numeral systems. On a Generalization of the Average Distance Classifier. Performance Analysis and Comparison of Machine and Deep Learning Algorithms for IoT Data Classification. Rethinking Sparse Gaussian Processes: Bayesian Ap …
Spatial-Temporal Multi-Cue Network for Continuous Sign Language Recognition
| Title | Spatial-Temporal Multi-Cue Network for Continuous Sign Language Recognition |
| Authors | Hao Zhou, Wengang Zhou, Yun Zhou, Houqiang Li |
| Abstract | Despite the recent success of deep learning in continuous sign language recognition (CSLR), deep models typically focus on the most discriminative features, ignoring other potentially non-trivial and informative contents. Such characteristic heavily constrains their capability to learn implicit visual grammars behind the collaboration of different visual cues (i,e., hand shape, facial expression and body posture). By injecting multi-cue learning into neural network design, we propose a spatial-temporal multi-cue (STMC) network to solve the vision-based sequence learning problem. Our STMC network consists of a spatial multi-cue (SMC) module and a temporal multi-cue (TMC) module. The SMC module is dedicated to spatial representation and explicitly decomposes visual features of different cues with the aid of a self-contained pose estimation branch. The TMC module models temporal correlations along two parallel paths, i.e., intra-cue and inter-cue, which aims to preserve the uniqueness and explore the collaboration of multiple cues. Finally, we design a joint optimization strategy to achieve the end-to-end sequence learning of the STMC network. To validate the effectiveness, we perform experiments on three large-scale CSLR benchmarks: PHOENIX-2014, CSL and PHOENIX-2014-T. Experimental results demonstrate that the proposed method achieves new state-of-the-art performance on all three benchmarks. |
| Tasks | Pose Estimation, Sign Language Recognition |
| Published | 2020-02-08 |
| URL | https://arxiv.org/abs/2002.03187v1 |
| https://arxiv.org/pdf/2002.03187v1.pdf | |
| PWC | https://paperswithcode.com/paper/spatial-temporal-multi-cue-network-for |
| Repo | |
| Framework | |
A tutorial on the range variant of asymmetric numeral systems
| Title | A tutorial on the range variant of asymmetric numeral systems |
| Authors | James Townsend |
| Abstract | This paper is intended to be an accessible introduction to the range variant of Asymmetric Numeral Systems (rANS). This version of ANS can be used as a drop in replacement for traditional arithmetic coding (AC). Implementing rANS is more straightforward than AC, and this paper includes pseudo-code which could be converted without too much effort into a working implementation. An example implementation, based on this tutorial, is available at https://raw.githubusercontent.com/j-towns/ans-notes/master/rans.py. After reading (and understanding) this tutorial, the reader should understand how rANS works, and be able to implement it and prove that it attains a near optimal compression rate. |
| Tasks | |
| Published | 2020-01-24 |
| URL | https://arxiv.org/abs/2001.09186v1 |
| https://arxiv.org/pdf/2001.09186v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-tutorial-on-the-range-variant-of-asymmetric |
| Repo | |
| Framework | |
On a Generalization of the Average Distance Classifier
| Title | On a Generalization of the Average Distance Classifier |
| Authors | Sarbojit Roy, Soham Sarkar, Subhajit Dutta |
| Abstract | In high dimension, low sample size (HDLSS)settings, the simple average distance classifier based on the Euclidean distance performs poorly if differences between the locations get masked by the scale differences. To rectify this issue, modifications to the average distance classifier was proposed by Chan and Hall (2009). However, the existing classifiers cannot discriminate when the populations differ in other aspects than locations and scales. In this article, we propose some simple transformations of the average distance classifier to tackle this issue. The resulting classifiers perform quite well even when the underlying populations have the same location and scale. The high-dimensional behaviour of the proposed classifiers is studied theoretically. Numerical experiments with a variety of simulated as well as real data sets exhibit the usefulness of the proposed methodology. |
| Tasks | |
| Published | 2020-01-08 |
| URL | https://arxiv.org/abs/2001.02430v1 |
| https://arxiv.org/pdf/2001.02430v1.pdf | |
| PWC | https://paperswithcode.com/paper/on-a-generalization-of-the-average-distance |
| Repo | |
| Framework | |
Performance Analysis and Comparison of Machine and Deep Learning Algorithms for IoT Data Classification
| Title | Performance Analysis and Comparison of Machine and Deep Learning Algorithms for IoT Data Classification |
| Authors | Meysam Vakili, Mohammad Ghamsari, Masoumeh Rezaei |
| Abstract | In recent years, the growth of Internet of Things (IoT) as an emerging technology has been unbelievable. The number of networkenabled devices in IoT domains is increasing dramatically, leading to the massive production of electronic data. These data contain valuable information which can be used in various areas, such as science, industry, business and even social life. To extract and analyze this information and make IoT systems smart, the only choice is entering artificial intelligence (AI) world and leveraging the power of machine learning and deep learning techniques. This paper evaluates the performance of 11 popular machine and deep learning algorithms for classification task using six IoT-related datasets. These algorithms are compared according to several performance evaluation metrics including precision, recall, f1-score, accuracy, execution time, ROC-AUC score and confusion matrix. A specific experiment is also conducted to assess the convergence speed of developed models. The comprehensive experiments indicated that, considering all performance metrics, Random Forests performed better than other machine learning models, while among deep learning models, ANN and CNN achieved more interesting results. |
| Tasks | |
| Published | 2020-01-27 |
| URL | https://arxiv.org/abs/2001.09636v1 |
| https://arxiv.org/pdf/2001.09636v1.pdf | |
| PWC | https://paperswithcode.com/paper/performance-analysis-and-comparison-of |
| Repo | |
| Framework | |
Rethinking Sparse Gaussian Processes: Bayesian Approaches to Inducing-Variable Approximations
| Title | Rethinking Sparse Gaussian Processes: Bayesian Approaches to Inducing-Variable Approximations |
| Authors | Simone Rossi, Markus Heinonen, Edwin V. Bonilla, Zheyang Shen, Maurizio Filippone |
| Abstract | Variational inference techniques based on inducing variables provide an elegant framework for scalable posterior estimation in Gaussian process (GP) models. Most previous works treat the locations of the inducing variables, i.e. the inducing inputs, as variational hyperparameters, and these are then optimized together with GP covariance hyper-parameters. While some approaches point to the benefits of a Bayesian treatment of GP hyper-parameters, this has been largely overlooked for the inducing inputs. In this work, we show that treating both inducing locations and GP hyper-parameters in a Bayesian way, by inferring their full posterior, further significantly improves performance. Based on stochastic gradient Hamiltonian Monte Carlo, we develop a fully Bayesian approach to scalable GP and deep GP models, and demonstrate its competitive performance through an extensive experimental campaign across several regression and classification problems. |
| Tasks | Gaussian Processes |
| Published | 2020-03-06 |
| URL | https://arxiv.org/abs/2003.03080v2 |
| https://arxiv.org/pdf/2003.03080v2.pdf | |
| PWC | https://paperswithcode.com/paper/rethinking-sparse-gaussian-processes-bayesian |
| Repo | |
| Framework | |
Multiple Flat Projections for Cross-manifold Clustering
| Title | Multiple Flat Projections for Cross-manifold Clustering |
| Authors | Lan Bai, Yuan-Hai Shao, Wei-Jie Chen, Zhen Wang, Nai-Yang Deng |
| Abstract | Cross-manifold clustering is a hard topic and many traditional clustering methods fail because of the cross-manifold structures. In this paper, we propose a Multiple Flat Projections Clustering (MFPC) to deal with cross-manifold clustering problems. In our MFPC, the given samples are projected into multiple subspaces to discover the global structures of the implicit manifolds. Thus, the cross-manifold clusters are distinguished from the various projections. Further, our MFPC is extended to nonlinear manifold clustering via kernel tricks to deal with more complex cross-manifold clustering. A series of non-convex matrix optimization problems in MFPC are solved by a proposed recursive algorithm. The synthetic tests show that our MFPC works on the cross-manifold structures well. Moreover, experimental results on the benchmark datasets show the excellent performance of our MFPC compared with some state-of-the-art clustering methods. |
| Tasks | |
| Published | 2020-02-17 |
| URL | https://arxiv.org/abs/2002.06739v1 |
| https://arxiv.org/pdf/2002.06739v1.pdf | |
| PWC | https://paperswithcode.com/paper/multiple-flat-projections-for-cross-manifold |
| Repo | |
| Framework | |
Online high rank matrix completion
| Title | Online high rank matrix completion |
| Authors | Jicong Fan, Madeleine Udell |
| Abstract | Recent advances in matrix completion enable data imputation in full-rank matrices by exploiting low dimensional (nonlinear) latent structure. In this paper, we develop a new model for high rank matrix completion (HRMC), together with batch and online methods to fit the model and out-of-sample extension to complete new data. The method works by (implicitly) mapping the data into a high dimensional polynomial feature space using the kernel trick; importantly, the data occupies a low dimensional subspace in this feature space, even when the original data matrix is of full-rank. We introduce an explicit parametrization of this low dimensional subspace, and an online fitting procedure, to reduce computational complexity compared to the state of the art. The online method can also handle streaming or sequential data and adapt to non-stationary latent structure. We provide guidance on the sampling rate required these methods to succeed. Experimental results on synthetic data and motion capture data validate the performance of the proposed methods. |
| Tasks | Imputation, Matrix Completion, Motion Capture |
| Published | 2020-02-20 |
| URL | https://arxiv.org/abs/2002.08934v1 |
| https://arxiv.org/pdf/2002.08934v1.pdf | |
| PWC | https://paperswithcode.com/paper/online-high-rank-matrix-completion-1 |
| Repo | |
| Framework | |
Machine Understandable Policies and GDPR Compliance Checking
| Title | Machine Understandable Policies and GDPR Compliance Checking |
| Authors | Piero A. Bonatti, Sabrina Kirrane, Iliana M. Petrova, Luigi Sauro |
| Abstract | The European General Data Protection Regulation (GDPR) calls for technical and organizational measures to support its implementation. Towards this end, the SPECIAL H2020 project aims to provide a set of tools that can be used by data controllers and processors to automatically check if personal data processing and sharing complies with the obligations set forth in the GDPR. The primary contributions of the project include: (i) a policy language that can be used to express consent, business policies, and regulatory obligations; and (ii) two different approaches to automated compliance checking that can be used to demonstrate that data processing performed by data controllers / processors complies with consent provided by data subjects, and business processes comply with regulatory obligations set forth in the GDPR. |
| Tasks | |
| Published | 2020-01-24 |
| URL | https://arxiv.org/abs/2001.08930v1 |
| https://arxiv.org/pdf/2001.08930v1.pdf | |
| PWC | https://paperswithcode.com/paper/machine-understandable-policies-and-gdpr |
| Repo | |
| Framework | |
Proceedings 8th International Workshop on Theorem Proving Components for Educational Software
| Title | Proceedings 8th International Workshop on Theorem Proving Components for Educational Software |
| Authors | Pedro Quaresma, Walther Neuper, João Marcos |
| Abstract | This EPTCS volume contains the proceedings of the ThEdu’19 workshop, promoted on August 25, 2019, as a satellite event of CADE-27, in Natal, Brazil. Representing the eighth installment of the ThEdu series, ThEdu’19 was a vibrant workshop, with an invited talk by Sarah Winkler, four contributions, and the first edition of a Geometry Automated Provers Competition. After the workshop an open call for papers was issued and attracted seven submissions, six of which have been accepted by the reviewers, and collected in the present post-proceedings volume. The ThEdu series pursues the smooth transition from an intuitive way of doing mathematics at secondary school to a more formal approach to the subject in STEM education, while favoring software support for this transition by exploiting the power of theorem-proving technologies. The volume editors hope that this collection of papers will further promote the development of theorem-proving-based software, and that it will collaborate on improving mutual understanding between computer mathematicians and stakeholders in education. |
| Tasks | Automated Theorem Proving |
| Published | 2020-02-27 |
| URL | https://arxiv.org/abs/2002.11895v1 |
| https://arxiv.org/pdf/2002.11895v1.pdf | |
| PWC | https://paperswithcode.com/paper/proceedings-8th-international-workshop-on |
| Repo | |
| Framework | |
Learning to Prove Theorems by Learning to Generate Theorems
| Title | Learning to Prove Theorems by Learning to Generate Theorems |
| Authors | Mingzhe Wang, Jia Deng |
| Abstract | We consider the task of automated theorem proving, a key AI task. Deep learning has shown promise for training theorem provers, but there are limited human-written theorems and proofs available for supervised learning. To address this limitation, we propose to learn a neural generator that automatically synthesizes theorems and proofs for the purpose of training a theorem prover. Experiments on real-world tasks demonstrate that synthetic data from our approach improves the theorem prover and advances the state of the art of automated theorem proving in Metamath. |
| Tasks | Automated Theorem Proving |
| Published | 2020-02-17 |
| URL | https://arxiv.org/abs/2002.07019v1 |
| https://arxiv.org/pdf/2002.07019v1.pdf | |
| PWC | https://paperswithcode.com/paper/learning-to-prove-theorems-by-learning-to-1 |
| Repo | |
| Framework | |
Stochasticity of Deterministic Gradient Descent: Large Learning Rate for Multiscale Objective Function
| Title | Stochasticity of Deterministic Gradient Descent: Large Learning Rate for Multiscale Objective Function |
| Authors | Lingkai Kong, Molei Tao |
| Abstract | This article suggests that deterministic Gradient Descent, which does not use any stochastic gradient approximation, can still exhibit stochastic behaviors. In particular, it shows that if the objective function exhibit multiscale behaviors, then in a large learning rate regime which only resolves the macroscopic but not the microscopic details of the objective, the deterministic GD dynamics can become chaotic and convergent not to a local minimizer but to a statistical distribution. A sufficient condition is also established for approximating this long-time statistical limit by a rescaled Gibbs distribution. Both theoretical and numerical demonstrations are provided, and the theoretical part relies on the construction of a stochastic map that uses bounded noise (as opposed to discretized diffusions). |
| Tasks | |
| Published | 2020-02-14 |
| URL | https://arxiv.org/abs/2002.06189v1 |
| https://arxiv.org/pdf/2002.06189v1.pdf | |
| PWC | https://paperswithcode.com/paper/stochasticity-of-deterministic-gradient |
| Repo | |
| Framework | |
Correlated Initialization for Correlated Data
| Title | Correlated Initialization for Correlated Data |
| Authors | Johannes Schneider |
| Abstract | Spatial data exhibits the property that nearby points are correlated. This holds also for learnt representations across layers, but not for commonly used weight initialization methods. Our theoretical analysis reveals for uncorrelated initialization that (i) flow through layers suffers from much more rapid decrease and (ii) training of individual parameters is subject to more ``zig-zagging’'. We propose multiple methods for correlated initialization. For CNNs, they yield accuracy gains of several per cent in the absence of regularization. Even for properly tuned L2-regularization gains are often possible. | |
| Tasks | L2 Regularization |
| Published | 2020-03-09 |
| URL | https://arxiv.org/abs/2003.04422v1 |
| https://arxiv.org/pdf/2003.04422v1.pdf | |
| PWC | https://paperswithcode.com/paper/correlated-initialization-for-correlated-data |
| Repo | |
| Framework | |
cube2net: Efficient Query-Specific Network Construction with Data Cube Organization
| Title | cube2net: Efficient Query-Specific Network Construction with Data Cube Organization |
| Authors | Carl Yang, Mengxiong Liu, Frank He, Jian Peng, Jiawei Han |
| Abstract | Networks are widely used to model objects with interactions and have enabled various downstream applications. However, in the real world, network mining is often done on particular query sets of objects, which does not require the construction and computation of networks including all objects in the datasets. In this work, for the first time, we propose to address the problem of query-specific network construction, to break the efficiency bottlenecks of existing network mining algorithms and facilitate various downstream tasks. To deal with real-world massive networks with complex attributes, we propose to leverage the well-developed data cube technology to organize network objects w.r.t. their essential attributes. An efficient reinforcement learning algorithm is then developed to automatically explore the data cube structures and construct the optimal query-specific networks. With extensive experiments of two classic network mining tasks on different real-world large datasets, we show that our proposed cube2net pipeline is general, and much more effective and efficient in query-specific network construction, compared with other methods without the leverage of data cube or reinforcement learning. |
| Tasks | |
| Published | 2020-01-18 |
| URL | https://arxiv.org/abs/2002.00841v1 |
| https://arxiv.org/pdf/2002.00841v1.pdf | |
| PWC | https://paperswithcode.com/paper/cube2net-efficient-query-specific-network |
| Repo | |
| Framework | |
EventMapper: Detecting Real-World Physical Events Using Corroborative and Probabilistic Sources
| Title | EventMapper: Detecting Real-World Physical Events Using Corroborative and Probabilistic Sources |
| Authors | Abhijit Suprem, Calton Pu |
| Abstract | The ubiquity of social media makes it a rich source for physical event detection, such as disasters, and as a potential resource for crisis management resource allocation. There have been some recent works on leveraging social media sources for retrospective, after-the-fact event detection of large events such as earthquakes or hurricanes. Similarly, there is a long history of using traditional physical sensors such as climate satellites to perform regional event detection. However, combining social media with corroborative physical sensors for real-time, accurate, and global physical detection has remained unexplored. This paper presents EventMapper, a framework to support event recognition of small yet equally costly events (landslides, flooding, wildfires). EventMapper integrates high-latency, high-accuracy corroborative sources such as physical sensors with low-latency, noisy probabilistic sources such as social media streams to deliver real-time, global event recognition. Furthermore, EventMapper is resilient to the concept drift phenomenon, where machine learning models require continuous fine-tuning to maintain high performance. By exploiting the common features of probabilistic and corroborative sources, EventMapper automates machine learning model updates, maintenance, and fine-tuning. We describe three applications built on EventMapper for landslide, wildfire, and flooding detection. |
| Tasks | |
| Published | 2020-01-23 |
| URL | https://arxiv.org/abs/2001.08700v1 |
| https://arxiv.org/pdf/2001.08700v1.pdf | |
| PWC | https://paperswithcode.com/paper/eventmapper-detecting-real-world-physical |
| Repo | |
| Framework | |
Merging of Ontologies Through Merging of Their Rules
| Title | Merging of Ontologies Through Merging of Their Rules |
| Authors | Olegs Verhodubs |
| Abstract | Ontology merging is important, but not always effective. The main reason, why ontology merging is not effective, is that ontology merging is performed without considering goals. Goals define the way, in which ontologies to be merged more effectively. The paper illustrates ontology merging by means of rules, which are generated from these ontologies. This is necessary for further use in expert systems. |
| Tasks | |
| Published | 2020-01-13 |
| URL | https://arxiv.org/abs/2001.04326v1 |
| https://arxiv.org/pdf/2001.04326v1.pdf | |
| PWC | https://paperswithcode.com/paper/merging-of-ontologies-through-merging-of |
| Repo | |
| Framework | |
