Paper Group ANR 456
Temporal Information Extraction by Predicting Relative Time-lines. How Many Pairwise Preferences Do We Need to Rank A Graph Consistently?. Interactive in-base street model edit: how common GIS software and a database can serve as a custom Graphical User Interface. FA-RPN: Floating Region Proposals for Face Detection. Sensitivity and Generalization …
Temporal Information Extraction by Predicting Relative Time-lines
| Title | Temporal Information Extraction by Predicting Relative Time-lines |
| Authors | Artuur Leeuwenberg, Marie-Francine Moens |
| Abstract | The current leading paradigm for temporal information extraction from text consists of three phases: (1) recognition of events and temporal expressions, (2) recognition of temporal relations among them, and (3) time-line construction from the temporal relations. In contrast to the first two phases, the last phase, time-line construction, received little attention and is the focus of this work. In this paper, we propose a new method to construct a linear time-line from a set of (extracted) temporal relations. But more importantly, we propose a novel paradigm in which we directly predict start and end-points for events from the text, constituting a time-line without going through the intermediate step of prediction of temporal relations as in earlier work. Within this paradigm, we propose two models that predict in linear complexity, and a new training loss using TimeML-style annotations, yielding promising results. |
| Tasks | Temporal Information Extraction |
| Published | 2018-08-28 |
| URL | http://arxiv.org/abs/1808.09401v1 |
| http://arxiv.org/pdf/1808.09401v1.pdf | |
| PWC | https://paperswithcode.com/paper/temporal-information-extraction-by-predicting |
| Repo | |
| Framework | |
How Many Pairwise Preferences Do We Need to Rank A Graph Consistently?
| Title | How Many Pairwise Preferences Do We Need to Rank A Graph Consistently? |
| Authors | Aadirupa Saha, Rakesh Shivanna, Chiranjib Bhattacharyya |
| Abstract | We consider the problem of optimal recovery of true ranking of $n$ items from a randomly chosen subset of their pairwise preferences. It is well known that without any further assumption, one requires a sample size of $\Omega(n^2)$ for the purpose. We analyze the problem with an additional structure of relational graph $G([n],E)$ over the $n$ items added with an assumption of \emph{locality}: Neighboring items are similar in their rankings. Noting the preferential nature of the data, we choose to embed not the graph, but, its \emph{strong product} to capture the pairwise node relationships. Furthermore, unlike existing literature that uses Laplacian embedding for graph based learning problems, we use a richer class of graph embeddings—\emph{orthonormal representations}—that includes (normalized) Laplacian as its special case. Our proposed algorithm, {\it Pref-Rank}, predicts the underlying ranking using an SVM based approach over the chosen embedding of the product graph, and is the first to provide \emph{statistical consistency} on two ranking losses: \emph{Kendall’s tau} and \emph{Spearman’s footrule}, with a required sample complexity of $O(n^2 \chi(\bar{G}))^{\frac{2}{3}}$ pairs, $\chi(\bar{G})$ being the \emph{chromatic number} of the complement graph $\bar{G}$. Clearly, our sample complexity is smaller for dense graphs, with $\chi(\bar G)$ characterizing the degree of node connectivity, which is also intuitive due to the locality assumption e.g. $O(n^\frac{4}{3})$ for union of $k$-cliques, or $O(n^\frac{5}{3})$ for random and power law graphs etc.—a quantity much smaller than the fundamental limit of $\Omega(n^2)$ for large $n$. This, for the first time, relates ranking complexity to structural properties of the graph. We also report experimental evaluations on different synthetic and real datasets, where our algorithm is shown to outperform the state-of-the-art methods. |
| Tasks | |
| Published | 2018-11-06 |
| URL | http://arxiv.org/abs/1811.02161v2 |
| http://arxiv.org/pdf/1811.02161v2.pdf | |
| PWC | https://paperswithcode.com/paper/how-many-pairwise-preferences-do-we-need-to |
| Repo | |
| Framework | |
Interactive in-base street model edit: how common GIS software and a database can serve as a custom Graphical User Interface
| Title | Interactive in-base street model edit: how common GIS software and a database can serve as a custom Graphical User Interface |
| Authors | Remi Cura, Julien Perret, Nicolas Paparoditis |
| Abstract | Our modern world produces an increasing quantity of data, and especially geospatial data, with advance of sensing technologies, and growing complexity and organisation of vector data. Tools are needed to efficiently create and edit those vector geospatial data. Procedural generation has been a tool of choice to generate strongly organised data, yet it may be hard to control. Because those data may be involved to take consequence-full real life decisions, user interactions are required to check data and edit it. The classical process to do so would be to build an adhoc Graphical User Interface (GUI) tool adapted for the model and method being used. This task is difficult, takes a large amount of resources, and is very specific to one model, making it hard to share and re-use. Besides, many common generic GUI already exists to edit vector data, each having its specialities. We propose a change of paradigm; instead of building a specific tool for one task, we use common GIS software as GUIs, and deport the specific interactions from the software to within the database. In this paradigm, GIS software simply modify geometry and attributes of database layers, and those changes are used by the database to perform automated task. This new paradigm has many advantages. The first one is genericity. With in-base interaction, any GIS software can be used to perform edition, whatever the software is a Desktop sofware or a web application. The second is concurrency and coherency. Because interaction is in-base, use of database features allows seamless multi-user work, and can guarantee that the data is in a coherent state. Last we propose tools to facilitate multi-user edits, both during the edit phase (each user knows what areas are edited by other users), and before and after edit (planning of edit, analyse of edited areas). |
| Tasks | |
| Published | 2018-01-17 |
| URL | http://arxiv.org/abs/1801.05800v1 |
| http://arxiv.org/pdf/1801.05800v1.pdf | |
| PWC | https://paperswithcode.com/paper/interactive-in-base-street-model-edit-how |
| Repo | |
| Framework | |
FA-RPN: Floating Region Proposals for Face Detection
| Title | FA-RPN: Floating Region Proposals for Face Detection |
| Authors | Mahyar Najibi, Bharat Singh, Larry S. Davis |
| Abstract | We propose a novel approach for generating region proposals for performing face-detection. Instead of classifying anchor boxes using features from a pixel in the convolutional feature map, we adopt a pooling-based approach for generating region proposals. However, pooling hundreds of thousands of anchors which are evaluated for generating proposals becomes a computational bottleneck during inference. To this end, an efficient anchor placement strategy for reducing the number of anchor-boxes is proposed. We then show that proposals generated by our network (Floating Anchor Region Proposal Network, FA-RPN) are better than RPN for generating region proposals for face detection. We discuss several beneficial features of FA-RPN proposals like iterative refinement, placement of fractional anchors and changing anchors which can be enabled without making any changes to the trained model. Our face detector based on FA-RPN obtains 89.4% mAP with a ResNet-50 backbone on the WIDER dataset. |
| Tasks | Face Detection |
| Published | 2018-12-13 |
| URL | http://arxiv.org/abs/1812.05586v1 |
| http://arxiv.org/pdf/1812.05586v1.pdf | |
| PWC | https://paperswithcode.com/paper/fa-rpn-floating-region-proposals-for-face |
| Repo | |
| Framework | |
Sensitivity and Generalization in Neural Networks: an Empirical Study
| Title | Sensitivity and Generalization in Neural Networks: an Empirical Study |
| Authors | Roman Novak, Yasaman Bahri, Daniel A. Abolafia, Jeffrey Pennington, Jascha Sohl-Dickstein |
| Abstract | In practice it is often found that large over-parameterized neural networks generalize better than their smaller counterparts, an observation that appears to conflict with classical notions of function complexity, which typically favor smaller models. In this work, we investigate this tension between complexity and generalization through an extensive empirical exploration of two natural metrics of complexity related to sensitivity to input perturbations. Our experiments survey thousands of models with various fully-connected architectures, optimizers, and other hyper-parameters, as well as four different image classification datasets. We find that trained neural networks are more robust to input perturbations in the vicinity of the training data manifold, as measured by the norm of the input-output Jacobian of the network, and that it correlates well with generalization. We further establish that factors associated with poor generalization $-$ such as full-batch training or using random labels $-$ correspond to lower robustness, while factors associated with good generalization $-$ such as data augmentation and ReLU non-linearities $-$ give rise to more robust functions. Finally, we demonstrate how the input-output Jacobian norm can be predictive of generalization at the level of individual test points. |
| Tasks | Data Augmentation, Image Classification |
| Published | 2018-02-23 |
| URL | http://arxiv.org/abs/1802.08760v3 |
| http://arxiv.org/pdf/1802.08760v3.pdf | |
| PWC | https://paperswithcode.com/paper/sensitivity-and-generalization-in-neural |
| Repo | |
| Framework | |
Machines hear better when they have ears
| Title | Machines hear better when they have ears |
| Authors | Deepak Baby, Sarah Verhulst |
| Abstract | Deep-neural-network (DNN) based noise suppression systems yield significant improvements over conventional approaches such as spectral subtraction and non-negative matrix factorization, but do not generalize well to noise conditions they were not trained for. In comparison to DNNs, humans show remarkable noise suppression capabilities that yield successful speech intelligibility under various adverse listening conditions and negative signal-to-noise ratios (SNRs). Motivated by the excellent human performance, this paper explores whether numerical models that simulate human cochlear signal processing can be combined with DNNs to improve the robustness of DNN based noise suppression systems. Five cochlear models were coupled to fully-connected and recurrent NN-based noise suppression systems and were trained and evaluated for a variety of noise conditions using objective metrics: perceptual speech quality (PESQ), segmental SNR and cepstral distance. The simulations show that biophysically-inspired cochlear models improve the generalizability of DNN-based noise suppression systems for unseen noise and negative SNRs. This approach thus leads to robust noise suppression systems that are less sensitive to the noise type and noise level. Because cochlear models capture the intrinsic nonlinearities and dynamics of peripheral auditory processing, it is shown here that accounting for their deterministic signal processing improves machine hearing and avoids overtraining of multi-layer DNNs. We hence conclude that machines hear better when realistic cochlear models are used at the input of DNNs. |
| Tasks | |
| Published | 2018-06-01 |
| URL | http://arxiv.org/abs/1806.01145v2 |
| http://arxiv.org/pdf/1806.01145v2.pdf | |
| PWC | https://paperswithcode.com/paper/machines-hear-better-when-they-have-ears |
| Repo | |
| Framework | |
Identifying Bias in AI using Simulation
| Title | Identifying Bias in AI using Simulation |
| Authors | Daniel McDuff, Roger Cheng, Ashish Kapoor |
| Abstract | Machine learned models exhibit bias, often because the datasets used to train them are biased. This presents a serious problem for the deployment of such technology, as the resulting models might perform poorly on populations that are minorities within the training set and ultimately present higher risks to them. We propose to use high-fidelity computer simulations to interrogate and diagnose biases within ML classifiers. We present a framework that leverages Bayesian parameter search to efficiently characterize the high dimensional feature space and more quickly identify weakness in performance. We apply our approach to an example domain, face detection, and show that it can be used to help identify demographic biases in commercial face application programming interfaces (APIs). |
| Tasks | Face Detection |
| Published | 2018-09-30 |
| URL | http://arxiv.org/abs/1810.00471v1 |
| http://arxiv.org/pdf/1810.00471v1.pdf | |
| PWC | https://paperswithcode.com/paper/identifying-bias-in-ai-using-simulation |
| Repo | |
| Framework | |
A Fast and Accurate System for Face Detection, Identification, and Verification
| Title | A Fast and Accurate System for Face Detection, Identification, and Verification |
| Authors | Rajeev Ranjan, Ankan Bansal, Jingxiao Zheng, Hongyu Xu, Joshua Gleason, Boyu Lu, Anirudh Nanduri, Jun-Cheng Chen, Carlos D. Castillo, Rama Chellappa |
| Abstract | The availability of large annotated datasets and affordable computation power have led to impressive improvements in the performance of CNNs on various object detection and recognition benchmarks. These, along with a better understanding of deep learning methods, have also led to improved capabilities of machine understanding of faces. CNNs are able to detect faces, locate facial landmarks, estimate pose, and recognize faces in unconstrained images and videos. In this paper, we describe the details of a deep learning pipeline for unconstrained face identification and verification which achieves state-of-the-art performance on several benchmark datasets. We propose a novel face detector, Deep Pyramid Single Shot Face Detector (DPSSD), which is fast and capable of detecting faces with large scale variations (especially tiny faces). We give design details of the various modules involved in automatic face recognition: face detection, landmark localization and alignment, and face identification/verification. We provide evaluation results of the proposed face detector on challenging unconstrained face detection datasets. Then, we present experimental results for IARPA Janus Benchmarks A, B and C (IJB-A, IJB-B, IJB-C), and the Janus Challenge Set 5 (CS5). |
| Tasks | Face Detection, Face Identification, Face Recognition, Object Detection, Robust Face Recognition |
| Published | 2018-09-20 |
| URL | http://arxiv.org/abs/1809.07586v1 |
| http://arxiv.org/pdf/1809.07586v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-fast-and-accurate-system-for-face-detection |
| Repo | |
| Framework | |
Identification of temporal transition of functional states using recurrent neural networks from functional MRI
| Title | Identification of temporal transition of functional states using recurrent neural networks from functional MRI |
| Authors | Hongming Li, Yong Fan |
| Abstract | Dynamic functional connectivity analysis provides valuable information for understanding brain functional activity underlying different cognitive processes. Besides sliding window based approaches, a variety of methods have been developed to automatically split the entire functional MRI scan into segments by detecting change points of functional signals to facilitate better characterization of temporally dynamic functional connectivity patterns. However, these methods are based on certain assumptions for the functional signals, such as Gaussian distribution, which are not necessarily suitable for the fMRI data. In this study, we develop a deep learning based framework for adaptively detecting temporally dynamic functional state transitions in a data-driven way without any explicit modeling assumptions, by leveraging recent advances in recurrent neural networks (RNNs) for sequence modeling. Particularly, we solve this problem in an anomaly detection framework with an assumption that the functional profile of one single time point could be reliably predicted based on its preceding profiles within stable functional state, while large prediction errors would occur around change points of functional states. We evaluate the proposed method using both task and resting-state fMRI data obtained from the human connectome project and experimental results have demonstrated that the proposed change point detection method could effectively identify change points between different task events and split the resting-state fMRI into segments with distinct functional connectivity patterns. |
| Tasks | Anomaly Detection, Change Point Detection |
| Published | 2018-09-14 |
| URL | http://arxiv.org/abs/1809.05560v1 |
| http://arxiv.org/pdf/1809.05560v1.pdf | |
| PWC | https://paperswithcode.com/paper/identification-of-temporal-transition-of |
| Repo | |
| Framework | |
Supervisory Control of Probabilistic Discrete Event Systems under Partial Observation
| Title | Supervisory Control of Probabilistic Discrete Event Systems under Partial Observation |
| Authors | Weilin Deng, Jingkai Yang, Daowen Qiu |
| Abstract | The supervisory control of probabilistic discrete event systems (PDESs) is investigated under the assumptions that the supervisory controller (supervisor) is probabilistic and has a partial observation. The probabilistic P-supervisor is defined, which specifies a probability distribution on the control patterns for each observation. The notions of the probabilistic controllability and observability are proposed and demonstrated to be a necessary and sufficient conditions for the existence of the probabilistic P-supervisors. Moreover, the polynomial verification algorithms for the probabilistic controllability and observability are put forward. In addition, the infimal probabilistic controllable and observable superlanguage is introduced and computed as the solution of the optimal control problem of PDESs. Several examples are presented to illustrate the results obtained. |
| Tasks | |
| Published | 2018-05-17 |
| URL | http://arxiv.org/abs/1805.07196v1 |
| http://arxiv.org/pdf/1805.07196v1.pdf | |
| PWC | https://paperswithcode.com/paper/supervisory-control-of-probabilistic-discrete |
| Repo | |
| Framework | |
Using Aspect Extraction Approaches to Generate Review Summaries and User Profiles
| Title | Using Aspect Extraction Approaches to Generate Review Summaries and User Profiles |
| Authors | Christopher Mitcheltree, Veronica Wharton, Avneesh Saluja |
| Abstract | Reviews of products or services on Internet marketplace websites contain a rich amount of information. Users often wish to survey reviews or review snippets from the perspective of a certain aspect, which has resulted in a large body of work on aspect identification and extraction from such corpora. In this work, we evaluate a newly-proposed neural model for aspect extraction on two practical tasks. The first is to extract canonical sentences of various aspects from reviews, and is judged by human evaluators against alternatives. A $k$-means baseline does remarkably well in this setting. The second experiment focuses on the suitability of the recovered aspect distributions to represent users by the reviews they have written. Through a set of review reranking experiments, we find that aspect-based profiles can largely capture notions of user preferences, by showing that divergent users generate markedly different review rankings. |
| Tasks | Aspect Extraction |
| Published | 2018-04-23 |
| URL | http://arxiv.org/abs/1804.08666v1 |
| http://arxiv.org/pdf/1804.08666v1.pdf | |
| PWC | https://paperswithcode.com/paper/using-aspect-extraction-approaches-to |
| Repo | |
| Framework | |
Fighting Contextual Bandits with Stochastic Smoothing
| Title | Fighting Contextual Bandits with Stochastic Smoothing |
| Authors | Young Hun Jung, Ambuj Tewari |
| Abstract | We introduce a new stochastic smoothing perspective to study adversarial contextual bandit problems. We propose a general algorithm template that represents random perturbation based algorithms and identify several perturbation distributions that lead to strong regret bounds. Using the idea of smoothness, we provide an $O(\sqrt{T})$ zero-order bound for the vanilla algorithm and an $O(L^{*2/3}_{T})$ first-order bound for the clipped version. These bounds hold when the algorithms use with a variety of distributions that have a bounded hazard rate. Our algorithm template includes EXP4 as a special case corresponding to the Gumbel perturbation. Our regret bounds match existing results for EXP4 without relying on the specific properties of the algorithm. |
| Tasks | Multi-Armed Bandits |
| Published | 2018-10-11 |
| URL | https://arxiv.org/abs/1810.05188v2 |
| https://arxiv.org/pdf/1810.05188v2.pdf | |
| PWC | https://paperswithcode.com/paper/fighting-contextual-bandits-with-stochastic |
| Repo | |
| Framework | |
Short Term Electric Load Forecast with Artificial Neural Networks
| Title | Short Term Electric Load Forecast with Artificial Neural Networks |
| Authors | Cristian Vasar, Iosif Szeidert, Ioan Filip, Gabriela Prostean |
| Abstract | This paper presents issues regarding short term electric load forecasting using feedforward and Elman recurrent neural networks. The study cases were developed using measured data representing electrical energy consume from Banat area. There were considered 35 different types of structure for both feedforward and recurrent network cases. For each type of neural network structure were performed many trainings and best solution was selected. The issue of forecasting the load on short term is essential in the effective energetic consume management in an open market environment. |
| Tasks | Load Forecasting |
| Published | 2018-04-18 |
| URL | http://arxiv.org/abs/1804.06660v1 |
| http://arxiv.org/pdf/1804.06660v1.pdf | |
| PWC | https://paperswithcode.com/paper/short-term-electric-load-forecast-with |
| Repo | |
| Framework | |
Learning Manifolds from Non-stationary Streaming Data
| Title | Learning Manifolds from Non-stationary Streaming Data |
| Authors | Suchismit Mahapatra, Varun Chandola |
| Abstract | Streaming adaptations of manifold learning based dimensionality reduction methods, such as {\em Isomap}, typically assume that the underlying data distribution is stationary. Such methods are not equipped to detect or handle sudden changes or gradual drifts in the distribution that generates the stream. We prove that a Gaussian Process Regression (GPR) model that uses a manifold-specific kernel function and is trained on an initial batch of sufficient size, can closely approximate the state-of-art streaming Isomap algorithm. The predictive variance obtained from the GPR prediction is then shown to be an effective detector of changes in the underlying data distribution. Results on several synthetic and real data sets show that the resulting algorithm can effectively learn lower dimensional representation of high dimensional data in a streaming setting, while identifying shifts in the generative distribution. |
| Tasks | Dimensionality Reduction |
| Published | 2018-04-24 |
| URL | https://arxiv.org/abs/1804.08833v2 |
| https://arxiv.org/pdf/1804.08833v2.pdf | |
| PWC | https://paperswithcode.com/paper/learning-manifolds-from-non-stationary |
| Repo | |
| Framework | |
Transfer Learning versus Multi-agent Learning regarding Distributed Decision-Making in Highway Traffic
| Title | Transfer Learning versus Multi-agent Learning regarding Distributed Decision-Making in Highway Traffic |
| Authors | Mark Schutera, Niklas Goby, Dirk Neumann, Markus Reischl |
| Abstract | Transportation and traffic are currently undergoing a rapid increase in terms of both scale and complexity. At the same time, an increasing share of traffic participants are being transformed into agents driven or supported by artificial intelligence resulting in mixed-intelligence traffic. This work explores the implications of distributed decision-making in mixed-intelligence traffic. The investigations are carried out on the basis of an online-simulated highway scenario, namely the MIT \emph{DeepTraffic} simulation. In the first step traffic agents are trained by means of a deep reinforcement learning approach, being deployed inside an elitist evolutionary algorithm for hyperparameter search. The resulting architectures and training parameters are then utilized in order to either train a single autonomous traffic agent and transfer the learned weights onto a multi-agent scenario or else to conduct multi-agent learning directly. Both learning strategies are evaluated on different ratios of mixed-intelligence traffic. The strategies are assessed according to the average speed of all agents driven by artificial intelligence. Traffic patterns that provoke a reduction in traffic flow are analyzed with respect to the different strategies. |
| Tasks | Decision Making, Transfer Learning |
| Published | 2018-10-19 |
| URL | http://arxiv.org/abs/1810.08515v1 |
| http://arxiv.org/pdf/1810.08515v1.pdf | |
| PWC | https://paperswithcode.com/paper/transfer-learning-versus-multi-agent-learning |
| Repo | |
| Framework | |
