Paper Group ANR 70
Nonsparse learning with latent variables. Image Reconstruction using Matched Wavelet Estimated from Data Sensed Compressively using Partial Canonical Identity Matrix. Interactive Music Generation with Positional Constraints using Anticipation-RNNs. Adversarial Feature Augmentation for Unsupervised Domain Adaptation. Deep Domain Adaptation by Geodes …
Nonsparse learning with latent variables
| Title | Nonsparse learning with latent variables |
| Authors | Zemin Zheng, Jinchi Lv, Wei Lin |
| Abstract | As a popular tool for producing meaningful and interpretable models, large-scale sparse learning works efficiently when the underlying structures are indeed or close to sparse. However, naively applying the existing regularization methods can result in misleading outcomes due to model misspecification. In particular, the direct sparsity assumption on coefficient vectors has been questioned in real applications. Therefore, we consider nonsparse learning with the conditional sparsity structure that the coefficient vector becomes sparse after taking out the impacts of certain unobservable latent variables. A new methodology of nonsparse learning with latent variables (NSL) is proposed to simultaneously recover the significant observable predictors and latent factors as well as their effects. We explore a common latent family incorporating population principal components and derive the convergence rates of both sample principal components and their score vectors that hold for a wide class of distributions. With the properly estimated latent variables, properties including model selection consistency and oracle inequalities under various prediction and estimation losses are established for the proposed methodology. Our new methodology and results are evidenced by simulation and real data examples. |
| Tasks | Model Selection, Sparse Learning |
| Published | 2017-10-07 |
| URL | http://arxiv.org/abs/1710.02704v1 |
| http://arxiv.org/pdf/1710.02704v1.pdf | |
| PWC | https://paperswithcode.com/paper/nonsparse-learning-with-latent-variables |
| Repo | |
| Framework | |
Image Reconstruction using Matched Wavelet Estimated from Data Sensed Compressively using Partial Canonical Identity Matrix
| Title | Image Reconstruction using Matched Wavelet Estimated from Data Sensed Compressively using Partial Canonical Identity Matrix |
| Authors | Naushad Ansari, Anubha Gupta |
| Abstract | This paper proposes a joint framework wherein lifting-based, separable, image-matched wavelets are estimated from compressively sensed (CS) images and used for the reconstruction of the same. Matched wavelet can be easily designed if full image is available. Also matched wavelet may provide better reconstruction results in CS application compared to standard wavelet sparsifying basis. Since in CS application, we have compressively sensed image instead of full image, existing methods of designing matched wavelet cannot be used. Thus, we propose a joint framework that estimates matched wavelet from the compressively sensed images and also reconstructs full images. This paper has three significant contributions. First, lifting-based, image-matched separable wavelet is designed from compressively sensed images and is also used to reconstruct the same. Second, a simple sensing matrix is employed to sample data at sub-Nyquist rate such that sensing and reconstruction time is reduced considerably without any noticeable degradation in the reconstruction performance. Third, a new multi-level L-Pyramid wavelet decomposition strategy is provided for separable wavelet implementation on images that leads to improved reconstruction performance. Compared to CS-based reconstruction using standard wavelets with Gaussian sensing matrix and with existing wavelet decomposition strategy, the proposed methodology provides faster and better image reconstruction in compressive sensing application. |
| Tasks | Compressive Sensing, Image Reconstruction |
| Published | 2017-02-07 |
| URL | http://arxiv.org/abs/1702.01970v1 |
| http://arxiv.org/pdf/1702.01970v1.pdf | |
| PWC | https://paperswithcode.com/paper/image-reconstruction-using-matched-wavelet |
| Repo | |
| Framework | |
Interactive Music Generation with Positional Constraints using Anticipation-RNNs
| Title | Interactive Music Generation with Positional Constraints using Anticipation-RNNs |
| Authors | Gaëtan Hadjeres, Frank Nielsen |
| Abstract | Recurrent Neural Networks (RNNS) are now widely used on sequence generation tasks due to their ability to learn long-range dependencies and to generate sequences of arbitrary length. However, their left-to-right generation procedure only allows a limited control from a potential user which makes them unsuitable for interactive and creative usages such as interactive music generation. This paper introduces a novel architecture called Anticipation-RNN which possesses the assets of the RNN-based generative models while allowing to enforce user-defined positional constraints. We demonstrate its efficiency on the task of generating melodies satisfying positional constraints in the style of the soprano parts of the J.S. Bach chorale harmonizations. Sampling using the Anticipation-RNN is of the same order of complexity than sampling from the traditional RNN model. This fast and interactive generation of musical sequences opens ways to devise real-time systems that could be used for creative purposes. |
| Tasks | Music Generation |
| Published | 2017-09-19 |
| URL | http://arxiv.org/abs/1709.06404v1 |
| http://arxiv.org/pdf/1709.06404v1.pdf | |
| PWC | https://paperswithcode.com/paper/interactive-music-generation-with-positional |
| Repo | |
| Framework | |
Adversarial Feature Augmentation for Unsupervised Domain Adaptation
| Title | Adversarial Feature Augmentation for Unsupervised Domain Adaptation |
| Authors | Riccardo Volpi, Pietro Morerio, Silvio Savarese, Vittorio Murino |
| Abstract | Recent works showed that Generative Adversarial Networks (GANs) can be successfully applied in unsupervised domain adaptation, where, given a labeled source dataset and an unlabeled target dataset, the goal is to train powerful classifiers for the target samples. In particular, it was shown that a GAN objective function can be used to learn target features indistinguishable from the source ones. In this work, we extend this framework by (i) forcing the learned feature extractor to be domain-invariant, and (ii) training it through data augmentation in the feature space, namely performing feature augmentation. While data augmentation in the image space is a well established technique in deep learning, feature augmentation has not yet received the same level of attention. We accomplish it by means of a feature generator trained by playing the GAN minimax game against source features. Results show that both enforcing domain-invariance and performing feature augmentation lead to superior or comparable performance to state-of-the-art results in several unsupervised domain adaptation benchmarks. |
| Tasks | Data Augmentation, Domain Adaptation, Unsupervised Domain Adaptation |
| Published | 2017-11-23 |
| URL | http://arxiv.org/abs/1711.08561v2 |
| http://arxiv.org/pdf/1711.08561v2.pdf | |
| PWC | https://paperswithcode.com/paper/adversarial-feature-augmentation-for |
| Repo | |
| Framework | |
Deep Domain Adaptation by Geodesic Distance Minimization
| Title | Deep Domain Adaptation by Geodesic Distance Minimization |
| Authors | Yifei Wang, Wen Li, Dengxin Dai, Luc Van Gool |
| Abstract | In this paper, we propose a new approach called Deep LogCORAL for unsupervised visual domain adaptation. Our work builds on the recently proposed Deep CORAL method, which proposed to train a convolutional neural network and simultaneously minimize the Euclidean distance of convariance matrices between the source and target domains. We propose to use the Riemannian distance, approximated by Log-Euclidean distance, to replace the naive Euclidean distance in Deep CORAL. We also consider first-order information, and minimize the distance of mean vectors between two domains. We build an end-to-end model, in which we minimize both the classification loss, and the domain difference based on the first and second order information between two domains. Our experiments on the benchmark Office dataset demonstrate the improvements of our newly proposed Deep LogCORAL approach over the Deep CORAL method, as well as further improvement when optimizing both orders of information. |
| Tasks | Domain Adaptation |
| Published | 2017-07-13 |
| URL | http://arxiv.org/abs/1707.09842v2 |
| http://arxiv.org/pdf/1707.09842v2.pdf | |
| PWC | https://paperswithcode.com/paper/deep-domain-adaptation-by-geodesic-distance |
| Repo | |
| Framework | |
Person Recognition in Personal Photo Collections
| Title | Person Recognition in Personal Photo Collections |
| Authors | Seong Joon Oh, Rodrigo Benenson, Mario Fritz, Bernt Schiele |
| Abstract | People nowadays share large parts of their personal lives through social media. Being able to automatically recognise people in personal photos may greatly enhance user convenience by easing photo album organisation. For human identification task, however, traditional focus of computer vision has been face recognition and pedestrian re-identification. Person recognition in social media photos sets new challenges for computer vision, including non-cooperative subjects (e.g. backward viewpoints, unusual poses) and great changes in appearance. To tackle this problem, we build a simple person recognition framework that leverages convnet features from multiple image regions (head, body, etc.). We propose new recognition scenarios that focus on the time and appearance gap between training and testing samples. We present an in-depth analysis of the importance of different features according to time and viewpoint generalisability. In the process, we verify that our simple approach achieves the state of the art result on the PIPA benchmark, arguably the largest social media based benchmark for person recognition to date with diverse poses, viewpoints, social groups, and events. Compared the conference version of the paper, this paper additionally presents (1) analysis of a face recogniser (DeepID2+), (2) new method naeil2 that combines the conference version method naeil and DeepID2+ to achieve state of the art results even compared to post-conference works, (3) discussion of related work since the conference version, (4) additional analysis including the head viewpoint-wise breakdown of performance, and (5) results on the open-world setup. |
| Tasks | Face Recognition, Person Recognition |
| Published | 2017-10-09 |
| URL | http://arxiv.org/abs/1710.03224v2 |
| http://arxiv.org/pdf/1710.03224v2.pdf | |
| PWC | https://paperswithcode.com/paper/person-recognition-in-personal-photo |
| Repo | |
| Framework | |
Enhancements of linked data expressiveness for ontologies
| Title | Enhancements of linked data expressiveness for ontologies |
| Authors | Renato Fabbri |
| Abstract | The semantic web has received many contributions of researchers as ontologies which, in this context, i.e. within RDF linked data, are formalized conceptualizations that might use different protocols, such as RDFS, OWL DL and OWL FULL. In this article, we describe new expressive techniques which were found necessary after elaborating dozens of OWL ontologies for the scientific academy, the State and the civil society. They consist in: 1) stating possible uses a property might have without incurring into axioms or restrictions; 2) assigning a level of priority for an element (class, property, triple); 3) correct depiction in diagrams of relations between classes, between individuals which are imperative, and between individuals which are optional; 4) a convenient association between OWL classes and SKOS concepts. We propose specific rules to accomplish these enhancements and exemplify both its use and the difficulties that arise because these techniques are currently not established as standards to the ontology designer. |
| Tasks | |
| Published | 2017-10-27 |
| URL | http://arxiv.org/abs/1710.09952v1 |
| http://arxiv.org/pdf/1710.09952v1.pdf | |
| PWC | https://paperswithcode.com/paper/enhancements-of-linked-data-expressiveness |
| Repo | |
| Framework | |
Building a Web-Scale Dependency-Parsed Corpus from CommonCrawl
| Title | Building a Web-Scale Dependency-Parsed Corpus from CommonCrawl |
| Authors | Alexander Panchenko, Eugen Ruppert, Stefano Faralli, Simone Paolo Ponzetto, Chris Biemann |
| Abstract | We present DepCC, the largest-to-date linguistically analyzed corpus in English including 365 million documents, composed of 252 billion tokens and 7.5 billion of named entity occurrences in 14.3 billion sentences from a web-scale crawl of the \textsc{Common Crawl} project. The sentences are processed with a dependency parser and with a named entity tagger and contain provenance information, enabling various applications ranging from training syntax-based word embeddings to open information extraction and question answering. We built an index of all sentences and their linguistic meta-data enabling quick search across the corpus. We demonstrate the utility of this corpus on the verb similarity task by showing that a distributional model trained on our corpus yields better results than models trained on smaller corpora, like Wikipedia. This distributional model outperforms the state of art models of verb similarity trained on smaller corpora on the SimVerb3500 dataset. |
| Tasks | Open Information Extraction, Question Answering, Word Embeddings |
| Published | 2017-10-04 |
| URL | http://arxiv.org/abs/1710.01779v2 |
| http://arxiv.org/pdf/1710.01779v2.pdf | |
| PWC | https://paperswithcode.com/paper/building-a-web-scale-dependency-parsed-corpus |
| Repo | |
| Framework | |
Maintaining cooperation in complex social dilemmas using deep reinforcement learning
| Title | Maintaining cooperation in complex social dilemmas using deep reinforcement learning |
| Authors | Adam Lerer, Alexander Peysakhovich |
| Abstract | Social dilemmas are situations where individuals face a temptation to increase their payoffs at a cost to total welfare. Building artificially intelligent agents that achieve good outcomes in these situations is important because many real world interactions include a tension between selfish interests and the welfare of others. We show how to modify modern reinforcement learning methods to construct agents that act in ways that are simple to understand, nice (begin by cooperating), provokable (try to avoid being exploited), and forgiving (try to return to mutual cooperation). We show both theoretically and experimentally that such agents can maintain cooperation in Markov social dilemmas. Our construction does not require training methods beyond a modification of self-play, thus if an environment is such that good strategies can be constructed in the zero-sum case (eg. Atari) then we can construct agents that solve social dilemmas in this environment. |
| Tasks | |
| Published | 2017-07-04 |
| URL | http://arxiv.org/abs/1707.01068v4 |
| http://arxiv.org/pdf/1707.01068v4.pdf | |
| PWC | https://paperswithcode.com/paper/maintaining-cooperation-in-complex-social |
| Repo | |
| Framework | |
Ambiguity set and learning via Bregman and Wasserstein
| Title | Ambiguity set and learning via Bregman and Wasserstein |
| Authors | Xin Guo, Johnny Hong, Nan Yang |
| Abstract | Construction of ambiguity set in robust optimization relies on the choice of divergences between probability distributions. In distribution learning, choosing appropriate probability distributions based on observed data is critical for approximating the true distribution. To improve the performance of machine learning models, there has recently been interest in designing objective functions based on Lp-Wasserstein distance rather than the classical Kullback-Leibler (KL) divergence. In this paper, we derive concentration and asymptotic results using Bregman divergence. We propose a novel asymmetric statistical divergence called Wasserstein-Bregman divergence as a generalization of L2-Wasserstein distance. We discuss how these results can be applied to the construction of ambiguity set in robust optimization. |
| Tasks | |
| Published | 2017-05-23 |
| URL | http://arxiv.org/abs/1705.08056v1 |
| http://arxiv.org/pdf/1705.08056v1.pdf | |
| PWC | https://paperswithcode.com/paper/ambiguity-set-and-learning-via-bregman-and |
| Repo | |
| Framework | |
Computational Models of Tutor Feedback in Language Acquisition
| Title | Computational Models of Tutor Feedback in Language Acquisition |
| Authors | Jens Nevens, Michael Spranger |
| Abstract | This paper investigates the role of tutor feedback in language learning using computational models. We compare two dominant paradigms in language learning: interactive learning and cross-situational learning - which differ primarily in the role of social feedback such as gaze or pointing. We analyze the relationship between these two paradigms and propose a new mixed paradigm that combines the two paradigms and allows to test algorithms in experiments that combine no feedback and social feedback. To deal with mixed feedback experiments, we develop new algorithms and show how they perform with respect to traditional knn and prototype approaches. |
| Tasks | Language Acquisition |
| Published | 2017-07-07 |
| URL | http://arxiv.org/abs/1707.02230v1 |
| http://arxiv.org/pdf/1707.02230v1.pdf | |
| PWC | https://paperswithcode.com/paper/computational-models-of-tutor-feedback-in |
| Repo | |
| Framework | |
Improving Slot Filling Performance with Attentive Neural Networks on Dependency Structures
| Title | Improving Slot Filling Performance with Attentive Neural Networks on Dependency Structures |
| Authors | Lifu Huang, Avirup Sil, Heng Ji, Radu Florian |
| Abstract | Slot Filling (SF) aims to extract the values of certain types of attributes (or slots, such as person:cities_of_residence) for a given entity from a large collection of source documents. In this paper we propose an effective DNN architecture for SF with the following new strategies: (1). Take a regularized dependency graph instead of a raw sentence as input to DNN, to compress the wide contexts between query and candidate filler; (2). Incorporate two attention mechanisms: local attention learned from query and candidate filler, and global attention learned from external knowledge bases, to guide the model to better select indicative contexts to determine slot type. Experiments show that this framework outperforms state-of-the-art on both relation extraction (16% absolute F-score gain) and slot filling validation for each individual system (up to 8.5% absolute F-score gain). |
| Tasks | Relation Extraction, Slot Filling |
| Published | 2017-07-04 |
| URL | http://arxiv.org/abs/1707.01075v1 |
| http://arxiv.org/pdf/1707.01075v1.pdf | |
| PWC | https://paperswithcode.com/paper/improving-slot-filling-performance-with |
| Repo | |
| Framework | |
Numerical optimization for Artificial Retina Algorithm
| Title | Numerical optimization for Artificial Retina Algorithm |
| Authors | Maxim Borisyak, Andrey Ustyuzhanin, Denis Derkach, Mikhail Belous |
| Abstract | High-energy physics experiments rely on reconstruction of the trajectories of particles produced at the interaction point. This is a challenging task, especially in the high track multiplicity environment generated by p-p collisions at the LHC energies. A typical event includes hundreds of signal examples (interesting decays) and a significant amount of noise (uninteresting examples). This work describes a modification of the Artificial Retina algorithm for fast track finding: numerical optimization methods were adopted for fast local track search. This approach allows for considerable reduction of the total computational time per event. Test results on simplified simulated model of LHCb VELO (VErtex LOcator) detector are presented. Also this approach is well-suited for implementation of paralleled computations as GPGPU which look very attractive in the context of upcoming detector upgrades. |
| Tasks | |
| Published | 2017-09-25 |
| URL | http://arxiv.org/abs/1709.08610v2 |
| http://arxiv.org/pdf/1709.08610v2.pdf | |
| PWC | https://paperswithcode.com/paper/numerical-optimization-for-artificial-retina |
| Repo | |
| Framework | |
Markov Properties for Graphical Models with Cycles and Latent Variables
| Title | Markov Properties for Graphical Models with Cycles and Latent Variables |
| Authors | Patrick Forré, Joris M. Mooij |
| Abstract | We investigate probabilistic graphical models that allow for both cycles and latent variables. For this we introduce directed graphs with hyperedges (HEDGes), generalizing and combining both marginalized directed acyclic graphs (mDAGs) that can model latent (dependent) variables, and directed mixed graphs (DMGs) that can model cycles. We define and analyse several different Markov properties that relate the graphical structure of a HEDG with a probability distribution on a corresponding product space over the set of nodes, for example factorization properties, structural equations properties, ordered/local/global Markov properties, and marginal versions of these. The various Markov properties for HEDGes are in general not equivalent to each other when cycles or hyperedges are present, in contrast with the simpler case of directed acyclic graphical (DAG) models (also known as Bayesian networks). We show how the Markov properties for HEDGes - and thus the corresponding graphical Markov models - are logically related to each other. |
| Tasks | |
| Published | 2017-10-24 |
| URL | http://arxiv.org/abs/1710.08775v1 |
| http://arxiv.org/pdf/1710.08775v1.pdf | |
| PWC | https://paperswithcode.com/paper/markov-properties-for-graphical-models-with |
| Repo | |
| Framework | |
Differentially Private Learning of Undirected Graphical Models using Collective Graphical Models
| Title | Differentially Private Learning of Undirected Graphical Models using Collective Graphical Models |
| Authors | Garrett Bernstein, Ryan McKenna, Tao Sun, Daniel Sheldon, Michael Hay, Gerome Miklau |
| Abstract | We investigate the problem of learning discrete, undirected graphical models in a differentially private way. We show that the approach of releasing noisy sufficient statistics using the Laplace mechanism achieves a good trade-off between privacy, utility, and practicality. A naive learning algorithm that uses the noisy sufficient statistics “as is” outperforms general-purpose differentially private learning algorithms. However, it has three limitations: it ignores knowledge about the data generating process, rests on uncertain theoretical foundations, and exhibits certain pathologies. We develop a more principled approach that applies the formalism of collective graphical models to perform inference over the true sufficient statistics within an expectation-maximization framework. We show that this learns better models than competing approaches on both synthetic data and on real human mobility data used as a case study. |
| Tasks | |
| Published | 2017-06-14 |
| URL | http://arxiv.org/abs/1706.04646v1 |
| http://arxiv.org/pdf/1706.04646v1.pdf | |
| PWC | https://paperswithcode.com/paper/differentially-private-learning-of-undirected |
| Repo | |
| Framework | |
