Paper Group ANR 126
ARTiS: Appearance-based Action Recognition in Task Space for Real-Time Human-Robot Collaboration. Graded Entailment for Compositional Distributional Semantics. Learning Crosslingual Word Embeddings without Bilingual Corpora. Generating machine-executable plans from end-user’s natural-language instructions. A Nonparametric Bayesian Approach for Spok …
ARTiS: Appearance-based Action Recognition in Task Space for Real-Time Human-Robot Collaboration
| Title | ARTiS: Appearance-based Action Recognition in Task Space for Real-Time Human-Robot Collaboration |
| Authors | Markus Eich, Sareh Shirazi, Gordon Wyeth |
| Abstract | To have a robot actively supporting a human during a collaborative task, it is crucial that robots are able to identify the current action in order to predict the next one. Common approaches make use of high-level knowledge, such as object affordances, semantics or understanding of actions in terms of pre- and post-conditions. These approaches often require hand-coded a priori knowledge, time- and resource-intensive or supervised learning techniques. We propose to reframe this problem as an appearance-based place recognition problem. In our framework, we regard sequences of visual images of human actions as a map in analogy to the visual place recognition problem. Observing the task for the second time, our approach is able to recognize pre-observed actions in a one-shot learning approach and is thereby able to recognize the current observation in the task space. We propose two new methods for creating and aligning action observations within a task map. We compare and verify our approaches with real data of humans assembling several types of IKEA flat packs. |
| Tasks | One-Shot Learning, Temporal Action Localization, Visual Place Recognition |
| Published | 2016-10-18 |
| URL | http://arxiv.org/abs/1610.05432v2 |
| http://arxiv.org/pdf/1610.05432v2.pdf | |
| PWC | https://paperswithcode.com/paper/artis-appearance-based-action-recognition-in |
| Repo | |
| Framework | |
Graded Entailment for Compositional Distributional Semantics
| Title | Graded Entailment for Compositional Distributional Semantics |
| Authors | Desislava Bankova, Bob Coecke, Martha Lewis, Daniel Marsden |
| Abstract | The categorical compositional distributional model of natural language provides a conceptually motivated procedure to compute the meaning of sentences, given grammatical structure and the meanings of its words. This approach has outperformed other models in mainstream empirical language processing tasks. However, until recently it has lacked the crucial feature of lexical entailment – as do other distributional models of meaning. In this paper we solve the problem of entailment for categorical compositional distributional semantics. Taking advantage of the abstract categorical framework allows us to vary our choice of model. This enables the introduction of a notion of entailment, exploiting ideas from the categorical semantics of partial knowledge in quantum computation. The new model of language uses density matrices, on which we introduce a novel robust graded order capturing the entailment strength between concepts. This graded measure emerges from a general framework for approximate entailment, induced by any commutative monoid. Quantum logic embeds in our graded order. Our main theorem shows that entailment strength lifts compositionally to the sentence level, giving a lower bound on sentence entailment. We describe the essential properties of graded entailment such as continuity, and provide a procedure for calculating entailment strength. |
| Tasks | |
| Published | 2016-01-19 |
| URL | http://arxiv.org/abs/1601.04908v2 |
| http://arxiv.org/pdf/1601.04908v2.pdf | |
| PWC | https://paperswithcode.com/paper/graded-entailment-for-compositional |
| Repo | |
| Framework | |
Learning Crosslingual Word Embeddings without Bilingual Corpora
| Title | Learning Crosslingual Word Embeddings without Bilingual Corpora |
| Authors | Long Duong, Hiroshi Kanayama, Tengfei Ma, Steven Bird, Trevor Cohn |
| Abstract | Crosslingual word embeddings represent lexical items from different languages in the same vector space, enabling transfer of NLP tools. However, previous attempts had expensive resource requirements, difficulty incorporating monolingual data or were unable to handle polysemy. We address these drawbacks in our method which takes advantage of a high coverage dictionary in an EM style training algorithm over monolingual corpora in two languages. Our model achieves state-of-the-art performance on bilingual lexicon induction task exceeding models using large bilingual corpora, and competitive results on the monolingual word similarity and cross-lingual document classification task. |
| Tasks | Cross-Lingual Document Classification, Document Classification, Word Embeddings |
| Published | 2016-06-30 |
| URL | http://arxiv.org/abs/1606.09403v1 |
| http://arxiv.org/pdf/1606.09403v1.pdf | |
| PWC | https://paperswithcode.com/paper/learning-crosslingual-word-embeddings-without |
| Repo | |
| Framework | |
Generating machine-executable plans from end-user’s natural-language instructions
| Title | Generating machine-executable plans from end-user’s natural-language instructions |
| Authors | Rui Liu, Xiaoli Zhang |
| Abstract | It is critical for advanced manufacturing machines to autonomously execute a task by following an end-user’s natural language (NL) instructions. However, NL instructions are usually ambiguous and abstract so that the machines may misunderstand and incorrectly execute the task. To address this NL-based human-machine communication problem and enable the machines to appropriately execute tasks by following the end-user’s NL instructions, we developed a Machine-Executable-Plan-Generation (exePlan) method. The exePlan method conducts task-centered semantic analysis to extract task-related information from ambiguous NL instructions. In addition, the method specifies machine execution parameters to generate a machine-executable plan by interpreting abstract NL instructions. To evaluate the exePlan method, an industrial robot Baxter was instructed by NL to perform three types of industrial tasks {‘drill a hole’, ‘clean a spot’, ‘install a screw’}. The experiment results proved that the exePlan method was effective in generating machine-executable plans from the end-user’s NL instructions. Such a method has the promise to endow a machine with the ability of NL-instructed task execution. |
| Tasks | |
| Published | 2016-11-20 |
| URL | http://arxiv.org/abs/1611.06468v1 |
| http://arxiv.org/pdf/1611.06468v1.pdf | |
| PWC | https://paperswithcode.com/paper/generating-machine-executable-plans-from-end |
| Repo | |
| Framework | |
A Nonparametric Bayesian Approach for Spoken Term detection by Example Query
| Title | A Nonparametric Bayesian Approach for Spoken Term detection by Example Query |
| Authors | Amir Hossein Harati Nejad Torbati, Joseph Picone |
| Abstract | State of the art speech recognition systems use data-intensive context-dependent phonemes as acoustic units. However, these approaches do not translate well to low resourced languages where large amounts of training data is not available. For such languages, automatic discovery of acoustic units is critical. In this paper, we demonstrate the application of nonparametric Bayesian models to acoustic unit discovery. We show that the discovered units are correlated with phonemes and therefore are linguistically meaningful. We also present a spoken term detection (STD) by example query algorithm based on these automatically learned units. We show that our proposed system produces a P@N of 61.2% and an EER of 13.95% on the TIMIT dataset. The improvement in the EER is 5% while P@N is only slightly lower than the best reported system in the literature. |
| Tasks | Speech Recognition |
| Published | 2016-06-20 |
| URL | http://arxiv.org/abs/1606.05967v1 |
| http://arxiv.org/pdf/1606.05967v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-nonparametric-bayesian-approach-for-spoken |
| Repo | |
| Framework | |
CONDITOR1: Topic Maps and DITA labelling tool for textual documents with historical information
| Title | CONDITOR1: Topic Maps and DITA labelling tool for textual documents with historical information |
| Authors | Piedad Garrido, Jesus Tramullas, Manuel Coll |
| Abstract | Conditor is a software tool which works with textual documents containing historical information. The purpose of this work two-fold: firstly to show the validity of the developed engine to correctly identify and label the entities of the universe of discourse with a labelled-combined XTM-DITA model. Secondly to explain the improvements achieved in the information retrieval process thanks to the use of a object-oriented database (JPOX) as well as its integration into the Lucene-type database search process to not only accomplish more accurate searches, but to also help the future development of a recommender system. We finish with a brief demo in a 3D-graph of the results of the aforementioned search. |
| Tasks | Information Retrieval, Recommendation Systems |
| Published | 2016-03-23 |
| URL | http://arxiv.org/abs/1603.07313v1 |
| http://arxiv.org/pdf/1603.07313v1.pdf | |
| PWC | https://paperswithcode.com/paper/conditor1-topic-maps-and-dita-labelling-tool |
| Repo | |
| Framework | |
Deep Learning with Darwin: Evolutionary Synthesis of Deep Neural Networks
| Title | Deep Learning with Darwin: Evolutionary Synthesis of Deep Neural Networks |
| Authors | Mohammad Javad Shafiee, Akshaya Mishra, Alexander Wong |
| Abstract | Taking inspiration from biological evolution, we explore the idea of “Can deep neural networks evolve naturally over successive generations into highly efficient deep neural networks?” by introducing the notion of synthesizing new highly efficient, yet powerful deep neural networks over successive generations via an evolutionary process from ancestor deep neural networks. The architectural traits of ancestor deep neural networks are encoded using synaptic probability models, which can be viewed as the DNA' of these networks. New descendant networks with differing network architectures are synthesized based on these synaptic probability models from the ancestor networks and computational environmental factor models, in a random manner to mimic heredity, natural selection, and random mutation. These offspring networks are then trained into fully functional networks, like one would train a newborn, and have more efficient, more diverse network architectures than their ancestor networks, while achieving powerful modeling capabilities. Experimental results for the task of visual saliency demonstrated that the synthesized evolved’ offspring networks can achieve state-of-the-art performance while having network architectures that are significantly more efficient (with a staggering $\sim$48-fold decrease in synapses by the fourth generation) compared to the original ancestor network. |
| Tasks | |
| Published | 2016-06-14 |
| URL | http://arxiv.org/abs/1606.04393v3 |
| http://arxiv.org/pdf/1606.04393v3.pdf | |
| PWC | https://paperswithcode.com/paper/deep-learning-with-darwin-evolutionary |
| Repo | |
| Framework | |
Dynamic Stacked Generalization for Node Classification on Networks
| Title | Dynamic Stacked Generalization for Node Classification on Networks |
| Authors | Zhen Han, Alyson Wilson |
| Abstract | We propose a novel stacked generalization (stacking) method as a dynamic ensemble technique using a pool of heterogeneous classifiers for node label classification on networks. The proposed method assigns component models a set of functional coefficients, which can vary smoothly with certain topological features of a node. Compared to the traditional stacking model, the proposed method can dynamically adjust the weights of individual models as we move across the graph and provide a more versatile and significantly more accurate stacking model for label prediction on a network. We demonstrate the benefits of the proposed model using both a simulation study and real data analysis. |
| Tasks | Node Classification |
| Published | 2016-10-16 |
| URL | http://arxiv.org/abs/1610.04804v1 |
| http://arxiv.org/pdf/1610.04804v1.pdf | |
| PWC | https://paperswithcode.com/paper/dynamic-stacked-generalization-for-node |
| Repo | |
| Framework | |
Reactive Policies with Planning for Action Languages
| Title | Reactive Policies with Planning for Action Languages |
| Authors | Zeynep G. Saribatur, Thomas Eiter |
| Abstract | We describe a representation in a high-level transition system for policies that express a reactive behavior for the agent. We consider a target decision component that figures out what to do next and an (online) planning capability to compute the plans needed to reach these targets. Our representation allows one to analyze the flow of executing the given reactive policy, and to determine whether it works as expected. Additionally, the flexibility of the representation opens a range of possibilities for designing behaviors. |
| Tasks | |
| Published | 2016-03-31 |
| URL | http://arxiv.org/abs/1603.09495v1 |
| http://arxiv.org/pdf/1603.09495v1.pdf | |
| PWC | https://paperswithcode.com/paper/reactive-policies-with-planning-for-action |
| Repo | |
| Framework | |
Learning in games with continuous action sets and unknown payoff functions
| Title | Learning in games with continuous action sets and unknown payoff functions |
| Authors | Panayotis Mertikopoulos, Zhengyuan Zhou |
| Abstract | This paper examines the convergence of no-regret learning in games with continuous action sets. For concreteness, we focus on learning via “dual averaging”, a widely used class of no-regret learning schemes where players take small steps along their individual payoff gradients and then “mirror” the output back to their action sets. In terms of feedback, we assume that players can only estimate their payoff gradients up to a zero-mean error with bounded variance. To study the convergence of the induced sequence of play, we introduce the notion of variational stability, and we show that stable equilibria are locally attracting with high probability whereas globally stable equilibria are globally attracting with probability 1. We also discuss some applications to mixed-strategy learning in finite games, and we provide explicit estimates of the method’s convergence speed. |
| Tasks | |
| Published | 2016-08-25 |
| URL | http://arxiv.org/abs/1608.07310v2 |
| http://arxiv.org/pdf/1608.07310v2.pdf | |
| PWC | https://paperswithcode.com/paper/learning-in-games-with-continuous-action-sets |
| Repo | |
| Framework | |
Discrete Deep Feature Extraction: A Theory and New Architectures
| Title | Discrete Deep Feature Extraction: A Theory and New Architectures |
| Authors | Thomas Wiatowski, Michael Tschannen, Aleksandar Stanić, Philipp Grohs, Helmut Bölcskei |
| Abstract | First steps towards a mathematical theory of deep convolutional neural networks for feature extraction were made—for the continuous-time case—in Mallat, 2012, and Wiatowski and B"olcskei, 2015. This paper considers the discrete case, introduces new convolutional neural network architectures, and proposes a mathematical framework for their analysis. Specifically, we establish deformation and translation sensitivity results of local and global nature, and we investigate how certain structural properties of the input signal are reflected in the corresponding feature vectors. Our theory applies to general filters and general Lipschitz-continuous non-linearities and pooling operators. Experiments on handwritten digit classification and facial landmark detection—including feature importance evaluation—complement the theoretical findings. |
| Tasks | Facial Landmark Detection, Feature Importance |
| Published | 2016-05-26 |
| URL | http://arxiv.org/abs/1605.08283v1 |
| http://arxiv.org/pdf/1605.08283v1.pdf | |
| PWC | https://paperswithcode.com/paper/discrete-deep-feature-extraction-a-theory-and |
| Repo | |
| Framework | |
A Tutorial on Online Supervised Learning with Applications to Node Classification in Social Networks
| Title | A Tutorial on Online Supervised Learning with Applications to Node Classification in Social Networks |
| Authors | Alexander Rakhlin, Karthik Sridharan |
| Abstract | We revisit the elegant observation of T. Cover ‘65 which, perhaps, is not as well-known to the broader community as it should be. The first goal of the tutorial is to explain—through the prism of this elementary result—how to solve certain sequence prediction problems by modeling sets of solutions rather than the unknown data-generating mechanism. We extend Cover’s observation in several directions and focus on computational aspects of the proposed algorithms. The applicability of the methods is illustrated on several examples, including node classification in a network. The second aim of this tutorial is to demonstrate the following phenomenon: it is possible to predict as well as a combinatorial “benchmark” for which we have a certain multiplicative approximation algorithm, even if the exact computation of the benchmark given all the data is NP-hard. The proposed prediction methods, therefore, circumvent some of the computational difficulties associated with finding the best model given the data. These difficulties arise rather quickly when one attempts to develop a probabilistic model for graph-based or other problems with a combinatorial structure. |
| Tasks | Node Classification |
| Published | 2016-08-31 |
| URL | http://arxiv.org/abs/1608.09014v1 |
| http://arxiv.org/pdf/1608.09014v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-tutorial-on-online-supervised-learning-with |
| Repo | |
| Framework | |
Matrix Completion has No Spurious Local Minimum
| Title | Matrix Completion has No Spurious Local Minimum |
| Authors | Rong Ge, Jason D. Lee, Tengyu Ma |
| Abstract | Matrix completion is a basic machine learning problem that has wide applications, especially in collaborative filtering and recommender systems. Simple non-convex optimization algorithms are popular and effective in practice. Despite recent progress in proving various non-convex algorithms converge from a good initial point, it remains unclear why random or arbitrary initialization suffices in practice. We prove that the commonly used non-convex objective function for \textit{positive semidefinite} matrix completion has no spurious local minima — all local minima must also be global. Therefore, many popular optimization algorithms such as (stochastic) gradient descent can provably solve positive semidefinite matrix completion with \textit{arbitrary} initialization in polynomial time. The result can be generalized to the setting when the observed entries contain noise. We believe that our main proof strategy can be useful for understanding geometric properties of other statistical problems involving partial or noisy observations. |
| Tasks | Matrix Completion, Recommendation Systems |
| Published | 2016-05-24 |
| URL | http://arxiv.org/abs/1605.07272v4 |
| http://arxiv.org/pdf/1605.07272v4.pdf | |
| PWC | https://paperswithcode.com/paper/matrix-completion-has-no-spurious-local |
| Repo | |
| Framework | |
Stereotyping and Bias in the Flickr30K Dataset
| Title | Stereotyping and Bias in the Flickr30K Dataset |
| Authors | Emiel van Miltenburg |
| Abstract | An untested assumption behind the crowdsourced descriptions of the images in the Flickr30K dataset (Young et al., 2014) is that they “focus only on the information that can be obtained from the image alone” (Hodosh et al., 2013, p. 859). This paper presents some evidence against this assumption, and provides a list of biases and unwarranted inferences that can be found in the Flickr30K dataset. Finally, it considers methods to find examples of these, and discusses how we should deal with stereotype-driven descriptions in future applications. |
| Tasks | |
| Published | 2016-05-19 |
| URL | http://arxiv.org/abs/1605.06083v1 |
| http://arxiv.org/pdf/1605.06083v1.pdf | |
| PWC | https://paperswithcode.com/paper/stereotyping-and-bias-in-the-flickr30k |
| Repo | |
| Framework | |
Model Interpolation with Trans-dimensional Random Field Language Models for Speech Recognition
| Title | Model Interpolation with Trans-dimensional Random Field Language Models for Speech Recognition |
| Authors | Bin Wang, Zhijian Ou, Yong He, Akinori Kawamura |
| Abstract | The dominant language models (LMs) such as n-gram and neural network (NN) models represent sentence probabilities in terms of conditionals. In contrast, a new trans-dimensional random field (TRF) LM has been recently introduced to show superior performances, where the whole sentence is modeled as a random field. In this paper, we examine how the TRF models can be interpolated with the NN models, and obtain 12.1% and 17.9% relative error rate reductions over 6-gram LMs for English and Chinese speech recognition respectively through log-linear combination. |
| Tasks | Speech Recognition |
| Published | 2016-03-30 |
| URL | http://arxiv.org/abs/1603.09170v5 |
| http://arxiv.org/pdf/1603.09170v5.pdf | |
| PWC | https://paperswithcode.com/paper/model-interpolation-with-trans-dimensional |
| Repo | |
| Framework | |
