Paper Group NANR 114
Contextual memory bandit for pro-active dialog engagement. PickleTeam! at SemEval-2018 Task 2: English and Spanish Emoji Prediction from Tweets. LSD-Net: Look, Step and Detect for Joint Navigation and Multi-View Recognition with Deep Reinforcement Learning. Sanskrit Word Segmentation Using Character-level Recurrent and Convolutional Neural Networks …
Contextual memory bandit for pro-active dialog engagement
| Title | Contextual memory bandit for pro-active dialog engagement |
| Authors | julien perez, Tomi Silander |
| Abstract | An objective of pro-activity in dialog systems is to enhance the usability of conversational agents by enabling them to initiate conversation on their own. While dialog systems have become increasingly popular during the last couple of years, current task oriented dialog systems are still mainly reactive and users tend to initiate conversations. In this paper, we propose to introduce the paradigm of contextual bandits as framework for pro-active dialog systems. Contextual bandits have been the model of choice for the problem of reward maximization with partial feedback since they fit well to the task description. As a second contribution, we introduce and explore the notion of memory into this paradigm. We propose two differentiable memory models that act as parts of the parametric reward estimation function. The first one, Convolutional Selective Memory Networks, uses a selection of past interactions as part of the decision support. The second model, called Contextual Attentive Memory Network, implements a differentiable attention mechanism over the past interactions of the agent. The goal is to generalize the classic model of contextual bandits to settings where temporal information needs to be incorporated and leveraged in a learnable manner. Finally, we illustrate the usability and performance of our model for building a pro-active mobile assistant through an extensive set of experiments. |
| Tasks | Multi-Armed Bandits |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=SJiHOSeR- |
| https://openreview.net/pdf?id=SJiHOSeR- | |
| PWC | https://paperswithcode.com/paper/contextual-memory-bandit-for-pro-active |
| Repo | |
| Framework | |
PickleTeam! at SemEval-2018 Task 2: English and Spanish Emoji Prediction from Tweets
| Title | PickleTeam! at SemEval-2018 Task 2: English and Spanish Emoji Prediction from Tweets |
| Authors | Daphne Groot, R{'e}mon Kruizinga, Hennie Veldthuis, Simon de Wit, Hessel Haagsma |
| Abstract | We present a system for emoji prediction on English and Spanish tweets, prepared for the SemEval-2018 task on Multilingual Emoji Prediction. We compared the performance of an SVM, LSTM and an ensemble of these two. We found the SVM performed best on our development set with an accuracy of 61.3{%} for English and 83{%} for Spanish. The features used for the SVM are lowercased word n-grams in the range of 1 to 20, tokenised by a TweetTokenizer and stripped of stop words. On the test set, our model achieved an accuracy of 34{%} on English, with a slightly lower score of 29.7{%} accuracy on Spanish. |
| Tasks | Medical Diagnosis |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-1072/ |
| https://www.aclweb.org/anthology/S18-1072 | |
| PWC | https://paperswithcode.com/paper/pickleteam-at-semeval-2018-task-2-english-and |
| Repo | |
| Framework | |
LSD-Net: Look, Step and Detect for Joint Navigation and Multi-View Recognition with Deep Reinforcement Learning
| Title | LSD-Net: Look, Step and Detect for Joint Navigation and Multi-View Recognition with Deep Reinforcement Learning |
| Authors | N dinesh reddy |
| Abstract | Multi-view recognition is the task of classifying an object from multi-view image sequences. Instead of using a single-view for classification, humans generally navigate around a target object to learn its multi-view representation. Motivated by this human behavior, the next best view can be learned by combining object recognition with navigation in complex environments. Since deep reinforcement learning has proven successful in navigation tasks, we propose a novel multi-task reinforcement learning framework for joint multi-view recognition and navigation. Our method uses a hierarchical action space for multi-task reinforcement learning. The framework was evaluated with an environment created from the ModelNet40 dataset. Our results show improvements on object recognition and demonstrate human-like behavior on navigation. |
| Tasks | Object Recognition |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=S1FFLWWCZ |
| https://openreview.net/pdf?id=S1FFLWWCZ | |
| PWC | https://paperswithcode.com/paper/lsd-net-look-step-and-detect-for-joint |
| Repo | |
| Framework | |
Sanskrit Word Segmentation Using Character-level Recurrent and Convolutional Neural Networks
| Title | Sanskrit Word Segmentation Using Character-level Recurrent and Convolutional Neural Networks |
| Authors | Oliver Hellwig, Sebastian Nehrdich |
| Abstract | The paper introduces end-to-end neural network models that tokenize Sanskrit by jointly splitting compounds and resolving phonetic merges (Sandhi). Tokenization of Sanskrit depends on local phonetic and distant semantic features that are incorporated using convolutional and recurrent elements. Contrary to most previous systems, our models do not require feature engineering or extern linguistic resources, but operate solely on parallel versions of raw and segmented text. The models discussed in this paper clearly improve over previous approaches to Sanskrit word segmentation. As they are language agnostic, we will demonstrate that they also outperform the state of the art for the related task of German compound splitting. |
| Tasks | Feature Engineering, Tokenization |
| Published | 2018-10-01 |
| URL | https://www.aclweb.org/anthology/D18-1295/ |
| https://www.aclweb.org/anthology/D18-1295 | |
| PWC | https://paperswithcode.com/paper/sanskrit-word-segmentation-using-character |
| Repo | |
| Framework | |
Differential Privacy for Growing Databases
| Title | Differential Privacy for Growing Databases |
| Authors | Rachel Cummings, Sara Krehbiel, Kevin A. Lai, Uthaipon Tantipongpipat |
| Abstract | The large majority of differentially private algorithms focus on the static setting, where queries are made on an unchanging database. This is unsuitable for the myriad applications involving databases that grow over time. To address this gap in the literature, we consider the dynamic setting, in which new data arrive over time. Previous results in this setting have been limited to answering a single non-adaptive query repeatedly as the database grows. In contrast, we provide tools for richer and more adaptive analysis of growing databases. Our first contribution is a novel modification of the private multiplicative weights algorithm, which provides accurate analysis of exponentially many adaptive linear queries (an expressive query class including all counting queries) for a static database. Our modification maintains the accuracy guarantee of the static setting even as the database grows without bound. Our second contribution is a set of general results which show that many other private and accurate algorithms can be immediately extended to the dynamic setting by rerunning them at appropriate points of data growth with minimal loss of accuracy, even when data growth is unbounded. |
| Tasks | |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/8102-differential-privacy-for-growing-databases |
| http://papers.nips.cc/paper/8102-differential-privacy-for-growing-databases.pdf | |
| PWC | https://paperswithcode.com/paper/differential-privacy-for-growing-databases |
| Repo | |
| Framework | |
Programmable Triangulation Light Curtains
| Title | Programmable Triangulation Light Curtains |
| Authors | Jian Wang, Joseph Bartels, William Whittaker, Aswin C. Sankaranarayanan, Srinivasa G. Narasimhan |
| Abstract | A vehicle on a road or a robot in the field does not need a full-featured 3D depth sensor to detect potential collisions or monitor its blind spot. Instead, it needs to only monitor if any object comes within its near proximity which is an easier task than full depth scanning. We introduce a novel device that monitors the presence of objects on a virtual shell near the device, which we refer to as a light curtain. Light curtains offer a light-weight, resource-efficient and programmable approach to proximity awareness for obstacle avoidance and navigation. They also have additional benefits in terms of improving visibility in fog as well as flexibility in handling light fall-off. Our prototype for generating light curtains works by rapidly rotating a line sensor and a line laser, in synchrony. The device is capable of generating light curtains of various shapes with a range of 20-30m in sunlight (40m under cloudy skies and 50m indoors) and adapts dynamically to the demands of the task. We analyze properties of light curtains and various approaches to optimize their thickness as well as power requirements. We showcase the potential of light curtains using a range of real-world scenarios. |
| Tasks | |
| Published | 2018-09-01 |
| URL | http://openaccess.thecvf.com/content_ECCV_2018/html/Jian_Wang_Programmable_Light_Curtains_ECCV_2018_paper.html |
| http://openaccess.thecvf.com/content_ECCV_2018/papers/Jian_Wang_Programmable_Light_Curtains_ECCV_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/programmable-triangulation-light-curtains |
| Repo | |
| Framework | |
Ensemble of Translators with Automatic Selection of the Best Translation – the submission of FOKUS to the WMT 18 biomedical translation task –
| Title | Ensemble of Translators with Automatic Selection of the Best Translation – the submission of FOKUS to the WMT 18 biomedical translation task – |
| Authors | Cristian Grozea |
| Abstract | This paper describes the system of Fraunhofer FOKUS for the WMT 2018 biomedical translation task. Our approach, described here, was to automatically select the most promising translation from a set of candidates produced with NMT (Transformer) models. We selected the highest fidelity translation of each sentence by using a dictionary, stemming and a set of heuristics. Our method is simple, can use any machine translators, and requires no further training in addition to that already employed to build the NMT models. The downside is that the score did not increase over the best in ensemble, but was quite close to it (difference about 0.5 BLEU). |
| Tasks | Machine Translation |
| Published | 2018-10-01 |
| URL | https://www.aclweb.org/anthology/W18-6445/ |
| https://www.aclweb.org/anthology/W18-6445 | |
| PWC | https://paperswithcode.com/paper/ensemble-of-translators-with-automatic |
| Repo | |
| Framework | |
Baseline-corrected space-by-time non-negative matrix factorization for decoding single trial population spike trains
| Title | Baseline-corrected space-by-time non-negative matrix factorization for decoding single trial population spike trains |
| Authors | Arezoo Alizadeh, Marion Mutter, Thomas Münch, Arno Onken, Stefano Panzeri |
| Abstract | Activity of populations of sensory neurons carries stimulus information in both the temporal and the spatial dimensions. This poses the question of how to compactly represent all the information that the population codes carry across all these dimensions. Here, we developed an analytical method to factorize a large number of retinal ganglion cells’ spike trains into a robust low-dimensional representation that captures efficiently both their spatial and temporal information. In particular, we extended previously used single-trial space-by-time tensor decomposition based on non-negative matrix factorization to efficiently discount pre-stimulus baseline activity. On data recorded from retinal ganglion cells with strong pre-stimulus baseline, we showed that in situations were the stimulus elicits a strong change in firing rate, our extensions yield a boost in stimulus decoding performance. Our results thus suggest that taking into account the baseline can be important for finding a compact information-rich representation of neural activity. |
| Tasks | |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=Bki1Ct1AW |
| https://openreview.net/pdf?id=Bki1Ct1AW | |
| PWC | https://paperswithcode.com/paper/baseline-corrected-space-by-time-non-negative |
| Repo | |
| Framework | |
Language Modeling for Morphologically Rich Languages: Character-Aware Modeling for Word-Level Prediction
| Title | Language Modeling for Morphologically Rich Languages: Character-Aware Modeling for Word-Level Prediction |
| Authors | Daniela Gerz, Ivan Vuli{'c}, Edoardo Ponti, Jason Naradowsky, Roi Reichart, Anna Korhonen |
| Abstract | Neural architectures are prominent in the construction of language models (LMs). However, word-level prediction is typically agnostic of subword-level information (characters and character sequences) and operates over a closed vocabulary, consisting of a limited word set. Indeed, while subword-aware models boost performance across a variety of NLP tasks, previous work did not evaluate the ability of these models to assist next-word prediction in language modeling tasks. Such subword-level informed models should be particularly effective for morphologically-rich languages (MRLs) that exhibit high type-to-token ratios. In this work, we present a large-scale LM study on 50 typologically diverse languages covering a wide variety of morphological systems, and offer new LM benchmarks to the community, while considering subword-level information. The main technical contribution of our work is a novel method for injecting subword-level information into semantic word vectors, integrated into the neural language modeling training, to facilitate word-level prediction. We conduct experiments in the LM setting where the number of infrequent words is large, and demonstrate strong perplexity gains across our 50 languages, especially for morphologically-rich languages. Our code and data sets are publicly available. |
| Tasks | Dialogue Generation, Language Modelling, Speech Recognition, Text Generation |
| Published | 2018-01-01 |
| URL | https://www.aclweb.org/anthology/Q18-1032/ |
| https://www.aclweb.org/anthology/Q18-1032 | |
| PWC | https://paperswithcode.com/paper/language-modeling-for-morphologically-rich |
| Repo | |
| Framework | |
Accurate and Diverse Sampling of Sequences Based on a âBest of Manyâ Sample Objective
| Title | Accurate and Diverse Sampling of Sequences Based on a âBest of Manyâ Sample Objective |
| Authors | Apratim Bhattacharyya, Bernt Schiele, Mario Fritz |
| Abstract | For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence. This problem has been formalized as a sequence extrapolation problem, where a number of observations are used to predict the sequence into the future. Real-world scenarios demand a model of uncertainty of such predictions, as predictions become increasingly uncertain – in particular on long time horizons. While impressive results have been shown on point estimates, scenarios that induce multi-modal distributions over future sequences remain challenging. Our work addresses these challenges in a Gaussian Latent Variable model for sequence prediction. Our core contribution is a ``Best of Many’’ sample objective that leads to more accurate and more diverse predictions that better capture the true variations in real-world sequence data. Beyond our analysis of improved model fit, our models also empirically outperform prior work on three diverse tasks ranging from traffic scenes to weather data. | |
| Tasks | |
| Published | 2018-06-01 |
| URL | http://openaccess.thecvf.com/content_cvpr_2018/html/Bhattacharyya_Accurate_and_Diverse_CVPR_2018_paper.html |
| http://openaccess.thecvf.com/content_cvpr_2018/papers/Bhattacharyya_Accurate_and_Diverse_CVPR_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/accurate-and-diverse-sampling-of-sequences-1 |
| Repo | |
| Framework | |
Jiffy: A Convolutional Approach to Learning Time Series Similarity
| Title | Jiffy: A Convolutional Approach to Learning Time Series Similarity |
| Authors | Divya Shanmugam, Davis Blalock, John Guttag |
| Abstract | Computing distances between examples is at the core of many learning algorithms for time series. Consequently, a great deal of work has gone into designing effective time series distance measures. We present Jiffy, a simple and scalable distance metric for multivariate time series. Our approach is to reframe the task as a representation learning problem—rather than design an elaborate distance function, we use a CNN to learn an embedding such that the Euclidean distance is effective. By aggressively max-pooling and downsampling, we are able to construct this embedding using a highly compact neural network. Experiments on a diverse set of multivariate time series datasets show that our approach consistently outperforms existing methods. |
| Tasks | Representation Learning, Time Series |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=ryacTMZRZ |
| https://openreview.net/pdf?id=ryacTMZRZ | |
| PWC | https://paperswithcode.com/paper/jiffy-a-convolutional-approach-to-learning |
| Repo | |
| Framework | |
Rollenwechsel-English: a large-scale semantic role corpus
| Title | Rollenwechsel-English: a large-scale semantic role corpus |
| Authors | Asad Sayeed, Pavel Shkadzko, Vera Demberg |
| Abstract | |
| Tasks | Language Modelling |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1488/ |
| https://www.aclweb.org/anthology/L18-1488 | |
| PWC | https://paperswithcode.com/paper/rollenwechsel-english-a-large-scale-semantic |
| Repo | |
| Framework | |
Evidence Types, Credibility Factors, and Patterns or Soft Rules for Weighing Conflicting Evidence: Argument Mining in the Context of Legal Rules Governing Evidence Assessment
| Title | Evidence Types, Credibility Factors, and Patterns or Soft Rules for Weighing Conflicting Evidence: Argument Mining in the Context of Legal Rules Governing Evidence Assessment |
| Authors | Vern R. Walker, Dina Foerster, Julia Monica Ponce, Matthew Rosen |
| Abstract | This paper reports on the results of an empirical study of adjudicatory decisions about veterans{'} claims for disability benefits in the United States. It develops a typology of kinds of relevant evidence (argument premises) employed in cases, and it identifies factors that the tribunal considers when assessing the credibility or trustworthiness of individual items of evidence. It also reports on patterns or {``}soft rules{''} that the tribunal uses to comparatively weigh the probative value of conflicting evidence. These evidence types, credibility factors, and comparison patterns are developed to be inter-operable with legal rules governing the evidence assessment process in the U.S. This approach should be transferable to other legal and non-legal domains. | |
| Tasks | Argument Mining, Decision Making |
| Published | 2018-11-01 |
| URL | https://www.aclweb.org/anthology/W18-5209/ |
| https://www.aclweb.org/anthology/W18-5209 | |
| PWC | https://paperswithcode.com/paper/evidence-types-credibility-factors-and |
| Repo | |
| Framework | |
Genetic-Gated Networks for Deep Reinforcement Learning
| Title | Genetic-Gated Networks for Deep Reinforcement Learning |
| Authors | Simyung Chang, John Yang, Jaeseok Choi, Nojun Kwak |
| Abstract | We introduce the Genetic-Gated Networks (G2Ns), simple neural networks that combine a gate vector composed of binary genetic genes in the hidden layer(s) of networks. Our method can take both advantages of gradient-free optimization and gradient-based optimization methods, of which the former is effective for problems with multiple local minima, while the latter can quickly find local minima. In addition, multiple chromosomes can define different models, making it easy to construct multiple models and can be effectively applied to problems that require multiple models. We show that this G2N can be applied to typical reinforcement learning algorithms to achieve a large improvement in sample efficiency and performance. |
| Tasks | |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/7446-genetic-gated-networks-for-deep-reinforcement-learning |
| http://papers.nips.cc/paper/7446-genetic-gated-networks-for-deep-reinforcement-learning.pdf | |
| PWC | https://paperswithcode.com/paper/genetic-gated-networks-for-deep-reinforcement-1 |
| Repo | |
| Framework | |
Automatic Detection of Incoherent Speech for Diagnosing Schizophrenia
| Title | Automatic Detection of Incoherent Speech for Diagnosing Schizophrenia |
| Authors | Dan Iter, Jong Yoon, Dan Jurafsky |
| Abstract | Schizophrenia is a mental disorder which afflicts an estimated 0.7{%} of adults world wide. It affects many areas of mental function, often evident from incoherent speech. Diagnosing schizophrenia relies on subjective judgments resulting in disagreements even among trained clinicians. Recent studies have proposed the use of natural language processing for diagnosis by drawing on automatically-extracted linguistic features like discourse coherence and lexicon. Here, we present the first benchmark comparison of previously proposed coherence models for detecting symptoms of schizophrenia and evaluate their performance on a new dataset of recorded interviews between subjects and clinicians. We also present two alternative coherence metrics based on modern sentence embedding techniques that outperform the previous methods on our dataset. Lastly, we propose a novel computational model for reference incoherence based on ambiguous pronoun usage and show that it is a highly predictive feature on our data. While the number of subjects is limited in this pilot study, our results suggest new directions for diagnosing common symptoms of schizophrenia. |
| Tasks | Sentence Embedding, Word Embeddings |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/W18-0615/ |
| https://www.aclweb.org/anthology/W18-0615 | |
| PWC | https://paperswithcode.com/paper/automatic-detection-of-incoherent-speech-for |
| Repo | |
| Framework | |
