Paper Group ANR 982
Interpretable Deep Learning for Two-Prong Jet Classification with Jet Spectra. The Stroke Correspondence Problem, Revisited. Scalable Multi Corpora Neural Language Models for ASR. Multi-View Region Adaptive Multi-temporal DMM and RGB Action Recognition. Differentially private anonymized histograms. Off-policy Learning for Multiple Loggers. One Mode …
Interpretable Deep Learning for Two-Prong Jet Classification with Jet Spectra
| Title | Interpretable Deep Learning for Two-Prong Jet Classification with Jet Spectra |
| Authors | Amit Chakraborty, Sung Hak Lim, Mihoko M. Nojiri |
| Abstract | Classification of jets with deep learning has gained significant attention in recent times. However, the performance of deep neural networks is often achieved at the cost of interpretability. Here we propose an interpretable network trained on the jet spectrum $S_{2}(R)$ which is a two-point correlation function of the jet constituents. The spectrum can be derived from a functional Taylor series of an arbitrary jet classifier function of energy flows. An interpretable network can be obtained by truncating the series. The intermediate feature of the network is an infrared and collinear safe C-correlator which allows us to estimate the importance of a $S_{2}(R)$ deposit at an angular scale R in the classification. The performance of the architecture is comparable to that of a convolutional neural network (CNN) trained on jet images, although the number of inputs and complexity of architecture is significantly simpler than the CNN classifier. We consider two examples: one is the classification of two-prong jets which differ in color charge of the mother particle, and the other is a comparison between Pythia 8 and Herwig 7 generated jets. |
| Tasks | |
| Published | 2019-04-03 |
| URL | https://arxiv.org/abs/1904.02092v2 |
| https://arxiv.org/pdf/1904.02092v2.pdf | |
| PWC | https://paperswithcode.com/paper/interpretable-deep-learning-for-two-prong-jet |
| Repo | |
| Framework | |
The Stroke Correspondence Problem, Revisited
| Title | The Stroke Correspondence Problem, Revisited |
| Authors | Dominik Klein |
| Abstract | We revisit the stroke correspondence problem [13,14]. We optimize this algorithm by 1) evaluating suitable preprocessing (normalization) methods 2) extending the algorithm with an additional distance measure to handle Hiragana, Katakana and Kanji characters with a low number of strokes and c) simplify the stroke linking algorithms. Our contributions are implemented in the free, open-source library ctegaki and in the demo-tools jTegaki and Kanjicanvas. |
| Tasks | |
| Published | 2019-09-26 |
| URL | https://arxiv.org/abs/1909.11995v2 |
| https://arxiv.org/pdf/1909.11995v2.pdf | |
| PWC | https://paperswithcode.com/paper/the-stroke-correspondence-problem-revisited |
| Repo | |
| Framework | |
Scalable Multi Corpora Neural Language Models for ASR
| Title | Scalable Multi Corpora Neural Language Models for ASR |
| Authors | Anirudh Raju, Denis Filimonov, Gautam Tiwari, Guitang Lan, Ariya Rastrow |
| Abstract | Neural language models (NLM) have been shown to outperform conventional n-gram language models by a substantial margin in Automatic Speech Recognition (ASR) and other tasks. There are, however, a number of challenges that need to be addressed for an NLM to be used in a practical large-scale ASR system. In this paper, we present solutions to some of the challenges, including training NLM from heterogenous corpora, limiting latency impact and handling personalized bias in the second-pass rescorer. Overall, we show that we can achieve a 6.2% relative WER reduction using neural LM in a second-pass n-best rescoring framework with a minimal increase in latency. |
| Tasks | Speech Recognition |
| Published | 2019-07-02 |
| URL | https://arxiv.org/abs/1907.01677v1 |
| https://arxiv.org/pdf/1907.01677v1.pdf | |
| PWC | https://paperswithcode.com/paper/scalable-multi-corpora-neural-language-models |
| Repo | |
| Framework | |
Multi-View Region Adaptive Multi-temporal DMM and RGB Action Recognition
| Title | Multi-View Region Adaptive Multi-temporal DMM and RGB Action Recognition |
| Authors | Mahmoud Al-Faris, John P. Chiverton, Yanyan Yang, David L. Ndzi |
| Abstract | Human action recognition remains an important yet challenging task. This work proposes a novel action recognition system. It uses a novel Multiple View Region Adaptive Multi-resolution in time Depth Motion Map (MV-RAMDMM) formulation combined with appearance information. Multiple stream 3D Convolutional Neural Networks (CNNs) are trained on the different views and time resolutions of the region adaptive Depth Motion Maps. Multiple views are synthesised to enhance the view invariance. The region adaptive weights, based on localised motion, accentuate and differentiate parts of actions possessing faster motion. Dedicated 3D CNN streams for multi-time resolution appearance information (RGB) are also included. These help to identify and differentiate between small object interactions. A pre-trained 3D-CNN is used here with fine-tuning for each stream along with multiple class Support Vector Machines (SVM)s. Average score fusion is used on the output. The developed approach is capable of recognising both human action and human-object interaction. Three public domain datasets including: MSR 3D Action,Northwestern UCLA multi-view actions and MSR 3D daily activity are used to evaluate the proposed solution. The experimental results demonstrate the robustness of this approach compared with state-of-the-art algorithms. |
| Tasks | Human-Object Interaction Detection, Temporal Action Localization |
| Published | 2019-04-12 |
| URL | http://arxiv.org/abs/1904.06074v1 |
| http://arxiv.org/pdf/1904.06074v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-view-region-adaptive-multi-temporal-dmm |
| Repo | |
| Framework | |
Differentially private anonymized histograms
| Title | Differentially private anonymized histograms |
| Authors | Ananda Theertha Suresh |
| Abstract | For a dataset of label-count pairs, an anonymized histogram is the multiset of counts. Anonymized histograms appear in various potentially sensitive contexts such as password-frequency lists, degree distribution in social networks, and estimation of symmetric properties of discrete distributions. Motivated by these applications, we propose the first differentially private mechanism to release anonymized histograms that achieves near-optimal privacy utility trade-off both in terms of number of items and the privacy parameter. Further, if the underlying histogram is given in a compact format, the proposed algorithm runs in time sub-linear in the number of items. For anonymized histograms generated from unknown discrete distributions, we show that the released histogram can be directly used for estimating symmetric properties of the underlying distribution. |
| Tasks | |
| Published | 2019-10-08 |
| URL | https://arxiv.org/abs/1910.03553v2 |
| https://arxiv.org/pdf/1910.03553v2.pdf | |
| PWC | https://paperswithcode.com/paper/differentially-private-anonymized-histograms |
| Repo | |
| Framework | |
Off-policy Learning for Multiple Loggers
| Title | Off-policy Learning for Multiple Loggers |
| Authors | Li He, Long Xia, Wei Zeng, Zhi-Ming Ma, Yihong Zhao, Dawei Yin |
| Abstract | It is well known that the historical logs are used for evaluating and learning policies in interactive systems, e.g. recommendation, search, and online advertising. Since direct online policy learning usually harms user experiences, it is more crucial to apply off-policy learning in real-world applications instead. Though there have been some existing works, most are focusing on learning with one single historical policy. However, in practice, usually a number of parallel experiments, e.g. multiple AB tests, are performed simultaneously. To make full use of such historical data, learning policies from multiple loggers becomes necessary. Motivated by this, in this paper, we investigate off-policy learning when the training data coming from multiple historical policies. Specifically, policies, e.g. neural networks, can be learned directly from multi-logger data, with counterfactual estimators. In order to understand the generalization ability of such estimator better, we conduct generalization error analysis for the empirical risk minimization problem. We then introduce the generalization error bound as the new risk function, which can be reduced to a constrained optimization problem. Finally, we give the corresponding learning algorithm for the new constrained problem, where we can appeal to the minimax problems to control the constraints. Extensive experiments on benchmark datasets demonstrate that the proposed methods achieve better performances than the state-of-the-arts. |
| Tasks | |
| Published | 2019-07-23 |
| URL | https://arxiv.org/abs/1907.09652v2 |
| https://arxiv.org/pdf/1907.09652v2.pdf | |
| PWC | https://paperswithcode.com/paper/off-policy-learning-for-multiple-loggers |
| Repo | |
| Framework | |
One Model to Learn Both: Zero Pronoun Prediction and Translation
| Title | One Model to Learn Both: Zero Pronoun Prediction and Translation |
| Authors | Longyue Wang, Zhaopeng Tu, Xing Wang, Shuming Shi |
| Abstract | Zero pronouns (ZPs) are frequently omitted in pro-drop languages, but should be recalled in non-pro-drop languages. This discourse phenomenon poses a significant challenge for machine translation (MT) when translating texts from pro-drop to non-pro-drop languages. In this paper, we propose a unified and discourse-aware ZP translation approach for neural MT models. Specifically, we jointly learn to predict and translate ZPs in an end-to-end manner, allowing both components to interact with each other. In addition, we employ hierarchical neural networks to exploit discourse-level context, which is beneficial for ZP prediction and thus translation. Experimental results on both Chinese-English and Japanese-English data show that our approach significantly and accumulatively improves both translation performance and ZP prediction accuracy over not only baseline but also previous works using external ZP prediction models. Extensive analyses confirm that the performance improvement comes from the alleviation of different kinds of errors especially caused by subjective ZPs. |
| Tasks | Machine Translation |
| Published | 2019-09-01 |
| URL | https://arxiv.org/abs/1909.00369v1 |
| https://arxiv.org/pdf/1909.00369v1.pdf | |
| PWC | https://paperswithcode.com/paper/one-model-to-learn-both-zero-pronoun |
| Repo | |
| Framework | |
Open Research Knowledge Graph: Next Generation Infrastructure for Semantic Scholarly Knowledge
| Title | Open Research Knowledge Graph: Next Generation Infrastructure for Semantic Scholarly Knowledge |
| Authors | Mohamad Yaser Jaradeh, Allard Oelen, Kheir Eddine Farfar, Manuel Prinz, Jennifer D’Souza, Gábor Kismihók, Markus Stocker, Sören Auer |
| Abstract | Despite improved digital access to scholarly knowledge in recent decades, scholarly communication remains exclusively document-based. In this form, scholarly knowledge is hard to process automatically. In this paper, we present the first steps towards a knowledge graph based infrastructure that acquires scholarly knowledge in machine actionable form thus enabling new possibilities for scholarly knowledge curation, publication and processing. The primary contribution is to present, evaluate and discuss multi-modal scholarly knowledge acquisition, combining crowdsourced and automated techniques. We present the results of the first user evaluation of the infrastructure with the participants of a recent international conference. Results suggest that users were intrigued by the novelty of the proposed infrastructure and by the possibilities for innovative scholarly knowledge processing it could enable. |
| Tasks | Knowledge Graphs |
| Published | 2019-01-30 |
| URL | https://arxiv.org/abs/1901.10816v3 |
| https://arxiv.org/pdf/1901.10816v3.pdf | |
| PWC | https://paperswithcode.com/paper/open-research-knowledge-graph-towards-machine |
| Repo | |
| Framework | |
Minimal Solvers for Rectifying from Radially-Distorted Conjugate Translations
| Title | Minimal Solvers for Rectifying from Radially-Distorted Conjugate Translations |
| Authors | James Pritts, Zuzana Kukelova, Viktor Larsson, Yaroslava Lochman, Ondřej Chum |
| Abstract | This paper introduces minimal solvers that jointly solve for affine-rectification and radial lens undistortion from the image of translated and reflected coplanar features. The proposed solvers use the invariant that the affine-rectified image of the meet of the joins of radially-distorted conjugately-translated point correspondences is on the line at infinity. The hidden-variable trick from algebraic geometry is used to reformulate and simplify the constraints so that the generated solvers are stable, small and fast. Multiple solvers are proposed to accommodate various local feature types and sampling strategies, and, remarkably, three of the proposed solvers can recover rectification and lens undistortion from only one radially-distorted conjugately-translated affine-covariant region correspondence. Synthetic and real-image experiments confirm that the proposed solvers demonstrate superior robustness to noise compared to the state of the art. Accurate rectifications on imagery taken with narrow to fisheye field-of-view lenses demonstrate the wide applicability of the proposed method. The method is fully automatic. |
| Tasks | |
| Published | 2019-11-04 |
| URL | https://arxiv.org/abs/1911.01507v1 |
| https://arxiv.org/pdf/1911.01507v1.pdf | |
| PWC | https://paperswithcode.com/paper/minimal-solvers-for-rectifying-from-radially-1 |
| Repo | |
| Framework | |
Controlling the Output Length of Neural Machine Translation
| Title | Controlling the Output Length of Neural Machine Translation |
| Authors | Surafel Melaku Lakew, Mattia Di Gangi, Marcello Federico |
| Abstract | The recent advances introduced by neural machine translation (NMT) are rapidly expanding the application fields of machine translation, as well as reshaping the quality level to be targeted. In particular, if translations have to fit some given layout, quality should not only be measured in terms of adequacy and fluency, but also length. Exemplary cases are the translation of document files, subtitles, and scripts for dubbing, where the output length should ideally be as close as possible to the length of the input text. This paper addresses for the first time, to the best of our knowledge, the problem of controlling the output length in NMT. We investigate two methods for biasing the output length with a transformer architecture: i) conditioning the output to a given target-source length-ratio class and ii) enriching the transformer positional embedding with length information. Our experiments show that both methods can induce the network to generate shorter translations, as well as acquiring interpretable linguistic skills. |
| Tasks | Machine Translation |
| Published | 2019-10-23 |
| URL | https://arxiv.org/abs/1910.10408v2 |
| https://arxiv.org/pdf/1910.10408v2.pdf | |
| PWC | https://paperswithcode.com/paper/controlling-the-output-length-of-neural |
| Repo | |
| Framework | |
Deep Neural Networks Ensemble for Detecting Medication Mentions in Tweets
| Title | Deep Neural Networks Ensemble for Detecting Medication Mentions in Tweets |
| Authors | Davy Weissenbacher, Abeed Sarker, Ari Klein, Karen O’Connor, Arjun Magge Ranganatha, Graciela Gonzalez-Hernandez |
| Abstract | Objective: After years of research, Twitter posts are now recognized as an important source of patient-generated data, providing unique insights into population health. A fundamental step to incorporating Twitter data in pharmacoepidemiological research is to automatically recognize medication mentions in tweets. Given that lexical searches for medication names may fail due to misspellings or ambiguity with common words, we propose a more advanced method to recognize them. Methods: We present Kusuri, an Ensemble Learning classifier, able to identify tweets mentioning drug products and dietary supplements. Kusuri (“medication” in Japanese) is composed of two modules. First, four different classifiers (lexicon-based, spelling-variant-based, pattern-based and one based on a weakly-trained neural network) are applied in parallel to discover tweets potentially containing medication names. Second, an ensemble of deep neural networks encoding morphological, semantical and long-range dependencies of important words in the tweets discovered is used to make the final decision. Results: On a balanced (50-50) corpus of 15,005 tweets, Kusuri demonstrated performances close to human annotators with 93.7% F1-score, the best score achieved thus far on this corpus. On a corpus made of all tweets posted by 113 Twitter users (98,959 tweets, with only 0.26% mentioning medications), Kusuri obtained 76.3% F1-score. There is not a prior drug extraction system that compares running on such an extremely unbalanced dataset. Conclusion: The system identifies tweets mentioning drug names with performance high enough to ensure its usefulness and ready to be integrated in larger natural language processing systems. |
| Tasks | |
| Published | 2019-04-10 |
| URL | https://arxiv.org/abs/1904.05308v2 |
| https://arxiv.org/pdf/1904.05308v2.pdf | |
| PWC | https://paperswithcode.com/paper/deep-neural-networks-ensemble-for-detecting |
| Repo | |
| Framework | |
DRASIC: Distributed Recurrent Autoencoder for Scalable Image Compression
| Title | DRASIC: Distributed Recurrent Autoencoder for Scalable Image Compression |
| Authors | Enmao Diao, Jie Ding, Vahid Tarokh |
| Abstract | We propose a new architecture for distributed image compression from a group of distributed data sources. The work is motivated by practical needs of data-driven codec design, low power consumption, robustness, and data privacy. The proposed architecture, which we refer to as Distributed Recurrent Autoencoder for Scalable Image Compression (DRASIC), is able to train distributed encoders and one joint decoder on correlated data sources. Its compression capability is much better than the method of training codecs separately. Meanwhile, the performance of our distributed system with 10 distributed sources is only within 2 dB peak signal-to-noise ratio (PSNR) of the performance of a single codec trained with all data sources. We experiment distributed sources with different correlations and show how our data-driven methodology well matches the Slepian-Wolf Theorem in Distributed Source Coding (DSC). To the best of our knowledge, this is the first data-driven DSC framework for general distributed code design with deep learning. |
| Tasks | Image Compression |
| Published | 2019-03-23 |
| URL | https://arxiv.org/abs/1903.09887v3 |
| https://arxiv.org/pdf/1903.09887v3.pdf | |
| PWC | https://paperswithcode.com/paper/distributed-lossy-image-compression-with |
| Repo | |
| Framework | |
You Shall Know a User by the Company It Keeps: Dynamic Representations for Social Media Users in NLP
| Title | You Shall Know a User by the Company It Keeps: Dynamic Representations for Social Media Users in NLP |
| Authors | Marco Del Tredici, Diego Marcheggiani, Sabine Schulte im Walde, Raquel Fernández |
| Abstract | Information about individuals can help to better understand what they say, particularly in social media where texts are short. Current approaches to modelling social media users pay attention to their social connections, but exploit this information in a static way, treating all connections uniformly. This ignores the fact, well known in sociolinguistics, that an individual may be part of several communities which are not equally relevant in all communicative situations. We present a model based on Graph Attention Networks that captures this observation. It dynamically explores the social graph of a user, computes a user representation given the most relevant connections for a target task, and combines it with linguistic information to make a prediction. We apply our model to three different tasks, evaluate it against alternative models, and analyse the results extensively, showing that it significantly outperforms other current methods. |
| Tasks | |
| Published | 2019-09-01 |
| URL | https://arxiv.org/abs/1909.00412v1 |
| https://arxiv.org/pdf/1909.00412v1.pdf | |
| PWC | https://paperswithcode.com/paper/you-shall-know-a-user-by-the-company-it-keeps |
| Repo | |
| Framework | |
Imbalanced multi-label classification using multi-task learning with extractive summarization
| Title | Imbalanced multi-label classification using multi-task learning with extractive summarization |
| Authors | John Brandt |
| Abstract | Extractive summarization and imbalanced multi-label classification often require vast amounts of training data to avoid overfitting. In situations where training data is expensive to generate, leveraging information between tasks is an attractive approach to increasing the amount of available information. This paper employs multi-task training of an extractive summarizer and an RNN-based classifier to improve summarization and classification accuracy by 50% and 75%, respectively, relative to RNN baselines. We hypothesize that concatenating sentence encodings based on document and class context increases generalizability for highly variable corpuses. |
| Tasks | Multi-Label Classification, Multi-Task Learning |
| Published | 2019-03-16 |
| URL | http://arxiv.org/abs/1903.06963v1 |
| http://arxiv.org/pdf/1903.06963v1.pdf | |
| PWC | https://paperswithcode.com/paper/imbalanced-multi-label-classification-using |
| Repo | |
| Framework | |
Extending Machine Language Models toward Human-Level Language Understanding
| Title | Extending Machine Language Models toward Human-Level Language Understanding |
| Authors | James L. McClelland, Felix Hill, Maja Rudolph, Jason Baldridge, Hinrich Schütze |
| Abstract | Language is central to human intelligence. We review recent breakthroughs in machine language processing and consider what remains to be achieved. Recent approaches rely on domain general principles of learning and representation captured in artificial neural networks. Most current models, however, focus too closely on language itself. In humans, language is part of a larger system for acquiring, representing, and communicating about objects and situations in the physical and social world, and future machine language models should emulate such a system. We describe existing machine models linking language to concrete situations, and point toward extensions to address more abstract cases. Human language processing exploits complementary learning systems, including a deep neural network-like learning system that learns gradually as machine systems do, as well as a fast-learning system that supports learning new information quickly. Adding such a system to machine language models will be an important further step toward truly human-like language understanding. |
| Tasks | |
| Published | 2019-12-12 |
| URL | https://arxiv.org/abs/1912.05877v1 |
| https://arxiv.org/pdf/1912.05877v1.pdf | |
| PWC | https://paperswithcode.com/paper/extending-machine-language-models-toward |
| Repo | |
| Framework | |
