Paper Group NAWR 7
Recurrent Attention Network on Memory for Aspect Sentiment Analysis. C-3MA: Tartu-Riga-Zurich Translation Systems for WMT17. Automatic Differentiation in PyTorch. Deep Recurrent Neural Network-Based Identification of Precursor microRNAs. Review highlights: opinion mining on reviews: a hybrid model for rule selection in aspect extraction. Context-Aw …
Recurrent Attention Network on Memory for Aspect Sentiment Analysis
| Title | Recurrent Attention Network on Memory for Aspect Sentiment Analysis |
| Authors | Peng Chen, Zhongqian Sun, Lidong Bing, Wei Yang |
| Abstract | We propose a novel framework based on neural networks to identify the sentiment of opinion targets in a comment/review. Our framework adopts multiple-attention mechanism to capture sentiment features separated by a long distance, so that it is more robust against irrelevant information. The results of multiple attentions are non-linearly combined with a recurrent neural network, which strengthens the expressive power of our model for handling more complications. The weighted-memory mechanism not only helps us avoid the labor-intensive feature engineering work, but also provides a tailor-made memory for different opinion targets of a sentence. We examine the merit of our model on four datasets: two are from SemEval2014, i.e. reviews of restaurants and laptops; a twitter dataset, for testing its performance on social media data; and a Chinese news comment dataset, for testing its language sensitivity. The experimental results show that our model consistently outperforms the state-of-the-art methods on different types of data. |
| Tasks | Aspect-Based Sentiment Analysis, Feature Engineering, Sentiment Analysis |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/D17-1047/ |
| https://www.aclweb.org/anthology/D17-1047 | |
| PWC | https://paperswithcode.com/paper/recurrent-attention-network-on-memory-for |
| Repo | https://github.com/lpq29743/RAM |
| Framework | tf |
C-3MA: Tartu-Riga-Zurich Translation Systems for WMT17
| Title | C-3MA: Tartu-Riga-Zurich Translation Systems for WMT17 |
| Authors | Mat{=\i}ss Rikters, Chantal Amrhein, Maksym Del, Mark Fishel |
| Abstract | |
| Tasks | Machine Translation, Word Embeddings |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/W17-4738/ |
| https://www.aclweb.org/anthology/W17-4738 | |
| PWC | https://paperswithcode.com/paper/c-3ma-tartu-riga-zurich-translation-systems |
| Repo | https://github.com/M4t1ss/C-3MA |
| Framework | none |
Automatic Differentiation in PyTorch
| Title | Automatic Differentiation in PyTorch |
| Authors | Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Alban Desmaison, Luca Antiga, Adam Lerer |
| Abstract | In this article, we describe an automatic differentiation module of PyTorch — a library designed to enable rapid research on machine learning models. It builds upon a few projects, most notably Lua Torch, Chainer, and HIPS Autograd, and provides a high performance environment with easy access to automatic differentiation of models executed on different devices (CPU and GPU). To make prototyping easier, PyTorch does not follow the symbolic approach used in many other deep learning frameworks, but focuses on differentiation of purely imperative programs, with a focus on extensibility and low overhead. Note that this preprint is a draft of certain sections from an upcoming paper covering all PyTorch features. |
| Tasks | Dimensionality Reduction |
| Published | 2017-10-28 |
| URL | https://openreview.net/forum?id=BJJsrmfCZ |
| https://openreview.net/pdf?id=BJJsrmfCZ | |
| PWC | https://paperswithcode.com/paper/automatic-differentiation-in-pytorch |
| Repo | https://github.com/pytorch/pytorch |
| Framework | pytorch |
Deep Recurrent Neural Network-Based Identification of Precursor microRNAs
| Title | Deep Recurrent Neural Network-Based Identification of Precursor microRNAs |
| Authors | Seunghyun Park, Seonwoo Min, Hyun-Soo Choi, Sungroh Yoon |
| Abstract | MicroRNAs (miRNAs) are small non-coding ribonucleic acids (RNAs) which play key roles in post-transcriptional gene regulation. Direct identification of mature miRNAs is infeasible due to their short lengths, and researchers instead aim at identifying precursor miRNAs (pre-miRNAs). Many of the known pre-miRNAs have distinctive stem-loop secondary structure, and structure-based filtering is usually the first step to predict the possibility of a given sequence being a pre-miRNA. To identify new pre-miRNAs that often have non-canonical structure, however, we need to consider additional features other than structure. To obtain such additional characteristics, existing computational methods rely on manual feature extraction, which inevitably limits the efficiency, robustness, and generalization of computational identification. To address the limitations of existing approaches, we propose a pre-miRNA identification method that incorporates (1) a deep recurrent neural network (RNN) for automated feature learning and classification, (2) multimodal architecture for seamless integration of prior knowledge (secondary structure), (3) an attention mechanism for improving long-term dependence modeling, and (4) an RNN-based class activation mapping for highlighting the learned representations that can contrast pre-miRNAs and non-pre-miRNAs. In our experiments with recent benchmarks, the proposed approach outperformed the compared state-of-the-art alternatives in terms of various performance metrics. |
| Tasks | |
| Published | 2017-12-01 |
| URL | http://papers.nips.cc/paper/6882-deep-recurrent-neural-network-based-identification-of-precursor-micrornas |
| http://papers.nips.cc/paper/6882-deep-recurrent-neural-network-based-identification-of-precursor-micrornas.pdf | |
| PWC | https://paperswithcode.com/paper/deep-recurrent-neural-network-based |
| Repo | https://github.com/eleventh83/deepMiRGene |
| Framework | none |
Review highlights: opinion mining on reviews: a hybrid model for rule selection in aspect extraction
| Title | Review highlights: opinion mining on reviews: a hybrid model for rule selection in aspect extraction |
| Authors | Amit Kushwaha, Shubham Chaudhary |
| Abstract | This paper proposes a methodology to extract key insights from user generated reviews. This work is based on Aspect Based Sentiment Analysis (ABSA) which predicts the sentiment of aspects mentioned in the text documents. The extracted aspects are fine-grained for the presentation form known as Review Highlights. The syntactic approach for extraction process suffers from the overlapping chunking rules which result in noise extraction. We introduce a hybrid technique which combines machine learning and rule based model. A multi-label classifier identifies the effective rules which efficiently parse aspects and opinions from texts. This selection of rules reduce the amount of noise in extraction tasks. This is a novel attempt to learn syntactic rule fitness from a corpus using machine learning for accurate aspect extraction. As the model learns the syntactic rule prediction from the corpus, it makes the extraction method domain independent. It also allows studying the quality of syntactic rules in a different corpus. |
| Tasks | Aspect-Based Sentiment Analysis, Aspect Extraction, Chunking, Extract Aspect, Extract aspect-polarity tuple, Opinion Mining, Sentiment Analysis |
| Published | 2017-10-18 |
| URL | https://dl.acm.org/citation.cfm?id=3158385 |
| http://vixra.org/pdf/1910.0514v1.pdf | |
| PWC | https://paperswithcode.com/paper/review-highlights-opinion-mining-on-reviews-a |
| Repo | https://github.com/yardstick17/AspectBasedSentimentAnalysis |
| Framework | none |
Context-Aware Representations for Knowledge Base Relation Extraction
| Title | Context-Aware Representations for Knowledge Base Relation Extraction |
| Authors | Daniil Sorokin, Iryna Gurevych |
| Abstract | We demonstrate that for sentence-level relation extraction it is beneficial to consider other relations in the sentential context while predicting the target relation. Our architecture uses an LSTM-based encoder to jointly learn representations for all relations in a single sentence. We combine the context representations with an attention mechanism to make the final prediction. We use the Wikidata knowledge base to construct a dataset of multiple relations per sentence and to evaluate our approach. Compared to a baseline system, our method results in an average error reduction of 24 on a held-out set of relations. The code and the dataset to replicate the experiments are made available at \url{https://github.com/ukplab/}. |
| Tasks | Question Answering, Relation Extraction |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/D17-1188/ |
| https://www.aclweb.org/anthology/D17-1188 | |
| PWC | https://paperswithcode.com/paper/context-aware-representations-for-knowledge |
| Repo | https://github.com/UKPLab/emnlp2017-relation-extraction |
| Framework | tf |
Convolutional Neural Networks (CNNs) for power system big data analysis
| Title | Convolutional Neural Networks (CNNs) for power system big data analysis |
| Authors | Siby Jose Plathottam, Hossein Salehfar, Prakash Ranganathan |
| Abstract | The concept of automated power system data analysis using Deep Neural Networks (as part of the routine tasks normally performed by Independent System Operators) is explored and developed in this paper. Specifically, we propose to use the widely-used Deep neural network architecture known as Convolutional Neural Networks (CNNs). To this end, a 2-D representation of power system data is developed and proposed. To show the relevance of the proposed concept, a multi-class multi-label classification problem is presented as an application example. Midcontinent ISO (MISO) data sets on wind power and load is used for this purpose. TensorFlow, an open source machine learning platform is used to construct the CNN and train the network. The results are discussed and compared with those from standard Feed Forward Networks for the same data. |
| Tasks | Multi-Label Classification |
| Published | 2017-09-17 |
| URL | https://ieeexplore.ieee.org/abstract/document/8107202 |
| https://ieeexplore.ieee.org/abstract/document/8107202 | |
| PWC | https://paperswithcode.com/paper/convolutional-neural-networks-cnns-for-power |
| Repo | https://github.com/sibyjackgrove/CNN-on-Wind-Power-Data |
| Framework | none |
Unstructured point cloud semantic labelingusing deep segmentation networks
| Title | Unstructured point cloud semantic labelingusing deep segmentation networks |
| Authors | A. Boulch, B. Le Saux and N. Audebert |
| Abstract | In this work, we describe a new, general, and efficient method for unstructured point cloud labeling. As the question of efficiently using deep Convolutional Neural Networks (CNNs) on 3D data is still a pending issue, we propose a framework which applies CNNs on multiple 2D image views (or snapshots) of the point cloud. The approach consists in three core ideas. (i) We pick many suitable snapshots of the point cloud. We generate two types of images: a Red-Green-Blue (RGB) view and a depth composite view containing geometric features. (ii) We then perform a pixel-wise labeling of each pair of 2D snapshots using fully convolutional networks. Different architectures are tested to achieve a profitable fusion of our heterogeneous inputs. (iii) Finally, we perform fast back-projection of the label predictions in the 3D space using efficient buffering to label every 3D point. Experiments show that our method is suitable for various types of point clouds such as Lidar or photogrammetric data. |
| Tasks | Semantic Segmentation |
| Published | 2017-04-23 |
| URL | http://blesaux.free.fr/papers/17-EG3DOR-SnapNet-BoulchLeSauxAudebert-compressed.pdf |
| http://blesaux.free.fr/papers/17-EG3DOR-SnapNet-BoulchLeSauxAudebert-compressed.pdf | |
| PWC | https://paperswithcode.com/paper/unstructured-point-cloud-semantic |
| Repo | https://github.com/aboulch/snapnet |
| Framework | tf |
Speeding up Reinforcement Learning-based Information Extraction Training using Asynchronous Methods
| Title | Speeding up Reinforcement Learning-based Information Extraction Training using Asynchronous Methods |
| Authors | Aditya Sharma, Zarana Parekh, Partha Talukdar |
| Abstract | RLIE-DQN is a recently proposed Reinforcement Learning-based Information Extraction (IE) technique which is able to incorporate external evidence during the extraction process. RLIE-DQN trains a single agent sequentially, training on one instance at a time. This results in significant training slowdown which is undesirable. We leverage recent advances in parallel RL training using asynchronous methods and propose RLIE-A3C. RLIE-A3C trains multiple agents in parallel and is able to achieve upto 6x training speedup over RLIE-DQN, while suffering no loss in average accuracy. |
| Tasks | |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/D17-1281/ |
| https://www.aclweb.org/anthology/D17-1281 | |
| PWC | https://paperswithcode.com/paper/speeding-up-reinforcement-learning-based |
| Repo | https://github.com/adi-sharma/RLIE_A3C |
| Framework | tf |
Explicit Document Modeling through Weighted Multiple-Instance Learning
| Title | Explicit Document Modeling through Weighted Multiple-Instance Learning |
| Authors | Nikolaos Pappas, Andrei Popescu-Belis |
| Abstract | Representing documents is a crucial component in many NLP tasks, for instance predicting aspect ratings in reviews. Previous methods for this task treat documents globally and do not acknowledge that target categories are often assigned by their authors with generally no indication of the specific sentences that motivate them. To address this issue, we adopt a weakly supervised learning model, which jointly learns to focus on relevant parts of a document according to the context along with a classifier for the target categories. Originated from the weighted multiple-instance regression (MIR) framework, the modellearns decomposable document vectors for each individual category and thus overcomesthe representational bottleneck in previous methods due to a fixed-length document vector. During prediction, the estimated relevance or saliency weights explicitly capture the contribution of each sentence to the predicted rating, thus offering an explanation of the rating. Our model achieves state-of-the-art performance on multi-aspect sentiment analysis, improving over several baselines. Moreover, the predicted saliency weights are close to human estimates obtained by crowdsourcing and increase the performance of lexical and topical features for review segmentation and summarization. |
| Tasks | Multiple Instance Learning, Sentiment Analysis |
| Published | 2017-02-17 |
| URL | https://jair.org/index.php/jair/article/view/11051/26228 |
| https://jair.org/index.php/jair/article/view/11051/26228 | |
| PWC | https://paperswithcode.com/paper/explicit-document-modeling-through-weighted |
| Repo | https://github.com/idiap/wmil-sgd |
| Framework | none |
VegFru: A Domain-Specific Dataset for Fine-Grained Visual Categorization
| Title | VegFru: A Domain-Specific Dataset for Fine-Grained Visual Categorization |
| Authors | Saihui Hou, Yushan Feng, Zilei Wang |
| Abstract | VegFru: A Domain-Specific Dataset for Fine-grained Visual Categorization In this paper, we propose a novel domain-specific dataset named VegFru for fine-grained visual categorization (FGVC). While the existing datasets for FGVC are mainly focused on animal breeds or man-made objects with limited labelled data, VegFru is a larger dataset consisting of vegetables and fruits which are closely associated with the daily life of everyone. Aiming at domestic cooking and food management, VegFru categorizes vegetables and fruits according to their eating characteristics, and each image contains at least one edible part of vegetables or fruits with the same cooking usage. Particularly, all the images are labelled hierarchically. The current version covers vegetables and fruits of 25 upper-level categories and 292 subordinate classes. And it contains more than 160,000 images in total and at least 200 images for each subordinate class. Accompanying the dataset, we also propose an effective framework called HybridNet to exploit the label hierarchy for FGVC. Specifically, multiple granularity features are first extracted by dealing with the hierarchical labels separately. And then they are fused through explicit operation, e.g., Compact Bilinear Pooling, to form a unified representation for the ultimate recognition. The experimental results on the novel VegFru, the public FGVC-Aircraft and CUB-200-2011 indicate that HybridNet achieves one of the top performance on these datasets. The dataset and code are available at https://github.com/hshustc/vegfru. |
| Tasks | Fine-Grained Visual Categorization |
| Published | 2017-10-01 |
| URL | http://openaccess.thecvf.com/content_iccv_2017/html/Hou_VegFru_A_Domain-Specific_ICCV_2017_paper.html |
| http://openaccess.thecvf.com/content_ICCV_2017/papers/Hou_VegFru_A_Domain-Specific_ICCV_2017_paper.pdf | |
| PWC | https://paperswithcode.com/paper/vegfru-a-domain-specific-dataset-for-fine |
| Repo | https://github.com/ustc-vim/vegfru |
| Framework | none |
Efficient, Compositional, Order-sensitive n-gram Embeddings
| Title | Efficient, Compositional, Order-sensitive n-gram Embeddings |
| Authors | Adam Poliak, Pushpendre Rastogi, M. Patrick Martin, Benjamin Van Durme |
| Abstract | We propose ECO: a new way to generate embeddings for phrases that is Efficient, Compositional, and Order-sensitive. Our method creates decompositional embeddings for words offline and combines them to create new embeddings for phrases in real time. Unlike other approaches, ECO can create embeddings for phrases not seen during training. We evaluate ECO on supervised and unsupervised tasks and demonstrate that creating phrase embeddings that are sensitive to word order can help downstream tasks. |
| Tasks | Word Embeddings |
| Published | 2017-04-01 |
| URL | https://www.aclweb.org/anthology/E17-2081/ |
| https://www.aclweb.org/anthology/E17-2081 | |
| PWC | https://paperswithcode.com/paper/efficient-compositional-order-sensitive-n |
| Repo | https://github.com/azpoliak/eco |
| Framework | none |
Real-time Document Localization in Natural Images by Recursive Application of a CNN.
| Title | Real-time Document Localization in Natural Images by Recursive Application of a CNN. |
| Authors | Khurram Javed, Faisal Shafait |
| Abstract | We propose a document segmentation algorithm that recursively uses convolutional neural networks to precisely localize a document in a natural image. The system can run in real-time on a mobile CPU, has minimal storage requirements, and achieves results comparable to the state of the art on the simple backgrounds and considerably better (Improving previous 86% to 94%) than the state of the art on the complex background of the “ICDAR 2015 SmartDoc Competition 1” dataset. |
| Tasks | |
| Published | 2017-11-13 |
| URL | https://sites.ualberta.ca/~kjaved/ |
| https://khurramjaved96.github.io/RecursiveCNN.pdf | |
| PWC | https://paperswithcode.com/paper/real-time-document-localization-in-natural |
| Repo | https://github.com/Khurramjaved96/Recursive-CNNs |
| Framework | pytorch |
Ego-splitting Framework: from Non-Overlapping to Overlapping Clusters
| Title | Ego-splitting Framework: from Non-Overlapping to Overlapping Clusters |
| Authors | Alessandro Epasto, Silvio Lattanzi, Renato Paes Leme |
| Abstract | We propose a new framework called Ego-Splitting for detecting clusters in complex networks which leverage the local structures known as ego-nets (i.e. the subgraph induced by the neighborhood of each node) to de-couple overlapping clusters. Ego-Splitting is a highly scalable and flexible framework, with provable theoretical guarantees, that reduces the complex overlapping clustering problem to a simpler and more amenable non-overlapping (partitioning) problem. We can solve community detection in graphs with tens of billions of edges and outperform previous solutions based on ego-nets analysis. More precisely, our framework works in two steps: a local ego-net analysis phase, and a global graph partitioning phase . In the local step, we first partition the nodes’ ego-nets using a partitioning algorithm. We then use the computed clusters to split each node into its persona nodes that represent the instantiations of the node in its communities. Then, in the global step, we partition the newly created graph to obtain an overlapping clustering of the original graph. |
| Tasks | Community Detection, graph partitioning |
| Published | 2017-08-13 |
| URL | https://www.kdd.org/kdd2017/papers/view/ego-splitting-framework-from-non-overlapping-to-overlapping-clusters |
| https://www.epasto.org/papers/kdd2017.pdf | |
| PWC | https://paperswithcode.com/paper/ego-splitting-framework-from-non-overlapping |
| Repo | https://github.com/benedekrozemberczki/karateclub |
| Framework | none |
Dealing with Co-reference in Neural Semantic Parsing
| Title | Dealing with Co-reference in Neural Semantic Parsing |
| Authors | Rik van Noord, Johan Bos |
| Abstract | |
| Tasks | Amr Parsing, Semantic Parsing |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/W17-7306/ |
| https://www.aclweb.org/anthology/W17-7306 | |
| PWC | https://paperswithcode.com/paper/dealing-with-co-reference-in-neural-semantic |
| Repo | https://github.com/RikVN/AMR |
| Framework | tf |
