Paper Group NANR 3
An Environment for Relational Annotation of Political Debates. v-trel: Vocabulary Trainer for Tracing Word Relations - An Implicit Crowdsourcing Approach. Compressing Unknown Images With Product Quantizer for Efficient Zero-Shot Classification. Marginalized Latent Semantic Encoder for Zero-Shot Learning. Estimation of Energy Prices in Turkey in the …
An Environment for Relational Annotation of Political Debates
| Title | An Environment for Relational Annotation of Political Debates |
| Authors | Andre Blessing, Nico Blokker, Sebastian Haunss, Jonas Kuhn, Gabriella Lapesa, Sebastian Pad{'o} |
| Abstract | This paper describes the MARDY corpus annotation environment developed for a collaboration between political science and computational linguistics. The tool realizes the complete workflow necessary for annotating a large newspaper text collection with rich information about claims (demands) raised by politicians and other actors, including claim and actor spans, relations, and polarities. In addition to the annotation GUI, the tool supports the identification of relevant documents, text pre-processing, user management, integration of external knowledge bases, annotation comparison and merging, statistical analysis, and the incorporation of machine learning models as {``}pseudo-annotators{''}. | |
| Tasks | |
| Published | 2019-07-01 |
| URL | https://www.aclweb.org/anthology/P19-3018/ |
| https://www.aclweb.org/anthology/P19-3018 | |
| PWC | https://paperswithcode.com/paper/an-environment-for-relational-annotation-of |
| Repo | |
| Framework | |
v-trel: Vocabulary Trainer for Tracing Word Relations - An Implicit Crowdsourcing Approach
| Title | v-trel: Vocabulary Trainer for Tracing Word Relations - An Implicit Crowdsourcing Approach |
| Authors | Verena Lyding, Christos Rodosthenous, Federico Sangati, Umair ul Hassan, Lionel Nicolas, Alex K{"o}nig, er, Jolita Horbacauskiene, Anisia Katinskaia |
| Abstract | In this paper, we present our work on developing a vocabulary trainer that uses exercises generated from language resources such as ConceptNet and crowdsources the responses of the learners to enrich the language resource. We performed an empirical evaluation of our approach with 60 non-native speakers over two days, which shows that new entries to expand Concept-Net can efficiently be gathered through vocabulary exercises on word relations. We also report on the feedback gathered from the users and an expert from language teaching, and discuss the potential of the vocabulary trainer application from the user and language learner perspective. The feedback suggests that v-trel has educational potential, while in its current state some shortcomings could be identified. |
| Tasks | |
| Published | 2019-09-01 |
| URL | https://www.aclweb.org/anthology/R19-1079/ |
| https://www.aclweb.org/anthology/R19-1079 | |
| PWC | https://paperswithcode.com/paper/v-trel-vocabulary-trainer-for-tracing-word |
| Repo | |
| Framework | |
Compressing Unknown Images With Product Quantizer for Efficient Zero-Shot Classification
| Title | Compressing Unknown Images With Product Quantizer for Efficient Zero-Shot Classification |
| Authors | Jin Li, Xuguang Lan, Yang Liu, Le Wang, Nanning Zheng |
| Abstract | For Zero-Shot Learning (ZSL), the Nearest Neighbor (NN) search is generally conducted for classification, which may cause unacceptable computational complexity for large-scale datasets. To compress zero-shot classes by the trained quantizer for efficient search, it tends to induce large quantization error because distributions between seen and unseen classes are different. However, as semantic attributes of classes are available in ZSL, both seen and unseen classes have the same distribution for one specific property, e.g., animals have or not have spots. Based on this intuition, a Product Quantization Zero-Shot Learning (PQZSL) method is proposed to learn embeddings as well as quantizers to compress visual features into compact codes for Approximate NN (ANN) search. Particularly, visual features are projected into an orthogonal semantic space, and then the Product Quantization (PQ) is utilized to quantize individual properties. Experimental results on five benchmark datasets demonstrate that unseen classes are represented by the Cartesian product of quantized properties with little quantization error. As classes in orthogonal common space are more discriminative, the classification based on PQZSL achieves state-of-the-art performance in Generalized Zero-Shot Learning (GZSL) task, meanwhile, the speed of ANN search is 10-100 times higher than traditional NN search. |
| Tasks | Quantization, Zero-Shot Learning |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Li_Compressing_Unknown_Images_With_Product_Quantizer_for_Efficient_Zero-Shot_Classification_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Compressing_Unknown_Images_With_Product_Quantizer_for_Efficient_Zero-Shot_Classification_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/compressing-unknown-images-with-product |
| Repo | |
| Framework | |
Marginalized Latent Semantic Encoder for Zero-Shot Learning
| Title | Marginalized Latent Semantic Encoder for Zero-Shot Learning |
| Authors | Zhengming Ding, Hongfu Liu |
| Abstract | Zero-shot learning has been well explored to precisely identify new unobserved classes through a visual-semantic function obtained from the existing objects. However, there exist two challenging obstacles: one is that the human-annotated semantics are insufficient to fully describe the visual samples; the other is the domain shift across existing and new classes. In this paper, we attempt to exploit the intrinsic relationship in the semantic manifold when given semantics are not enough to describe the visual objects, and enhance the generalization ability of the visual-semantic function with marginalized strategy. Specifically, we design a Marginalized Latent Semantic Encoder (MLSE), which is learned on the augmented seen visual features and the latent semantic representation. Meanwhile, latent semantics are discovered under an adaptive graph reconstruction scheme based on the provided semantics. Consequently, our proposed algorithm could enrich visual characteristics from seen classes, and well generalize to unobserved classes. Experimental results on zero-shot benchmarks demonstrate that the proposed model delivers superior performance over the state-of-the-art zero-shot learning approaches. |
| Tasks | Zero-Shot Learning |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Ding_Marginalized_Latent_Semantic_Encoder_for_Zero-Shot_Learning_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Ding_Marginalized_Latent_Semantic_Encoder_for_Zero-Shot_Learning_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/marginalized-latent-semantic-encoder-for-zero |
| Repo | |
| Framework | |
Estimation of Energy Prices in Turkey in the Nash-Cournot Framework
| Title | Estimation of Energy Prices in Turkey in the Nash-Cournot Framework |
| Authors | Prof. Dr. Rengin Ak; Assist. Prof. Dr. Armağan Türk; Assoc. Prof. Dr. Hasan İslatince |
| Abstract | Increase of mass production with the industrial revolution has increased the country’s dependence on energy. Even today, energy constitutes the main weight of industrialization. Especially for countries dependent on energy from abroad, energy prices are very important, and fluctuations in energy prices are decisive for the economies of these countries. Fluctuations in energy prices make the future course of energy prices important for both energy demanding countries and energy-supplying countries. There are many methods in the related literature. Our estimate of energy prices for Turkey will be based on the Nash-Cournot framework. In the Nash-Cournot framework, energy is considered as a homogeneous commodity and market equilibrium is determined by the capacity-setting decisions of the suppliers. The model indicates that competitors aim to produce more by reacting to higher prices. A fundamental offer-based stochastic model is being put forward in order to predict the energy prices and the mean price in a given period. Two ambiguous sources are addressed, these are energy producers and demand. Studies have shown that the expected number of prices increases significantly with the decrease in the number of firms in the market. Within this framework, factors determining the energy prices in Turkey will be determined and forecasts for the future will be made. |
| Tasks | |
| Published | 2019-07-30 |
| URL | https://ijbassnet.com/publication/256/details |
| https://ijbassnet.com/storage/app/publications/5d401a56a436011564482134.pdf | |
| PWC | https://paperswithcode.com/paper/estimation-of-energy-prices-in-turkey-in-the |
| Repo | |
| Framework | |
Single Shot Neural Architecture Search Via Direct Sparse Optimization
| Title | Single Shot Neural Architecture Search Via Direct Sparse Optimization |
| Authors | Xinbang Zhang, Zehao Huang, Naiyan Wang |
| Abstract | Recently Neural Architecture Search (NAS) has aroused great interest in both academia and industry, however it remains challenging because of its huge and non-continuous search space. Instead of applying evolutionary algorithm or reinforcement learning as previous works, this paper proposes a Direct Sparse Optimization NAS (DSO-NAS) method. In DSO-NAS, we provide a novel model pruning view to NAS problem. In specific, we start from a completely connected block, and then introduce scaling factors to scale the information flow between operations. Next, we impose sparse regularizations to prune useless connections in the architecture. Lastly, we derive an efficient and theoretically sound optimization method to solve it. Our method enjoys both advantages of differentiability and efficiency, therefore can be directly applied to large datasets like ImageNet. Particularly, On CIFAR-10 dataset, DSO-NAS achieves an average test error 2.84%, while on the ImageNet dataset DSO-NAS achieves 25.4% test error under 600M FLOPs with 8 GPUs in 18 hours. |
| Tasks | Neural Architecture Search |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=ryxjH3R5KQ |
| https://openreview.net/pdf?id=ryxjH3R5KQ | |
| PWC | https://paperswithcode.com/paper/single-shot-neural-architecture-search-via |
| Repo | |
| Framework | |
A Survey on CSI-Based Human Behavior Recognition in Through-the-Wall Scenario
| Title | A Survey on CSI-Based Human Behavior Recognition in Through-the-Wall Scenario |
| Authors | Zhengjie Wang, Kangkang Jiang, Yushan Hou, Zehua Huang, Wenwen Dou, Chengming Zhang, Yinjing Guo |
| Abstract | Recent years have witnessed increasing research interest in human behavior recognition as it provides attractive applications in various sensing scenarios. Among these encouraging implementations, device-free behavior recognition based on WiFi channel state information (CSI) has attracted significant attention due to the popularity of WiFi devices and abundant channel characteristics from CSI. Meanwhile, the CSI signal provides us with additional benefits because it can propagate through a wall. This through-the-wall and device-free scheme not only enables us to identify specific human actions but also to infer person activities by collecting the data from different rooms. This paper presents a survey on the state-of-art progresses in device-free through-the-wall human behavior recognition based on CSI. Specifically, this paper first introduces the basic concept of CSI and describes the signal variation caused by human behavior. Then, it illustrates that different human behaviors can cause signal transformation. Therefore, the unique map relationship between action and signal variation can be leveraged to recognize human behavior. Next, it provides the general architecture of through-the-wall behavior recognition and highlights its core characteristic. It investigates the state-of-art applications in various scenarios and analyzes specific design schemes and implementations. Afterward, it discusses the various across wall applications and makes a detailed comparison between non-through-the-wall and through-the-wall applications. Meanwhile, it analyzes many factors that affect recognition accuracy and emphasizes performance differences under across wall scenarios. Finally, this paper concludes by summarizing the issues and challenges faced and providing insights into the possible solution and future research trend. |
| Tasks | RF-based Action Recognition, RF-based Pose Estimation |
| Published | 2019-06-12 |
| URL | https://doi.org/10.1109/ACCESS.2019.2922244 |
| https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8735849 | |
| PWC | https://paperswithcode.com/paper/a-survey-on-csi-based-human-behavior |
| Repo | |
| Framework | |
Learning agents with prioritization and parameter noise in continuous state and action space
| Title | Learning agents with prioritization and parameter noise in continuous state and action space |
| Authors | Rajesh Devaraddi, G. Srinivasaraghavan |
| Abstract | Reinforcement Learning (RL) problem can be solved in two different ways - the Value function-based approach and the policy optimization-based approach - to eventually arrive at an optimal policy for the given environment. One of the recent breakthroughs in reinforcement learning is the use of deep neural networks as function approximators to approximate the value function or q-function in a reinforcement learning scheme. This has led to results with agents automatically learning how to play games like alpha-go showing better-than-human performance. Deep Q-learning networks (DQN) and Deep Deterministic Policy Gradient (DDPG) are two such methods that have shown state-of-the-art results in recent times. Among the many variants of RL, an important class of problems is where the state and action spaces are continuous — autonomous robots, autonomous vehicles, optimal control are all examples of such problems that can lend themselves naturally to reinforcement based algorithms, and have continuous state and action spaces. In this paper, we adapt and combine approaches such as DQN and DDPG in novel ways to outperform the earlier results for continuous state and action space problems. We believe these results are a valuable addition to the fast-growing body of results on Reinforcement Learning, more so for continuous state and action space problems. |
| Tasks | Autonomous Vehicles, Q-Learning |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=ryGiYoAqt7 |
| https://openreview.net/pdf?id=ryGiYoAqt7 | |
| PWC | https://paperswithcode.com/paper/learning-agents-with-prioritization-and |
| Repo | |
| Framework | |
Learning Mixed-Curvature Representations in Product Spaces
| Title | Learning Mixed-Curvature Representations in Product Spaces |
| Authors | Albert Gu, Frederic Sala, Beliz Gunel, Christopher Ré |
| Abstract | The quality of the representations achieved by embeddings is determined by how well the geometry of the embedding space matches the structure of the data. Euclidean space has been the workhorse space for embeddings; recently hyperbolic and spherical spaces are gaining popularity due to their ability to better embed new types of structured data—such as hierarchical data—but most data is not structured so uniformly. We address this problem by proposing embedding into a product manifold combining multiple copies of spherical, hyperbolic, and Euclidean spaces, providing a space of heterogeneous curvature suitable for a wide variety of structures. We introduce a heuristic to estimate the sectional curvature of graph data and directly determine the signature—the number of component spaces and their dimensions—of the product manifold. Empirically, we jointly learn the curvature and the embedding in the product space via Riemannian optimization. We discuss how to define and compute intrinsic quantities such as means—a challenging notion for product manifolds—and provably learnable optimization functions. On a range of datasets and reconstruction tasks, our product space embeddings outperform single Euclidean or hyperbolic spaces used in previous works, reducing distortion by 32.55% on a Facebook social network dataset. We learn word embeddings and find that a product of hyperbolic spaces in 50 dimensions consistently improves on baseline Euclidean and hyperbolic embeddings by 2.6 points in Spearman rank correlation on similarity tasks and 3.4 points on analogy accuracy. |
| Tasks | Word Embeddings |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=HJxeWnCcF7 |
| https://openreview.net/pdf?id=HJxeWnCcF7 | |
| PWC | https://paperswithcode.com/paper/learning-mixed-curvature-representations-in |
| Repo | |
| Framework | |
DEEP GEOMETRICAL GRAPH CLASSIFICATION
| Title | DEEP GEOMETRICAL GRAPH CLASSIFICATION |
| Authors | Mostafa Rahmani, Ping Li |
| Abstract | Most of the existing Graph Neural Networks (GNNs) are the mere extension of the Convolutional Neural Networks (CNNs) to graphs. Generally, they consist of several steps of message passing between the nodes followed by a global indiscriminate feature pooling function. In many data-sets, however, the nodes are unlabeled or their labels provide no information about the similarity between the nodes and the locations of the nodes in the graph. Accordingly, message passing may not propagate helpful information throughout the graph. We show that this conventional approach can fail to learn to perform even simple graph classification tasks. We alleviate this serious shortcoming of the GNNs by making them a two step method. In the first of the proposed approach, a graph embedding algorithm is utilized to obtain a continuous feature vector for each node of the graph. The embedding algorithm represents the graph as a point-cloud in the embedding space. In the second step, the GNN is applied to the point-cloud representation of the graph provided by the embedding method. The GNN learns to perform the given task by inferring the topological structure of the graph encoded in the spatial distribution of the embedded vectors. In addition, we extend the proposed approach to the graph clustering problem and a new architecture for graph clustering is proposed. Moreover, the spatial representation of the graph is utilized to design a graph pooling algorithm. We turn the problem of graph down-sampling into a column sampling problem, i.e., the sampling algorithm selects a subset of the nodes whose feature vectors preserve the spatial distribution of all the feature vectors. We apply the proposed approach to several popular benchmark data-sets and it is shown that the proposed geometrical approach strongly improves the state-of-the-art result for several data-sets. For instance, for the PTC data-set, we improve the state-of-the-art result for more than 22 %. |
| Tasks | Graph Classification, Graph Clustering, Graph Embedding |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=Hkes0iR9KX |
| https://openreview.net/pdf?id=Hkes0iR9KX | |
| PWC | https://paperswithcode.com/paper/deep-geometrical-graph-classification |
| Repo | |
| Framework | |
Flow-based Image-to-Image Translation with Feature Disentanglement
| Title | Flow-based Image-to-Image Translation with Feature Disentanglement |
| Authors | Ruho Kondo, Keisuke Kawano, Satoshi Koide, Takuro Kutsuna |
| Abstract | Learning non-deterministic dynamics and intrinsic factors from images obtained through physical experiments is at the intersection of machine learning and material science. Disentangling the origins of uncertainties involved in microstructure growth, for example, is of great interest because future states vary due to thermal fluctuation and other environmental factors. To this end we propose a flow-based image-to-image model, called Flow U-Net with Squeeze modules (FUNS), that allows us to disentangle the features while retaining the ability to generate highquality diverse images from condition images. Our model successfully captures probabilistic phenomena by incorporating a U-Net-like architecture into the flowbased model. In addition, our model automatically separates the diversity of target images into condition-dependent/independent parts. We demonstrate that the quality and diversity of the images generated for microstructure growth and CelebA datasets outperform existing variational generative models. |
| Tasks | Image-to-Image Translation |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/8670-flow-based-image-to-image-translation-with-feature-disentanglement |
| http://papers.nips.cc/paper/8670-flow-based-image-to-image-translation-with-feature-disentanglement.pdf | |
| PWC | https://paperswithcode.com/paper/flow-based-image-to-image-translation-with |
| Repo | |
| Framework | |
Collaborative Dialogue in Minecraft
| Title | Collaborative Dialogue in Minecraft |
| Authors | Anjali Narayan-Chen, Prashant Jayannavar, Julia Hockenmaier |
| Abstract | We wish to develop interactive agents that can communicate with humans to collaboratively solve tasks in grounded scenarios. Since computer games allow us to simulate such tasks without the need for physical robots, we define a Minecraft-based collaborative building task in which one player (A, the Architect) is shown a target structure and needs to instruct the other player (B, the Builder) to build this structure. Both players interact via a chat interface. A can observe B but cannot place blocks. We present the Minecraft Dialogue Corpus, a collection of 509 conversations and game logs. As a first step towards our goal of developing fully interactive agents for this task, we consider the subtask of Architect utterance generation, and show how challenging it is. |
| Tasks | |
| Published | 2019-07-01 |
| URL | https://www.aclweb.org/anthology/P19-1537/ |
| https://www.aclweb.org/anthology/P19-1537 | |
| PWC | https://paperswithcode.com/paper/collaborative-dialogue-in-minecraft |
| Repo | |
| Framework | |
A Fast and Accurate Partially Deterministic Morphological Analysis
| Title | A Fast and Accurate Partially Deterministic Morphological Analysis |
| Authors | Hajime Morita, Tomoya Iwakura |
| Abstract | This paper proposes a partially deterministic morphological analysis method for improved processing speed. Maximum matching is a fast deterministic method for morphological analysis. However, the method tends to decrease performance due to lack of consideration of contextual information. In order to use maximum matching safely, we propose the use of Context Independent Strings (CISs), which are strings that do not have ambiguity in terms of morphological analysis. Our method first identifies CISs in a sentence using maximum matching without contextual information, then analyzes the unprocessed part of the sentence using a bi-gram-based morphological analysis model. We evaluate the method on a Japanese morphological analysis task. The experimental results show a 30{%} reduction of running time while maintaining improved accuracy. |
| Tasks | Morphological Analysis |
| Published | 2019-09-01 |
| URL | https://www.aclweb.org/anthology/R19-1093/ |
| https://www.aclweb.org/anthology/R19-1093 | |
| PWC | https://paperswithcode.com/paper/a-fast-and-accurate-partially-deterministic |
| Repo | |
| Framework | |
With or without post-editing processes? Evidence for a gap in machine translation evaluation
| Title | With or without post-editing processes? Evidence for a gap in machine translation evaluation |
| Authors | Caroline Rossi, Emmanuelle Esperan{\c{c}}a-Rodier |
| Abstract | |
| Tasks | Machine Translation |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-7005/ |
| https://www.aclweb.org/anthology/W19-7005 | |
| PWC | https://paperswithcode.com/paper/with-or-without-post-editing-processes |
| Repo | |
| Framework | |
Long-Distance Dependencies Don’t Have to Be Long: Simplifying through Provably (Approximately) Optimal Permutations
| Title | Long-Distance Dependencies Don’t Have to Be Long: Simplifying through Provably (Approximately) Optimal Permutations |
| Authors | Rishi Bommasani |
| Abstract | Neural models at the sentence level often operate on the constituent words/tokens in a way that encodes the inductive bias of processing the input in a similar fashion to how humans do. However, there is no guarantee that the standard ordering of words is computationally efficient or optimal. To help mitigate this, we consider a dependency parse as a proxy for the inter-word dependencies in a sentence and simplify the sentence with respect to combinatorial objectives imposed on the sentence-parse pair. The associated optimization results in permuted sentences that are provably (approximately) optimal with respect to minimizing dependency parse lengths and that are demonstrably simpler. We evaluate our general-purpose permutations within a fine-tuning schema for the downstream task of subjectivity analysis. Our fine-tuned baselines reflect a new state of the art for the SUBJ dataset and the permutations we introduce lead to further improvements with a 2.0{%} increase in classification accuracy (absolute) and a 45{%} reduction in classification error (relative) over the previous state of the art. |
| Tasks | Subjectivity Analysis |
| Published | 2019-07-01 |
| URL | https://www.aclweb.org/anthology/P19-2012/ |
| https://www.aclweb.org/anthology/P19-2012 | |
| PWC | https://paperswithcode.com/paper/long-distance-dependencies-dont-have-to-be |
| Repo | |
| Framework | |
