Paper Group NANR 252
Beyond the One-Step Greedy Approach in Reinforcement Learning. SINAI at SemEval-2018 Task 1: Emotion Recognition in Tweets. Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing. A Weighted Sparse Sampling and Smoothing Frame Transition Approach for Semantic Fast-Forward First-Person Videos. CENTEMENT at SemEval-2018 Task 1: Cl …
Beyond the One-Step Greedy Approach in Reinforcement Learning
| Title | Beyond the One-Step Greedy Approach in Reinforcement Learning |
| Authors | Yonathan Efroni, Gal Dalal, Bruno Scherrer, Shie Mannor |
| Abstract | The famous Policy Iteration algorithm alternates between policy improvement and policy evaluation. Implementations of this algorithm with several variants of the latter evaluation stage, e.g, n-step and trace-based returns, have been analyzed in previous works. However, the case of multiple-step lookahead policy improvement, despite the recent increase in empirical evidence of its strength, has to our knowledge not been carefully analyzed yet. In this work, we introduce the first such analysis. Namely, we formulate variants of multiple-step policy improvement, derive new algorithms using these definitions and prove their convergence. Moreover, we show that recent prominent Reinforcement Learning algorithms are, in fact, instances of our framework. We thus shed light on their empirical success and give a recipe for deriving new algorithms for future study. |
| Tasks | |
| Published | 2018-07-01 |
| URL | https://icml.cc/Conferences/2018/Schedule?showEvent=2126 |
| http://proceedings.mlr.press/v80/efroni18a/efroni18a.pdf | |
| PWC | https://paperswithcode.com/paper/beyond-the-one-step-greedy-approach-in |
| Repo | |
| Framework | |
SINAI at SemEval-2018 Task 1: Emotion Recognition in Tweets
| Title | SINAI at SemEval-2018 Task 1: Emotion Recognition in Tweets |
| Authors | Flor Miriam Plaza-del-Arco, Salud Mar{'\i}a Jim{'e}nez-Zafra, Maite Martin, L. Alfonso Ure{~n}a-L{'o}pez |
| Abstract | Emotion classification is a new task that combines several disciplines including Artificial Intelligence and Psychology, although Natural Language Processing is perhaps the most challenging area. In this paper, we describe our participation in SemEval-2018 Task1: Affect in Tweets. In particular, we have participated in EI-oc, EI-reg and E-c subtasks for English and Spanish languages. |
| Tasks | Emotion Classification, Emotion Recognition |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-1017/ |
| https://www.aclweb.org/anthology/S18-1017 | |
| PWC | https://paperswithcode.com/paper/sinai-at-semeval-2018-task-1-emotion |
| Repo | |
| Framework | |
Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing
| Title | Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing |
| Authors | Daniel Fern{'a}ndez-Gonz{'a}lez, Carlos G{'o}mez-Rodr{'\i}guez |
| Abstract | We introduce novel dynamic oracles for training two of the most accurate known shift-reduce algorithms for constituent parsing: the top-down and in-order transition-based parsers. In both cases, the dynamic oracles manage to notably increase their accuracy, in comparison to that obtained by performing classic static training. In addition, by improving the performance of the state-of-the-art in-order shift-reduce parser, we achieve the best accuracy to date (92.0 F1) obtained by a fully-supervised single-model greedy shift-reduce constituent parser on the WSJ benchmark. |
| Tasks | |
| Published | 2018-10-01 |
| URL | https://www.aclweb.org/anthology/D18-1161/ |
| https://www.aclweb.org/anthology/D18-1161 | |
| PWC | https://paperswithcode.com/paper/dynamic-oracles-for-top-down-and-in-order-1 |
| Repo | |
| Framework | |
A Weighted Sparse Sampling and Smoothing Frame Transition Approach for Semantic Fast-Forward First-Person Videos
| Title | A Weighted Sparse Sampling and Smoothing Frame Transition Approach for Semantic Fast-Forward First-Person Videos |
| Authors | Michel Silva, Washington Ramos, João Ferreira, Felipe Chamone, Mario Campos, Erickson R. Nascimento |
| Abstract | Thanks to the advances in the technology of low-cost digital cameras and the popularity of the self-recording culture, the amount of visual data on the Internet is going to the opposite side of the available time and patience of the users. Thus, most of the uploaded videos are doomed to be forgotten and unwatched in a computer folder or website. In this work, we address the problem of creating smooth fast-forward videos without losing the relevant content. We present a new adaptive frame selection formulated as a weighted minimum reconstruction problem, which combined with a smoothing frame transition method accelerates first-person videos emphasizing the relevant segments and avoids visual discontinuities. The experiments show that our method is able to fast-forward videos to retain as much relevant information and smoothness as the state-of-the-art techniques in less time. We also present a new 80-hour multimodal (RGB-D, IMU, and GPS) dataset of first-person videos with annotations for recorder profile, frame scene, activities, interaction, and attention. |
| Tasks | |
| Published | 2018-06-01 |
| URL | http://openaccess.thecvf.com/content_cvpr_2018/html/Silva_A_Weighted_Sparse_CVPR_2018_paper.html |
| http://openaccess.thecvf.com/content_cvpr_2018/papers/Silva_A_Weighted_Sparse_CVPR_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/a-weighted-sparse-sampling-and-smoothing |
| Repo | |
| Framework | |
CENTEMENT at SemEval-2018 Task 1: Classification of Tweets using Multiple Thresholds with Self-correction and Weighted Conditional Probabilities
| Title | CENTEMENT at SemEval-2018 Task 1: Classification of Tweets using Multiple Thresholds with Self-correction and Weighted Conditional Probabilities |
| Authors | Tariq Ahmad, Allan Ramsay, Hanady Ahmed |
| Abstract | In this paper we present our contribution to SemEval-2018, a classifier for classifying multi-label emotions of Arabic and English tweets. We attempted {``}Affect in Tweets{''}, specifically Task E-c: Detecting Emotions (multi-label classification). Our method is based on preprocessing the tweets and creating word vectors combined with a self correction step to remove noise. We also make use of emotion specific thresholds. The final submission was selected upon the best performance achieved, selected when using a range of thresholds. Our system was evaluated on the Arabic and English datasets provided for the task by the competition organisers, where it ranked 2nd for the Arabic dataset (out of 14 entries) and 12th for the English dataset (out of 35 entries). | |
| Tasks | Multi-Label Classification, Opinion Mining, Sentiment Analysis |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-1030/ |
| https://www.aclweb.org/anthology/S18-1030 | |
| PWC | https://paperswithcode.com/paper/centement-at-semeval-2018-task-1 |
| Repo | |
| Framework | |
CENNLP at SemEval-2018 Task 2: Enhanced Distributed Representation of Text using Target Classes for Emoji Prediction Representation
| Title | CENNLP at SemEval-2018 Task 2: Enhanced Distributed Representation of Text using Target Classes for Emoji Prediction Representation |
| Authors | Naveen J R, Hariharan V, Barathi Ganesh H. B., An Kumar M, , Soman K P |
| Abstract | Emoji is one of the {``}fastest growing language {''} in pop-culture, especially in social media and it is very unlikely for its usage to decrease. These are generally used to bring an extra level of meaning to the texts, posted on social media platforms. Providing such an added info, gives more insights to the plain text, arising to hidden interpretation within the text. This paper explains our analysis on Task 2, {''} Multilingual Emoji Prediction{''} sharedtask conducted by Semeval-2018. In the task, a predicted emoji based on a piece of Twitter text are labelled under 20 different classes (most commonly used emojis) where these classes are learnt and further predicted are made for unseen Twitter text. In this work, we have experimented and analysed emojis predicted based on Twitter text, as a classification problem where the entailing emoji is considered as a label for every individual text data. We have implemented this using distributed representation of text through fastText. Also, we have made an effort to demonstrate how fastText framework can be useful in case of emoji prediction. This task is divide into two subtask, they are based on dataset presented in two different languages English and Spanish. | |
| Tasks | Opinion Mining, Sentiment Analysis |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-1078/ |
| https://www.aclweb.org/anthology/S18-1078 | |
| PWC | https://paperswithcode.com/paper/cennlp-at-semeval-2018-task-2-enhanced |
| Repo | |
| Framework | |
TAJJEB at SemEval-2018 Task 2: Traditional Approaches Just Do the Job with Emoji Prediction
| Title | TAJJEB at SemEval-2018 Task 2: Traditional Approaches Just Do the Job with Emoji Prediction |
| Authors | Angelo Basile, Kenny W. Lino |
| Abstract | Emojis are widely used on social media andunderstanding their meaning is important forboth practical purposes (e.g. opinion mining,sentiment detection) and theoretical purposes(e.g. how different L1 speakers use them, dothey have some syntax?); this paper presents aset of experiments that aim to predict a singleemoji from a tweet. We built different mod-els and we found that the test results are verydifferent from the validation results. |
| Tasks | Opinion Mining, Sentiment Analysis, Word Embeddings |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-1075/ |
| https://www.aclweb.org/anthology/S18-1075 | |
| PWC | https://paperswithcode.com/paper/tajjeb-at-semeval-2018-task-2-traditional |
| Repo | |
| Framework | |
What’s in Your Embedding, And How It Predicts Task Performance
| Title | What’s in Your Embedding, And How It Predicts Task Performance |
| Authors | Anna Rogers, Shashwath Hosur Ananthakrishna, Anna Rumshisky |
| Abstract | Attempts to find a single technique for general-purpose intrinsic evaluation of word embeddings have so far not been successful. We present a new approach based on scaled-up qualitative analysis of word vector neighborhoods that quantifies interpretable characteristics of a given model (e.g. its preference for synonyms or shared morphological forms as nearest neighbors). We analyze 21 such factors and show how they correlate with performance on 14 extrinsic and intrinsic task datasets (and also explain the lack of correlation between some of them). Our approach enables multi-faceted evaluation, parameter search, and generally {–} a more principled, hypothesis-driven approach to development of distributional semantic representations. |
| Tasks | Named Entity Recognition, Semantic Role Labeling, Word Embeddings |
| Published | 2018-08-01 |
| URL | https://www.aclweb.org/anthology/C18-1228/ |
| https://www.aclweb.org/anthology/C18-1228 | |
| PWC | https://paperswithcode.com/paper/whats-in-your-embedding-and-how-it-predicts |
| Repo | |
| Framework | |
High Performance Zero-Memory Overhead Direct Convolutions
| Title | High Performance Zero-Memory Overhead Direct Convolutions |
| Authors | Jiyuan Zhang, Franz Franchetti, Tze Meng Low |
| Abstract | The computation of convolution layers in deep neural networks typically rely on high performance routines that trade space for time by using additional memory (either for packing purposes or required as part of the algorithm) to improve performance. The problems with such an approach are two-fold. First, these routines incur additional memory overhead which reduces the overall size of the network that can fit on embedded devices with limited memory capacity. Second, these high performance routines were not optimized for performing convolution, which means that the performance obtained is usually less than conventionally expected. In this paper, we demonstrate that direct convolution, when implemented correctly, eliminates all memory overhead, and yields performance that is between 10% to 400% times better than existing high performance implementations of convolution layers on conventional and embedded CPU architectures. We also show that a high performance direct convolution exhibits better scaling performance, i.e. suffers less performance drop, when increasing the number of threads. |
| Tasks | |
| Published | 2018-07-01 |
| URL | https://icml.cc/Conferences/2018/Schedule?showEvent=2341 |
| http://proceedings.mlr.press/v80/zhang18d/zhang18d.pdf | |
| PWC | https://paperswithcode.com/paper/high-performance-zero-memory-overhead-direct |
| Repo | |
| Framework | |
A Fine-grained Large-scale Analysis of Coreference Projection
| Title | A Fine-grained Large-scale Analysis of Coreference Projection |
| Authors | Michal Nov{'a}k |
| Abstract | We perform a fine-grained large-scale analysis of coreference projection. By projecting gold coreference from Czech to English and vice versa on Prague Czech-English Dependency Treebank 2.0 Coref, we set an upper bound of a proposed projection approach for these two languages. We undertake a detailed thorough analysis that combines the analysis of projection{'}s subtasks with analysis of performance on individual mention types. The findings are accompanied with examples from the corpus. |
| Tasks | Opinion Mining, Part-Of-Speech Tagging, Semantic Role Labeling |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/W18-0709/ |
| https://www.aclweb.org/anthology/W18-0709 | |
| PWC | https://paperswithcode.com/paper/a-fine-grained-large-scale-analysis-of |
| Repo | |
| Framework | |
AugGAN: Cross Domain Adaptation with GAN-based Data Augmentation
| Title | AugGAN: Cross Domain Adaptation with GAN-based Data Augmentation |
| Authors | Sheng-Wei Huang, Che-Tsung Lin, Shu-Ping Chen, Yen-Yi Wu, Po-Hao Hsu, Shang-Hong Lai |
| Abstract | Deep learning based image-to-image translation methods aim at learning the joint distribution of the two domains and finding transformations between them. Despite recent GAN (Generative Adversarial Network) based methods have shown compelling visual results, they are prone to fail at preserving image-objects and maintaining translation consistency when faced with large and complex domain shifts, which reduces their practicality on tasks such as generating large-scale training data for different domains. To address this problem, we purpose a weakly supervised structure-aware image-to-image translation network, which is composed of encoders, generators, discriminators and parsing nets for the two domains, respectively, in a unified framework. The purposed network generates more visually plausible images of a different domain compared to the competing methods on different image-translation tasks. In addition, we quantitatively evaluate different methods by training Faster-RCNN and YOLO with datasets generated from the image-translation results and demonstrate significant improvement of the detection accuracies by using the proposed image-object preserving network. |
| Tasks | Data Augmentation, Domain Adaptation, Image-to-Image Translation |
| Published | 2018-09-01 |
| URL | http://openaccess.thecvf.com/content_ECCV_2018/html/Sheng-Wei_Huang_AugGAN_Cross_Domain_ECCV_2018_paper.html |
| http://openaccess.thecvf.com/content_ECCV_2018/papers/Sheng-Wei_Huang_AugGAN_Cross_Domain_ECCV_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/auggan-cross-domain-adaptation-with-gan-based |
| Repo | |
| Framework | |
Distilling the Posterior in Bayesian Neural Networks
| Title | Distilling the Posterior in Bayesian Neural Networks |
| Authors | Kuan-Chieh Wang, Paul Vicol, James Lucas, Li Gu, Roger Grosse, Richard Zemel |
| Abstract | Bayesian neural networks (BNNs) allow us to reason about uncertainty in a principled way. Stochastic Gradient Langevin Dynamics (SGLD) enables efficient BNN learning by drawing samples from the BNN posterior using mini-batches. However, SGLD and its extensions require storage of many copies of the model parameters, a potentially prohibitive cost, especially for large neural networks. We propose a framework, Adversarial Posterior Distillation, to distill the SGLD samples using a Generative Adversarial Network (GAN). At test-time, samples are generated by the GAN. We show that this distillation framework incurs no loss in performance on recent BNN applications including anomaly detection, active learning, and defense against adversarial attacks. By construction, our framework distills not only the Bayesian predictive distribution, but the posterior itself. This allows one to compute quantities such as the approximate model variance, which is useful in downstream tasks. To our knowledge, these are the first results applying MCMC-based BNNs to the aforementioned applications. |
| Tasks | Active Learning, Anomaly Detection |
| Published | 2018-07-01 |
| URL | https://icml.cc/Conferences/2018/Schedule?showEvent=2362 |
| http://proceedings.mlr.press/v80/wang18i/wang18i.pdf | |
| PWC | https://paperswithcode.com/paper/distilling-the-posterior-in-bayesian-neural |
| Repo | |
| Framework | |
CoDetect: Financial Fraud Detection With Anomaly Feature Detection
| Title | CoDetect: Financial Fraud Detection With Anomaly Feature Detection |
| Authors | DONGXU HUANG, DEJUN MU, LIBIN YANG, AND XIAOYAN CAI |
| Abstract | Financial fraud, such as money laundering, is known to be a serious process of crime that makes illegitimately obtained funds go to terrorism or other criminal activity. This kind of illegal activities involve complex networks of trade and financial transactions, which makes it difficult to detect the fraud entities and discover the features of fraud. Fortunately, trading/transaction network and features of entities in the network can be constructed from the complex networks of the trade and financial transactions. The trading/transaction network reveals the interaction between entities, and thus anomaly detection on trading networks can reveal the entities involved in the fraud activity; while features of entities are the description of entities, and anomaly detection on features can reflect details of the fraud activities. Thus, network and features provide complementary information for fraud detection, which has potential to improve fraud detection performance. However, the majority of existing methods focus on networks or features information separately, which does not utilize both information. In this paper, we propose a novel fraud detection framework, CoDetect, which can leverage both network information and feature information for financial fraud detection. In addition, the CoDetect can simultaneously detecting financial fraud activities and the feature patterns associated with the fraud activities. Extensive experiments on both synthetic data and real-world data demonstrate the efficiency and the effectiveness of the proposed framework in combating financial fraud, especially for money laundering. |
| Tasks | Anomaly Detection, Fraud Detection |
| Published | 2018-03-26 |
| URL | https://ieeexplore.ieee.org/abstract/document/8325544/ |
| https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8325544 | |
| PWC | https://paperswithcode.com/paper/codetect-financial-fraud-detection-with |
| Repo | |
| Framework | |
Temporal alignment and latent Gaussian process factor inference in population spike trains
| Title | Temporal alignment and latent Gaussian process factor inference in population spike trains |
| Authors | Lea Duncker, Maneesh Sahani |
| Abstract | We introduce a novel scalable approach to identifying common latent structure in neural population spike-trains, which allows for variability both in the trajectory and in the rate of progression of the underlying computation. Our approach is based on shared latent Gaussian processes (GPs) which are combined linearly, as in the Gaussian Process Factor Analysis (GPFA) algorithm. We extend GPFA to handle unbinned spike-train data by incorporating a continuous time point-process likelihood model, achieving scalability with a sparse variational approximation. Shared variability is separated into terms that express condition dependence, as well as trial-to-trial variation in trajectories. Finally, we introduce a nested GP formulation to capture variability in the rate of evolution along the trajectory. We show that the new method learns to recover latent trajectories in synthetic data, and can accurately identify the trial-to-trial timing of movement-related parameters from motor cortical data without any supervision. |
| Tasks | Gaussian Processes |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/8245-temporal-alignment-and-latent-gaussian-process-factor-inference-in-population-spike-trains |
| http://papers.nips.cc/paper/8245-temporal-alignment-and-latent-gaussian-process-factor-inference-in-population-spike-trains.pdf | |
| PWC | https://paperswithcode.com/paper/temporal-alignment-and-latent-gaussian |
| Repo | |
| Framework | |
Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization
| Title | Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization |
| Authors | Umut Simsekli, Cagatay Yildiz, Than Huy Nguyen, Taylan Cemgil, Gael Richard |
| Abstract | Recent studies have illustrated that stochastic gradient Markov Chain Monte Carlo techniques have a strong potential in non-convex optimization, where local and global convergence guarantees can be shown under certain conditions. By building up on this recent theory, in this study, we develop an asynchronous-parallel stochastic L-BFGS algorithm for non-convex optimization. The proposed algorithm is suitable for both distributed and shared-memory settings. We provide formal theoretical analysis and show that the proposed method achieves an ergodic convergence rate of ${\cal O}(1/\sqrt{N})$ ($N$ being the total number of iterations) and it can achieve a linear speedup under certain conditions. We perform several experiments on both synthetic and real datasets. The results support our theory and show that the proposed algorithm provides a significant speedup over the recently proposed synchronous distributed L-BFGS algorithm. |
| Tasks | |
| Published | 2018-07-01 |
| URL | https://icml.cc/Conferences/2018/Schedule?showEvent=1877 |
| http://proceedings.mlr.press/v80/simsekli18a/simsekli18a.pdf | |
| PWC | https://paperswithcode.com/paper/asynchronous-stochastic-quasi-newton-mcmc-for-1 |
| Repo | |
| Framework | |
