Paper Group ANR 1448
End to End Recognition System for Recognizing Offline Unconstrained Vietnamese Handwriting. Entity Summarization: State of the Art and Future Challenges. Low-Resource Parsing with Crosslingual Contextualized Representations. IRIS: Implicit Reinforcement without Interaction at Scale for Learning Control from Offline Robot Manipulation Data. Learning …
End to End Recognition System for Recognizing Offline Unconstrained Vietnamese Handwriting
| Title | End to End Recognition System for Recognizing Offline Unconstrained Vietnamese Handwriting |
| Authors | Anh Duc Le, Hung Tuan Nguyen, Masaki Nakagawa |
| Abstract | Inspired by recent successes in neural machine translation and image caption generation, we present an attention based encoder decoder model (AED) to recognize Vietnamese Handwritten Text. The model composes of two parts: a DenseNet for extracting invariant features, and a Long Short-Term Memory network (LSTM) with an attention model incorporated for generating output text (LSTM decoder), which are connected from the CNN part to the attention model. The input of the CNN part is a handwritten text image and the target of the LSTM decoder is the corresponding text of the input image. Our model is trained end-to-end to predict the text from a given input image since all the parts are differential components. In the experiment section, we evaluate our proposed AED model on the VNOnDB-Word and VNOnDB-Line datasets to verify its efficiency. The experiential results show that our model achieves 12.30% of word error rate without using any language model. This result is competitive with the handwriting recognition system provided by Google in the Vietnamese Online Handwritten Text Recognition competition. |
| Tasks | Language Modelling, Machine Translation |
| Published | 2019-05-14 |
| URL | https://arxiv.org/abs/1905.05381v1 |
| https://arxiv.org/pdf/1905.05381v1.pdf | |
| PWC | https://paperswithcode.com/paper/end-to-end-recognition-system-for-recognizing |
| Repo | |
| Framework | |
Entity Summarization: State of the Art and Future Challenges
| Title | Entity Summarization: State of the Art and Future Challenges |
| Authors | Qingxia Liu, Gong Cheng, Kalpa Gunaratna, Yuzhong Qu |
| Abstract | The increasing availability of semantic data, which is commonly represented as entity-property-value triples, has enabled novel information retrieval applications. However, the magnitude of semantic data, in particular the large number of triples describing an entity, could overload users with excessive amounts of information. This has motivated fruitful research on automated generation of summaries for entity descriptions to satisfy users’ information needs efficiently and effectively. We focus on this important topic of entity summarization, and present the first comprehensive survey of existing research. We review existing methods and evaluation efforts, and suggest directions for future work. |
| Tasks | Information Retrieval |
| Published | 2019-10-18 |
| URL | https://arxiv.org/abs/1910.08252v1 |
| https://arxiv.org/pdf/1910.08252v1.pdf | |
| PWC | https://paperswithcode.com/paper/entity-summarization-state-of-the-art-and |
| Repo | |
| Framework | |
Low-Resource Parsing with Crosslingual Contextualized Representations
| Title | Low-Resource Parsing with Crosslingual Contextualized Representations |
| Authors | Phoebe Mulcaire, Jungo Kasai, Noah A. Smith |
| Abstract | Despite advances in dependency parsing, languages with small treebanks still present challenges. We assess recent approaches to multilingual contextual word representations (CWRs), and compare them for crosslingual transfer from a language with a large treebank to a language with a small or nonexistent treebank, by sharing parameters between languages in the parser itself. We experiment with a diverse selection of languages in both simulated and truly low-resource scenarios, and show that multilingual CWRs greatly facilitate low-resource dependency parsing even without crosslingual supervision such as dictionaries or parallel text. Furthermore, we examine the non-contextual part of the learned language models (which we call a “decontextual probe”) to demonstrate that polyglot language models better encode crosslingual lexical correspondence compared to aligned monolingual language models. This analysis provides further evidence that polyglot training is an effective approach to crosslingual transfer. |
| Tasks | Dependency Parsing |
| Published | 2019-09-19 |
| URL | https://arxiv.org/abs/1909.08744v1 |
| https://arxiv.org/pdf/1909.08744v1.pdf | |
| PWC | https://paperswithcode.com/paper/low-resource-parsing-with-crosslingual |
| Repo | |
| Framework | |
IRIS: Implicit Reinforcement without Interaction at Scale for Learning Control from Offline Robot Manipulation Data
| Title | IRIS: Implicit Reinforcement without Interaction at Scale for Learning Control from Offline Robot Manipulation Data |
| Authors | Ajay Mandlekar, Fabio Ramos, Byron Boots, Silvio Savarese, Li Fei-Fei, Animesh Garg, Dieter Fox |
| Abstract | Learning from offline task demonstrations is a problem of great interest in robotics. For simple short-horizon manipulation tasks with modest variation in task instances, offline learning from a small set of demonstrations can produce controllers that successfully solve the task. However, leveraging a fixed batch of data can be problematic for larger datasets and longer-horizon tasks with greater variations. The data can exhibit substantial diversity and consist of suboptimal solution approaches. In this paper, we propose Implicit Reinforcement without Interaction at Scale (IRIS), a novel framework for learning from large-scale demonstration datasets. IRIS factorizes the control problem into a goal-conditioned low-level controller that imitates short demonstration sequences and a high-level goal selection mechanism that sets goals for the low-level and selectively combines parts of suboptimal solutions leading to more successful task completions. We evaluate IRIS across three datasets, including the RoboTurk Cans dataset collected by humans via crowdsourcing, and show that performant policies can be learned from purely offline learning. Additional results at https://sites.google.com/stanford.edu/iris/ . |
| Tasks | |
| Published | 2019-11-13 |
| URL | https://arxiv.org/abs/1911.05321v2 |
| https://arxiv.org/pdf/1911.05321v2.pdf | |
| PWC | https://paperswithcode.com/paper/iris-implicit-reinforcement-without |
| Repo | |
| Framework | |
Learning to Optimize with Unsupervised Learning: Training Deep Neural Networks for URLLC
| Title | Learning to Optimize with Unsupervised Learning: Training Deep Neural Networks for URLLC |
| Authors | Chengjian Sun, Chenyang Yang |
| Abstract | Learning the optimized solution as a function of environmental parameters is effective in solving numerical optimization in real time for time-sensitive applications. Existing works of learning to optimize train deep neural networks (DNN) with labels, and the learnt solution are inaccurate, which cannot be employed to ensure the stringent quality of service. In this paper, we propose a framework to learn the latent function with unsupervised deep learning, where the property that the optimal solution should satisfy is used as the “supervision signal” implicitly. The framework is applicable to both functional and variable optimization problems with constraints. We take a variable optimization problem in ultra-reliable and low-latency communications as an example, which demonstrates that the ultra-high reliability can be supported by the DNN without supervision labels. |
| Tasks | |
| Published | 2019-05-27 |
| URL | https://arxiv.org/abs/1905.11017v1 |
| https://arxiv.org/pdf/1905.11017v1.pdf | |
| PWC | https://paperswithcode.com/paper/learning-to-optimize-with-unsupervised |
| Repo | |
| Framework | |
PgNN: Physics-guided Neural Network for Fourier Ptychographic Microscopy
| Title | PgNN: Physics-guided Neural Network for Fourier Ptychographic Microscopy |
| Authors | Yongbing Zhang, Yangzhe Liu, Xiu Li, Shaowei Jiang, Krishna Dixit, Xinfeng Zhang, Xiangyang Ji |
| Abstract | Fourier ptychography (FP) is a newly developed computational imaging approach that achieves both high resolution and wide field of view by stitching a series of low-resolution images captured under angle-varied illumination. So far, many supervised data-driven models have been applied to solve inverse imaging problems. These models need massive amounts of data to train, and are limited by the dataset characteristics. In FP problems, generic datasets are always scarce, and the optical aberration varies greatly under different acquisition conditions. To address these dilemmas, we model the forward physical imaging process as an interpretable physics-guided neural network (PgNN), where the reconstructed image in the complex domain is considered as the learnable parameters of the neural network. Since the optimal parameters of the PgNN can be derived by minimizing the difference between the model-generated images and real captured angle-varied images corresponding to the same scene, the proposed PgNN can get rid of the problem of massive training data as in traditional supervised methods. Applying the alternate updating mechanism and the total variation regularization, PgNN can flexibly reconstruct images with improved performance. In addition, the Zernike mode is incorporated to compensate for optical aberrations to enhance the robustness of FP reconstructions. As a demonstration, we show our method can reconstruct images with smooth performance and detailed information in both simulated and experimental datasets. In particular, when validated in an extension of a high-defocus, high-exposure tissue section dataset, PgNN outperforms traditional FP methods with fewer artifacts and distinguishable structures. |
| Tasks | |
| Published | 2019-09-19 |
| URL | https://arxiv.org/abs/1909.08869v1 |
| https://arxiv.org/pdf/1909.08869v1.pdf | |
| PWC | https://paperswithcode.com/paper/pgnn-physics-guided-neural-network-for |
| Repo | |
| Framework | |
Improving fraud prediction with incremental data balancing technique for massive data streams
| Title | Improving fraud prediction with incremental data balancing technique for massive data streams |
| Authors | Rafiq Ahmed Mohammed, Kok-Wai Wong, Mohd Fairuz Shiratuddin, Xuequn Wang |
| Abstract | The performance of classification algorithms with a massive and highly imbalanced data stream depends upon efficient balancing strategy. Some techniques of balancing strategy have been applied in the past with Batch data to resolve the class imbalance problem. This paper proposes a new incremental data balancing framework which can work with massive imbalanced data streams. In this paper, we choose Racing Algorithm as an automated data balancing technique which optimizes the balancing techniques. We applied Random Forest classification algorithm which can deal with the massive data stream. We investigated the suitability of Racing Algorithm and Random Forest in the proposed framework. Applying new technique in the proposed framework on the European Credit Card dataset, provided better results than the Batch mode. The proposed framework is more scalable to handle online massive data streams. |
| Tasks | |
| Published | 2019-02-28 |
| URL | https://arxiv.org/abs/1903.00410v2 |
| https://arxiv.org/pdf/1903.00410v2.pdf | |
| PWC | https://paperswithcode.com/paper/improving-fraud-prediction-with-incremental |
| Repo | |
| Framework | |
An Iterative Scientific Machine Learning Approach for Discovery of Theories Underlying Physical Phenomena
| Title | An Iterative Scientific Machine Learning Approach for Discovery of Theories Underlying Physical Phenomena |
| Authors | Navid Zobeiry, Keith D. Humfeld |
| Abstract | Form a pure mathematical point of view, common functional forms representing different physical phenomena can be defined. For example, rates of chemical reactions, diffusion and heat transfer are all governed by exponential-type expressions. If machine learning is used for physical problems, inferred from domain knowledge, original features can be transformed in such a way that the end expressions are highly aligned and correlated with the underlying physics. This should significantly reduce the training effort in terms of iterations, architecture and the number of required data points. We extend this by approaching a problem from an agnostic position and propose a systematic and iterative methodology to discover theories underlying physical phenomena. At first, commonly observed functional forms of theoretical expressions are used to transform original features before conducting correlation analysis to output. Using random combinations of highly correlated expressions, training of Neural Networks (NN) are performed. By comparing the rates of convergence or mean error in training, expressions describing the underlying physical problems can be discovered, leading to extracting explicit analytic equations. This approach was used in three blind demonstrations for different physical phenomena. |
| Tasks | |
| Published | 2019-09-24 |
| URL | https://arxiv.org/abs/1909.13718v1 |
| https://arxiv.org/pdf/1909.13718v1.pdf | |
| PWC | https://paperswithcode.com/paper/an-iterative-scientific-machine-learning |
| Repo | |
| Framework | |
Casting Geometric Constraints in Semantic Segmentation as Semi-Supervised Learning
| Title | Casting Geometric Constraints in Semantic Segmentation as Semi-Supervised Learning |
| Authors | Sinisa Stekovic, Friedrich Fraundorfer, Vincent Lepetit |
| Abstract | We propose a simple yet effective method to learn to segment new indoor scenes from video frames: State-of-the-art methods trained on one dataset, even as large as the SUNRGB-D dataset, can perform poorly when applied to images that are not part of the dataset, because of the dataset bias, a common phenomenon in computer vision. To make semantic segmentation more useful in practice, one can exploit geometric constraints. Our main contribution is to show that these constraints can be cast conveniently as semi-supervised terms, which enforce the fact that the same class should be predicted for the projections of the same 3D location in different images. This is interesting as we can exploit general existing techniques developed for semi-supervised learning to efficiently incorporate the constraints. We show that this approach can efficiently and accurately learn to segment target sequences of ScanNet and our own target sequences using only annotations from SUNRGB-D, and geometric relations between the video frames of target sequences. |
| Tasks | Semantic Segmentation |
| Published | 2019-04-29 |
| URL | https://arxiv.org/abs/1904.12534v3 |
| https://arxiv.org/pdf/1904.12534v3.pdf | |
| PWC | https://paperswithcode.com/paper/casting-geometric-constraints-in-semantic |
| Repo | |
| Framework | |
Attending Form and Context to Generate Specialized Out-of-VocabularyWords Representations
| Title | Attending Form and Context to Generate Specialized Out-of-VocabularyWords Representations |
| Authors | Nicolas Garneau, Jean-Samuel Leboeuf, Yuval Pinter, Luc Lamontagne |
| Abstract | We propose a new contextual-compositional neural network layer that handles out-of-vocabulary (OOV) words in natural language processing (NLP) tagging tasks. This layer consists of a model that attends to both the character sequence and the context in which the OOV words appear. We show that our model learns to generate task-specific \textit{and} sentence-dependent OOV word representations without the need for pre-training on an embedding table, unlike previous attempts. We insert our layer in the state-of-the-art tagging model of \citet{plank2016multilingual} and thoroughly evaluate its contribution on 23 different languages on the task of jointly tagging part-of-speech and morphosyntactic attributes. Our OOV handling method successfully improves performances of this model on every language but one to achieve a new state-of-the-art on the Universal Dependencies Dataset 1.4. |
| Tasks | |
| Published | 2019-12-14 |
| URL | https://arxiv.org/abs/1912.06876v1 |
| https://arxiv.org/pdf/1912.06876v1.pdf | |
| PWC | https://paperswithcode.com/paper/attending-form-and-context-to-generate |
| Repo | |
| Framework | |
Exact inference in structured prediction
| Title | Exact inference in structured prediction |
| Authors | Kevin Bello, Jean Honorio |
| Abstract | Structured prediction can be thought of as a simultaneous prediction of multiple labels. This is often done by maximizing a score function on the space of labels, which decomposes as a sum of pairwise and unary potentials. The above is naturally modeled with a graph, where edges and vertices are related to pairwise and unary potentials, respectively. We consider the generative process proposed by Globerson et al. and apply it to general connected graphs. We analyze the structural conditions of the graph that allow for the exact recovery of the labels. Our results show that exact recovery is possible and achievable in polynomial time for a large class of graphs. In particular, we show that graphs that are bad expanders can be exactly recovered by adding small edge perturbations coming from the Erd\H{o}s-R'enyi model. Finally, as a byproduct of our analysis, we provide an extension of Cheeger’s inequality. |
| Tasks | Structured Prediction |
| Published | 2019-06-02 |
| URL | https://arxiv.org/abs/1906.00451v1 |
| https://arxiv.org/pdf/1906.00451v1.pdf | |
| PWC | https://paperswithcode.com/paper/190600451 |
| Repo | |
| Framework | |
Deep Dexterous Grasping of Novel Objects from a Single View
| Title | Deep Dexterous Grasping of Novel Objects from a Single View |
| Authors | Umit Rusen Aktas, Chao Zhao, Marek Kopicki, Ales Leonardis, Jeremy L. Wyatt |
| Abstract | Dexterous grasping of a novel object given a single view is an open problem. This paper makes several contributions to its solution. First, we present a simulator for generating and testing dexterous grasps. Second we present a data set, generated by this simulator, of 2.4 million simulated dexterous grasps of variations of 294 base objects drawn from 20 categories. Third, we present a basic architecture for generation and evaluation of dexterous grasps that may be trained in a supervised manner. Fourth, we present three different evaluative architectures, employing ResNet-50 or VGG16 as their visual backbone. Fifth, we train, and evaluate seventeen variants of generative-evaluative architectures on this simulated data set, showing improvement from 69.53% grasp success rate to 90.49%. Finally, we present a real robot implementation and evaluate the four most promising variants, executing 196 real robot grasps in total. We show that our best architectural variant achieves a grasp success rate of 87.8% on real novel objects seen from a single view, improving on a baseline of 57.1%. |
| Tasks | |
| Published | 2019-08-10 |
| URL | https://arxiv.org/abs/1908.04293v1 |
| https://arxiv.org/pdf/1908.04293v1.pdf | |
| PWC | https://paperswithcode.com/paper/deep-dexterous-grasping-of-novel-objects-from |
| Repo | |
| Framework | |
Uniform concentration and symmetrization for weak interactions
| Title | Uniform concentration and symmetrization for weak interactions |
| Authors | Andreas Maurer, Massimiliano Pontil |
| Abstract | The method to derive uniform bounds with Gaussian and Rademacher complexities is extended to the case where the sample average is replaced by a nonlinear statistic. Tight bounds are obtained for U-statistics, smoothened L-statistics and error functionals of l2-regularized algorithms. |
| Tasks | |
| Published | 2019-02-05 |
| URL | https://arxiv.org/abs/1902.01911v4 |
| https://arxiv.org/pdf/1902.01911v4.pdf | |
| PWC | https://paperswithcode.com/paper/uniform-concentration-and-symmetrization-for |
| Repo | |
| Framework | |
Uncertainty Analysis of VAE-GANs for Compressive Medical Imaging
| Title | Uncertainty Analysis of VAE-GANs for Compressive Medical Imaging |
| Authors | Vineet Edupuganti, Morteza Mardani, Joseph Cheng, Shreyas Vasanawala, John Pauly |
| Abstract | Reliable medical image recovery is crucial for accurate diagnoses and patient wellbeing. However, high resolution imaging from limited sensory data leaves substantial uncertainty about the authenticity of the recovered pixels. This study aims to quantify this uncertainty so as to guide radiologists about the confidence of their diagnoses. We put forth a probabilistic recovery scheme based on VAE-GANs, comprised of a VAE generator and multi-layer CNN discriminator, that maps out low-quality images with aliasing artifacts to diagnostic-quality ones. We leverage Stein’s Unbiased Risk Estimator (SURE) as a proxy for the prediction error, which includes a divergence term (trace of the end-to-end network Jacobian) quantifying the estimation uncertainty. Extensive empirical experiments are performed for the task of magnetic resonance (MR) image recovery using a dataset of pediatric Knee images. We statistically analyze the output distribution of the model using Monte Carlo sampling to gauge the extent of variance, bias, and error across reconstructions. The results indicate that uncertainty level is significantly influenced by the hyperparameter setting and network architecture. The key observations are that the pixel uncertainty level: 1) increases as the GAN loss weight rises; and 2) decreases as we add more recurrent units to the generator network (cascade of VAEs and data consistency layers). |
| Tasks | |
| Published | 2019-01-31 |
| URL | https://arxiv.org/abs/1901.11228v2 |
| https://arxiv.org/pdf/1901.11228v2.pdf | |
| PWC | https://paperswithcode.com/paper/vae-gans-for-probabilistic-compressive-image |
| Repo | |
| Framework | |
Large-Scale Markov Decision Problems via the Linear Programming Dual
| Title | Large-Scale Markov Decision Problems via the Linear Programming Dual |
| Authors | Yasin Abbasi-Yadkori, Peter L. Bartlett, Xi Chen, Alan Malek |
| Abstract | We consider the problem of controlling a fully specified Markov decision process (MDP), also known as the planning problem, when the state space is very large and calculating the optimal policy is intractable. Instead, we pursue the more modest goal of optimizing over some small family of policies. Specifically, we show that the family of policies associated with a low-dimensional approximation of occupancy measures yields a tractable optimization. Moreover, we propose an efficient algorithm, scaling with the size of the subspace but not the state space, that is able to find a policy with low excess loss relative to the best policy in this class. To the best of our knowledge, such results did not exist in the literature previously. We bound excess loss in the average cost and discounted cost cases, which are treated separately. Preliminary experiments show the effectiveness of the proposed algorithms in a queueing application. |
| Tasks | |
| Published | 2019-01-06 |
| URL | http://arxiv.org/abs/1901.01992v1 |
| http://arxiv.org/pdf/1901.01992v1.pdf | |
| PWC | https://paperswithcode.com/paper/large-scale-markov-decision-problems-via-the |
| Repo | |
| Framework | |
