Paper Group ANR 136
On the Limitation of Convolutional Neural Networks in Recognizing Negative Images. Robust Conditional Probabilities. Optimal Densification for Fast and Accurate Minwise Hashing. MORSE: Semantic-ally Drive-n MORpheme SEgment-er. A Gamut-Mapping Framework for Color-Accurate Reproduction of HDR Images. A causal framework for explaining the predictions …
On the Limitation of Convolutional Neural Networks in Recognizing Negative Images
| Title | On the Limitation of Convolutional Neural Networks in Recognizing Negative Images |
| Authors | Hossein Hosseini, Baicen Xiao, Mayoore Jaiswal, Radha Poovendran |
| Abstract | Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance on a variety of computer vision tasks, particularly visual classification problems, where new algorithms reported to achieve or even surpass the human performance. In this paper, we examine whether CNNs are capable of learning the semantics of training data. To this end, we evaluate CNNs on negative images, since they share the same structure and semantics as regular images and humans can classify them correctly. Our experimental results indicate that when training on regular images and testing on negative images, the model accuracy is significantly lower than when it is tested on regular images. This leads us to the conjecture that current training methods do not effectively train models to generalize the concepts. We then introduce the notion of semantic adversarial examples - transformed inputs that semantically represent the same objects, but the model does not classify them correctly - and present negative images as one class of such inputs. |
| Tasks | |
| Published | 2017-03-20 |
| URL | http://arxiv.org/abs/1703.06857v2 |
| http://arxiv.org/pdf/1703.06857v2.pdf | |
| PWC | https://paperswithcode.com/paper/on-the-limitation-of-convolutional-neural |
| Repo | |
| Framework | |
Robust Conditional Probabilities
| Title | Robust Conditional Probabilities |
| Authors | Yoav Wald, Amir Globerson |
| Abstract | Conditional probabilities are a core concept in machine learning. For example, optimal prediction of a label $Y$ given an input $X$ corresponds to maximizing the conditional probability of $Y$ given $X$. A common approach to inference tasks is learning a model of conditional probabilities. However, these models are often based on strong assumptions (e.g., log-linear models), and hence their estimate of conditional probabilities is not robust and is highly dependent on the validity of their assumptions. Here we propose a framework for reasoning about conditional probabilities without assuming anything about the underlying distributions, except knowledge of their second order marginals, which can be estimated from data. We show how this setting leads to guaranteed bounds on conditional probabilities, which can be calculated efficiently in a variety of settings, including structured-prediction. Finally, we apply them to semi-supervised deep learning, obtaining results competitive with variational autoencoders. |
| Tasks | Structured Prediction |
| Published | 2017-08-08 |
| URL | http://arxiv.org/abs/1708.02406v1 |
| http://arxiv.org/pdf/1708.02406v1.pdf | |
| PWC | https://paperswithcode.com/paper/robust-conditional-probabilities |
| Repo | |
| Framework | |
Optimal Densification for Fast and Accurate Minwise Hashing
| Title | Optimal Densification for Fast and Accurate Minwise Hashing |
| Authors | Anshumali Shrivastava |
| Abstract | Minwise hashing is a fundamental and one of the most successful hashing algorithm in the literature. Recent advances based on the idea of densification~\cite{Proc:OneHashLSH_ICML14,Proc:Shrivastava_UAI14} have shown that it is possible to compute $k$ minwise hashes, of a vector with $d$ nonzeros, in mere $(d + k)$ computations, a significant improvement over the classical $O(dk)$. These advances have led to an algorithmic improvement in the query complexity of traditional indexing algorithms based on minwise hashing. Unfortunately, the variance of the current densification techniques is unnecessarily high, which leads to significantly poor accuracy compared to vanilla minwise hashing, especially when the data is sparse. In this paper, we provide a novel densification scheme which relies on carefully tailored 2-universal hashes. We show that the proposed scheme is variance-optimal, and without losing the runtime efficiency, it is significantly more accurate than existing densification techniques. As a result, we obtain a significantly efficient hashing scheme which has the same variance and collision probability as minwise hashing. Experimental evaluations on real sparse and high-dimensional datasets validate our claims. We believe that given the significant advantages, our method will replace minwise hashing implementations in practice. |
| Tasks | |
| Published | 2017-03-14 |
| URL | http://arxiv.org/abs/1703.04664v1 |
| http://arxiv.org/pdf/1703.04664v1.pdf | |
| PWC | https://paperswithcode.com/paper/optimal-densification-for-fast-and-accurate |
| Repo | |
| Framework | |
MORSE: Semantic-ally Drive-n MORpheme SEgment-er
| Title | MORSE: Semantic-ally Drive-n MORpheme SEgment-er |
| Authors | Tarek Sakakini, Suma Bhat, Pramod Viswanath |
| Abstract | We present in this paper a novel framework for morpheme segmentation which uses the morpho-syntactic regularities preserved by word representations, in addition to orthographic features, to segment words into morphemes. This framework is the first to consider vocabulary-wide syntactico-semantic information for this task. We also analyze the deficiencies of available benchmarking datasets and introduce our own dataset that was created on the basis of compositionality. We validate our algorithm across datasets and present state-of-the-art results. |
| Tasks | |
| Published | 2017-02-07 |
| URL | http://arxiv.org/abs/1702.02212v3 |
| http://arxiv.org/pdf/1702.02212v3.pdf | |
| PWC | https://paperswithcode.com/paper/morse-semantic-ally-drive-n-morpheme-segment |
| Repo | |
| Framework | |
A Gamut-Mapping Framework for Color-Accurate Reproduction of HDR Images
| Title | A Gamut-Mapping Framework for Color-Accurate Reproduction of HDR Images |
| Authors | E. Sikudova, T. Pouli, A. Artusi, A. O. Akyuz, F. Banterle, Z. M. Mazlumoglu, E. Reinhard |
| Abstract | Few tone mapping operators (TMOs) take color management into consideration, limiting compression to luminance values only. This may lead to changes in image chroma and hues which are typically managed with a post-processing step. However, current post-processing techniques for tone reproduction do not explicitly consider the target display gamut. Gamut mapping on the other hand, deals with mapping images from one color gamut to another, usually smaller, gamut but has traditionally focused on smaller scale, chromatic changes. In this context, we present a novel gamut and tone management framework for color-accurate reproduction of high dynamic range (HDR) images, which is conceptually and computationally simple, parameter-free, and compatible with existing TMOs. In the CIE LCh color space, we compress chroma to fit the gamut of the output color space. This prevents hue and luminance shifts while taking gamut boundaries into consideration. We also propose a compatible lightness compression scheme that minimizes the number of color space conversions. Our results show that our gamut management method effectively compresses the chroma of tone mapped images, respecting the target gamut and without reducing image quality. |
| Tasks | |
| Published | 2017-11-24 |
| URL | http://arxiv.org/abs/1711.08925v1 |
| http://arxiv.org/pdf/1711.08925v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-gamut-mapping-framework-for-color-accurate |
| Repo | |
| Framework | |
A causal framework for explaining the predictions of black-box sequence-to-sequence models
| Title | A causal framework for explaining the predictions of black-box sequence-to-sequence models |
| Authors | David Alvarez-Melis, Tommi S. Jaakkola |
| Abstract | We interpret the predictions of any black-box structured input-structured output model around a specific input-output pair. Our method returns an “explanation” consisting of groups of input-output tokens that are causally related. These dependencies are inferred by querying the black-box model with perturbed inputs, generating a graph over tokens from the responses, and solving a partitioning problem to select the most relevant components. We focus the general approach on sequence-to-sequence problems, adopting a variational autoencoder to yield meaningful input perturbations. We test our method across several NLP sequence generation tasks. |
| Tasks | |
| Published | 2017-07-06 |
| URL | http://arxiv.org/abs/1707.01943v3 |
| http://arxiv.org/pdf/1707.01943v3.pdf | |
| PWC | https://paperswithcode.com/paper/a-causal-framework-for-explaining-the |
| Repo | |
| Framework | |
Consistency of Dirichlet Partitions
| Title | Consistency of Dirichlet Partitions |
| Authors | Braxton Osting, Todd Harry Reeb |
| Abstract | A Dirichlet $k$-partition of a domain $U \subseteq \mathbb{R}^d$ is a collection of $k$ pairwise disjoint open subsets such that the sum of their first Laplace-Dirichlet eigenvalues is minimal. A discrete version of Dirichlet partitions has been posed on graphs with applications in data analysis. Both versions admit variational formulations: solutions are characterized by minimizers of the Dirichlet energy of mappings from $U$ into a singular space $\Sigma_k \subseteq \mathbb{R}^k$. In this paper, we extend results of N.\ Garc'ia Trillos and D.\ Slep\v{c}ev to show that there exist solutions of the continuum problem arising as limits to solutions of a sequence of discrete problems. Specifically, a sequence of points ${x_i}{i \in \mathbb{N}}$ from $U$ is sampled i.i.d.\ with respect to a given probability measure $\nu$ on $U$ and for all $n \in \mathbb{N}$, a geometric graph $G_n$ is constructed from the first $n$ points $x_1, x_2, \ldots, x_n$ and the pairwise distances between the points. With probability one with respect to the choice of points ${x_i}{i \in \mathbb{N}}$, we show that as $n \to \infty$ the discrete Dirichlet energies for functions $G_n \to \Sigma_k$ $\Gamma$-converge to (a scalar multiple of) the continuum Dirichlet energy for functions $U \to \Sigma_k$ with respect to a metric coming from the theory of optimal transport. This, along with a compactness property for the aforementioned energies that we prove, implies the convergence of minimizers. When $\nu$ is the uniform distribution, our results also imply the statistical consistency statement that Dirichlet partitions of geometric graphs converge to partitions of the sampled space in the Hausdorff sense. |
| Tasks | |
| Published | 2017-08-18 |
| URL | http://arxiv.org/abs/1708.05472v1 |
| http://arxiv.org/pdf/1708.05472v1.pdf | |
| PWC | https://paperswithcode.com/paper/consistency-of-dirichlet-partitions |
| Repo | |
| Framework | |
On Nearest Neighbors in Non Local Means Denoising
| Title | On Nearest Neighbors in Non Local Means Denoising |
| Authors | Iuri Frosio, Jan Kautz |
| Abstract | To denoise a reference patch, the Non-Local-Means denoising filter processes a set of neighbor patches. Few Nearest Neighbors (NN) are used to limit the computational burden of the algorithm. Here here we show analytically that the NN approach introduces a bias in the denoised patch, and we propose a different neighbors’ collection criterion, named Statistical NN (SNN), to alleviate this issue. Our approach outperforms the traditional one in case of both white and colored noise: fewer SNNs generate images of higher quality, at a lower computational cost. |
| Tasks | Denoising |
| Published | 2017-11-20 |
| URL | http://arxiv.org/abs/1711.07568v1 |
| http://arxiv.org/pdf/1711.07568v1.pdf | |
| PWC | https://paperswithcode.com/paper/on-nearest-neighbors-in-non-local-means |
| Repo | |
| Framework | |
An Aposteriorical Clusterability Criterion for $k$-Means++ and Simplicity of Clustering
| Title | An Aposteriorical Clusterability Criterion for $k$-Means++ and Simplicity of Clustering |
| Authors | Mieczysław A. Kłopotek |
| Abstract | We define the notion of a well-clusterable data set combining the point of view of the objective of $k$-means clustering algorithm (minimising the centric spread of data elements) and common sense (clusters shall be separated by gaps). We identify conditions under which the optimum of $k$-means objective coincides with a clustering under which the data is separated by predefined gaps. We investigate two cases: when the whole clusters are separated by some gap and when only the cores of the clusters meet some separation condition. We overcome a major obstacle in using clusterability criteria due to the fact that known approaches to clusterability checking had the disadvantage that they are related to the optimal clustering which is NP hard to identify. Compared to other approaches to clusterability, the novelty consists in the possibility of an a posteriori (after running $k$-means) check if the data set is well-clusterable or not. As the $k$-means algorithm applied for this purpose has polynomial complexity so does therefore the appropriate check. Additionally, if $k$-means++ fails to identify a clustering that meets clusterability criteria, with high probability the data is not well-clusterable. |
| Tasks | Common Sense Reasoning |
| Published | 2017-04-24 |
| URL | http://arxiv.org/abs/1704.07139v2 |
| http://arxiv.org/pdf/1704.07139v2.pdf | |
| PWC | https://paperswithcode.com/paper/an-aposteriorical-clusterability-criterion |
| Repo | |
| Framework | |
End-to-End Multi-View Networks for Text Classification
| Title | End-to-End Multi-View Networks for Text Classification |
| Authors | Hongyu Guo, Colin Cherry, Jiang Su |
| Abstract | We propose a multi-view network for text classification. Our method automatically creates various views of its input text, each taking the form of soft attention weights that distribute the classifier’s focus among a set of base features. For a bag-of-words representation, each view focuses on a different subset of the text’s words. Aggregating many such views results in a more discriminative and robust representation. Through a novel architecture that both stacks and concatenates views, we produce a network that emphasizes both depth and width, allowing training to converge quickly. Using our multi-view architecture, we establish new state-of-the-art accuracies on two benchmark tasks. |
| Tasks | Text Classification |
| Published | 2017-04-19 |
| URL | http://arxiv.org/abs/1704.05907v1 |
| http://arxiv.org/pdf/1704.05907v1.pdf | |
| PWC | https://paperswithcode.com/paper/end-to-end-multi-view-networks-for-text |
| Repo | |
| Framework | |
Ensemble Distillation for Neural Machine Translation
| Title | Ensemble Distillation for Neural Machine Translation |
| Authors | Markus Freitag, Yaser Al-Onaizan, Baskaran Sankaran |
| Abstract | Knowledge distillation describes a method for training a student network to perform better by learning from a stronger teacher network. Translating a sentence with an Neural Machine Translation (NMT) engine is time expensive and having a smaller model speeds up this process. We demonstrate how to transfer the translation quality of an ensemble and an oracle BLEU teacher network into a single NMT system. Further, we present translation improvements from a teacher network that has the same architecture and dimensions of the student network. As the training of the student model is still expensive, we introduce a data filtering method based on the knowledge of the teacher model that not only speeds up the training, but also leads to better translation quality. Our techniques need no code change and can be easily reproduced with any NMT architecture to speed up the decoding process. |
| Tasks | Machine Translation |
| Published | 2017-02-06 |
| URL | http://arxiv.org/abs/1702.01802v2 |
| http://arxiv.org/pdf/1702.01802v2.pdf | |
| PWC | https://paperswithcode.com/paper/ensemble-distillation-for-neural-machine |
| Repo | |
| Framework | |
Simple Classification using Binary Data
| Title | Simple Classification using Binary Data |
| Authors | Deanna Needell, Rayan Saab, Tina Woolf |
| Abstract | Binary, or one-bit, representations of data arise naturally in many applications, and are appealing in both hardware implementations and algorithm design. In this work, we study the problem of data classification from binary data and propose a framework with low computation and resource costs. We illustrate the utility of the proposed approach through stylized and realistic numerical experiments, and provide a theoretical analysis for a simple case. We hope that our framework and analysis will serve as a foundation for studying similar types of approaches. |
| Tasks | |
| Published | 2017-07-06 |
| URL | http://arxiv.org/abs/1707.01945v1 |
| http://arxiv.org/pdf/1707.01945v1.pdf | |
| PWC | https://paperswithcode.com/paper/simple-classification-using-binary-data |
| Repo | |
| Framework | |
Semi-Supervised Learning with IPM-based GANs: an Empirical Study
| Title | Semi-Supervised Learning with IPM-based GANs: an Empirical Study |
| Authors | Tom Sercu, Youssef Mroueh |
| Abstract | We present an empirical investigation of a recent class of Generative Adversarial Networks (GANs) using Integral Probability Metrics (IPM) and their performance for semi-supervised learning. IPM-based GANs like Wasserstein GAN, Fisher GAN and Sobolev GAN have desirable properties in terms of theoretical understanding, training stability, and a meaningful loss. In this work we investigate how the design of the critic (or discriminator) influences the performance in semi-supervised learning. We distill three key take-aways which are important for good SSL performance: (1) the K+1 formulation, (2) avoiding batch normalization in the critic and (3) avoiding gradient penalty constraints on the classification layer. |
| Tasks | |
| Published | 2017-12-07 |
| URL | http://arxiv.org/abs/1712.02505v1 |
| http://arxiv.org/pdf/1712.02505v1.pdf | |
| PWC | https://paperswithcode.com/paper/semi-supervised-learning-with-ipm-based-gans |
| Repo | |
| Framework | |
What matters in a transferable neural network model for relation classification in the biomedical domain?
| Title | What matters in a transferable neural network model for relation classification in the biomedical domain? |
| Authors | Sunil Kumar Sahu, Ashish Anand |
| Abstract | Lack of sufficient labeled data often limits the applicability of advanced machine learning algorithms to real life problems. However efficient use of Transfer Learning (TL) has been shown to be very useful across domains. TL utilizes valuable knowledge learned in one task (source task), where sufficient data is available, to the task of interest (target task). In biomedical and clinical domain, it is quite common that lack of sufficient training data do not allow to fully exploit machine learning models. In this work, we present two unified recurrent neural models leading to three transfer learning frameworks for relation classification tasks. We systematically investigate effectiveness of the proposed frameworks in transferring the knowledge under multiple aspects related to source and target tasks, such as, similarity or relatedness between source and target tasks, and size of training data for source task. Our empirical results show that the proposed frameworks in general improve the model performance, however these improvements do depend on aspects related to source and target tasks. This dependence then finally determine the choice of a particular TL framework. |
| Tasks | Relation Classification, Transfer Learning |
| Published | 2017-08-11 |
| URL | http://arxiv.org/abs/1708.03446v2 |
| http://arxiv.org/pdf/1708.03446v2.pdf | |
| PWC | https://paperswithcode.com/paper/what-matters-in-a-transferable-neural-network |
| Repo | |
| Framework | |
No, This is not a Circle
| Title | No, This is not a Circle |
| Authors | Zoltán Kovács |
| Abstract | A popular curve shown in introductory maths textbooks, seems like a circle. But it is actually a different curve. This paper discusses some elementary approaches to identify the geometric object, including novel technological means by using GeoGebra. We demonstrate two ways to refute the false impression, two suggestions to find a correct conjecture, and four ways to confirm the result by proving it rigorously. All of the discussed approaches can be introduced in classrooms at various levels from middle school to high school. |
| Tasks | |
| Published | 2017-04-27 |
| URL | http://arxiv.org/abs/1704.08483v3 |
| http://arxiv.org/pdf/1704.08483v3.pdf | |
| PWC | https://paperswithcode.com/paper/no-this-is-not-a-circle |
| Repo | |
| Framework | |
