January 27, 2020

3566 words 17 mins read

Paper Group ANR 1163

Paper Group ANR 1163

Online Learning and Optimization Under a New Linear-Threshold Model with Negative Influence. Calibrated Surrogate Maximization of Linear-fractional Utility in Binary Classification. Constrained Bilinear Factorization Multi-view Subspace Clustering. Are Quantitative Features of Lung Nodules Reproducible at Different CT Acquisition and Reconstruction …

Online Learning and Optimization Under a New Linear-Threshold Model with Negative Influence

Title Online Learning and Optimization Under a New Linear-Threshold Model with Negative Influence
Authors Shuoguang Yang, Shatian Wang, Van-Anh Truong
Abstract We propose a new class of Linear Threshold Model-based information-diffusion model that incorporates the formation and spread of negative attitude. We call such models negativity-aware.. We show that in these models, the influence function is a monotone submodular function. Thus we can use the greedy algorithm to construct seed sets with constant approximation guarantees, when the objective is to select a seed set of fixed size $K$ to maximize total influence. Our models are flexible enough to account for both the features of local users and the features of the information being propagated in the diffusion. We analyze an online-learning setting for a multi-round influence-maximization problem, where an agent is actively learning the diffusion parameters over time while trying to maximize total cumulative positive influence. We assume that in each diffusion step, the agent can only observe whether a node becomes positively or negatively influenced, or remains inactive. In particular, he does not observe the particular edge that brought about the activation of a node, if any, as in the case of most models that assume Independent Cascade (IC)-based diffusions. This model of feedback is called node-level feedback, as opposed to the more common \emph{edge-level feedback} model in which he is able to observe, for each node, the edge through which that node is influenced. Under mild assumptions, we develop online learning algorithms that achieve cumulative expected regrets of order $O(T^{-c})$ for any $c<1$ where $T$ is the total number of rounds. These are the first regret guarantees for node-level feedback models of influence maximization of any kind. Furthermore, with mild assumptions, this result also improves the average regret of $O(\sqrt{\ln T / T})$ for the edge-level feedback models, thus providing a new performance benchmark.
Tasks
Published 2019-11-08
URL https://arxiv.org/abs/1911.03276v2
PDF https://arxiv.org/pdf/1911.03276v2.pdf
PWC https://paperswithcode.com/paper/online-learning-and-optimization-under-a-new
Repo
Framework

Calibrated Surrogate Maximization of Linear-fractional Utility in Binary Classification

Title Calibrated Surrogate Maximization of Linear-fractional Utility in Binary Classification
Authors Han Bao, Masashi Sugiyama
Abstract Complex classification performance metrics such as the F${}\beta$-measure and Jaccard index are often used, in order to handle class-imbalanced cases such as information retrieval and image segmentation. These performance metrics are not decomposable, that is, they cannot be expressed in a per-example manner, which hinders a straightforward application of M-estimation widely used in supervised learning. In this paper, we consider linear-fractional metrics, which are a family of classification performance metrics that encompasses many standard ones such as the F${}\beta$-measure and Jaccard index, and propose methods to directly maximize performances under those metrics. A clue to tackle their direct optimization is a calibrated surrogate utility, which is a tractable lower bound of the true utility function representing a given metric. We characterize sufficient conditions which make the surrogate maximization coincide with the maximization of the true utility. Simulation results on benchmark datasets validate the effectiveness of our calibrated surrogate maximization especially if the sample sizes are extremely small.
Tasks Calibration, Information Retrieval, Semantic Segmentation
Published 2019-05-29
URL https://arxiv.org/abs/1905.12511v2
PDF https://arxiv.org/pdf/1905.12511v2.pdf
PWC https://paperswithcode.com/paper/calibrated-surrogate-maximization-of-linear
Repo
Framework

Constrained Bilinear Factorization Multi-view Subspace Clustering

Title Constrained Bilinear Factorization Multi-view Subspace Clustering
Authors Qinghai Zheng, Jihua Zhu, Zhiqiang Tian, Zhongyu Li, Shanmin Pang, Xiuyi Jia
Abstract Multi-view clustering is an important and fundamental problem. Many multi-view subspace clustering methods have been proposed and achieved success in real-world applications, most of which assume that all views share a same coefficient matrix. However, the underlying information of multiview data are not exploited effectively under this assumption, since the coefficient matrices of different views should have the same clustering properties rather than be the same among multiple views. To this end, a novel Constrained Bilinear Factorization Multi-view Subspace Clustering (CBF-MSC) method is proposed in this paper. Specifically, the bilinear factorization with an orthonormality constraint and a low-rank constraint is employed for all coefficient matrices to make all coefficient matrices have the same trace-norm instead of being equivalent, so as to explore the consensus information of multi-view data more effectively. Finally, an algorithm based on the Augmented Lagrangian Multiplier (ALM) scheme with alternating direction minimization is designed to optimize the objective function. Comprehensive experiments tested on six benchmark datasets validate the effectiveness and competitiveness of the proposed approach compared with several state-of-the-art approaches.
Tasks Multi-view Subspace Clustering
Published 2019-06-19
URL https://arxiv.org/abs/1906.08107v1
PDF https://arxiv.org/pdf/1906.08107v1.pdf
PWC https://paperswithcode.com/paper/constrained-bilinear-factorization-multi-view
Repo
Framework

Are Quantitative Features of Lung Nodules Reproducible at Different CT Acquisition and Reconstruction Parameters?

Title Are Quantitative Features of Lung Nodules Reproducible at Different CT Acquisition and Reconstruction Parameters?
Authors Barbaros S. Erdal, Mutlu Demirer, Chiemezie C. Amadi, Gehan F. M. Ibrahim, Thomas P. O’Donnell, Rainer Grimmer, Andreas Wimmer, Kevin J. Little, Vikash Gupta, Matthew T. Bigelow, Luciano M. Prevedello, Richard D. White
Abstract Consistency and duplicability in Computed Tomography (CT) output is essential to quantitative imaging for lung cancer detection and monitoring. This study of CT-detected lung nodules investigated the reproducibility of volume-, density-, and texture-based features (outcome variables) over routine ranges of radiation-dose, reconstruction kernel, and slice thickness. CT raw data of 23 nodules were reconstructed using 320 acquisition/reconstruction conditions (combinations of 4 doses, 10 kernels, and 8 thicknesses). Scans at 12.5%, 25%, and 50% of protocol dose were simulated; reduced-dose and full-dose data were reconstructed using conventional filtered back-projection and iterative-reconstruction kernels at a range of thicknesses (0.6-5.0 mm). Full-dose/B50f kernel reconstructions underwent expert segmentation for reference Region-Of-Interest (ROI) and nodule volume per thickness; each ROI was applied to 40 corresponding images (combinations of 4 doses and 10 kernels). Typical texture analysis metrics (including 5 histogram features, 13 Gray Level Co-occurrence Matrix, 5 Run Length Matrix, 2 Neighboring Gray-Level Dependence Matrix, and 2 Neighborhood Gray-Tone Difference Matrix) were computed per ROI. Reconstruction conditions resulting in no significant change in volume, density, or texture metrics were identified as “compatible pairs” for a given outcome variable. Our results indicate that as thickness increases, volumetric reproducibility decreases, while reproducibility of histogram- and texture-based features across different acquisition and reconstruction parameters improves. In order to achieve concomitant reproducibility of volumetric and radiomic results across studies, balanced standardization of the imaging acquisition parameters is required.
Tasks Computed Tomography (CT), Texture Classification
Published 2019-08-14
URL https://arxiv.org/abs/1908.05667v1
PDF https://arxiv.org/pdf/1908.05667v1.pdf
PWC https://paperswithcode.com/paper/are-quantitative-features-of-lung-nodules
Repo
Framework

Feature Concatenation Multi-view Subspace Clustering

Title Feature Concatenation Multi-view Subspace Clustering
Authors Qinghai Zheng, Jihua Zhu, Zhongyu Li, Shanmin Pang, Jun Wang, Yaochen Li
Abstract Multi-view clustering aims to achieve more promising clustering results than single-view clustering by exploring the multi-view information. Since statistic properties of different views are diverse, even incompatible, few approaches implement multi-view clustering based on the concatenated features directly. However, feature concatenation is a natural way to combine multiple views. To this end, this paper proposes a novel multi-view subspace clustering approach dubbed Feature Concatenation Multi-view Subspace Clustering (FCMSC). Specifically, by exploring the consensus information, multi-view data are concatenated into a joint representation firstly, then, $l_{2,1}$-norm is integrated into the objective function to deal with the sample-specific and cluster-specific corruptions of multiple views for benefiting the clustering performance. Furthermore, by introducing graph Laplacians of multiple views, a graph regularized FCMSC is also introduced to explore both the consensus information and complementary information for clustering. It is noteworthy that the obtained coefficient matrix is not derived by directly applying the Low-Rank Representation (LRR) to the joint view representation simply. Finally, an effective algorithm based on the Augmented Lagrangian Multiplier (ALM) is designed to optimized the objective functions. Comprehensive experiments on six real world datasets illustrate the superiority of the proposed methods over several state-of-the-art approaches for multi-view clustering.
Tasks Multi-view Subspace Clustering
Published 2019-01-30
URL https://arxiv.org/abs/1901.10657v5
PDF https://arxiv.org/pdf/1901.10657v5.pdf
PWC https://paperswithcode.com/paper/feature-concatenation-multi-view-subspace
Repo
Framework

Tractable Minor-free Generalization of Planar Zero-field Ising Models

Title Tractable Minor-free Generalization of Planar Zero-field Ising Models
Authors Valerii Likhosherstov, Yury Maximov, Michael Chertkov
Abstract We present a new family of zero-field Ising models over $N$ binary variables/spins obtained by consecutive “gluing” of planar and $O(1)$-sized components and subsets of at most three vertices into a tree. The polynomial-time algorithm of the dynamic programming type for solving exact inference (computing partition function) and exact sampling (generating i.i.d. samples) consists in a sequential application of an efficient (for planar) or brute-force (for $O(1)$-sized) inference and sampling to the components as a black box. To illustrate the utility of the new family of tractable graphical models, we first build a polynomial algorithm for inference and sampling of zero-field Ising models over $K_{3,3}$-minor-free topologies and over $K_{5}$-minor-free topologies – both are extensions of the planar zero-field Ising models – which are neither genus - nor treewidth-bounded. Second, we demonstrate empirically an improvement in the approximation quality of the NP-hard problem of inference over the square-grid Ising model in a node-dependent non-zero “magnetic” field.
Tasks
Published 2019-10-22
URL https://arxiv.org/abs/1910.11142v1
PDF https://arxiv.org/pdf/1910.11142v1.pdf
PWC https://paperswithcode.com/paper/tractable-minor-free-generalization-of-planar
Repo
Framework

Color Texture Classification Based on Proposed Impulse-Noise Resistant Color Local Binary Patterns and Significant Points Selection Algorithm

Title Color Texture Classification Based on Proposed Impulse-Noise Resistant Color Local Binary Patterns and Significant Points Selection Algorithm
Authors Shervan Fekri-Ershad, Farshad Tajeripour
Abstract The main aim of this paper is to propose a color texture classification approach which uses color sensor information and texture features jointly. High accuracy, low noise sensitivity and low computational complexity are specified aims for our proposed approach. One of the efficient texture analysis operations is local binary patterns. The proposed approach includes two steps. First, a noise resistant version of color local binary patterns is proposed to decrease sensitivity to noise of LBP. This step is evaluated based on combination of color sensor information using AND operation. In second step, a significant points selection algorithm is proposed to select significant LBP. This phase decreases final computational complexity along with increasing accuracy rate. The Proposed approach is evaluated using Vistex, Outex, and KTH TIPS2a data sets. Our approach has been compared with some state of the art methods. It is experimentally demonstrated that the proposed approach achieves highest accuracy. In two other experiments, result show low noise sensitivity and low computational complexity of the proposed approach in comparison with previous versions of LBP. Rotation invariant, multi resolution, general usability are other advantages of our proposed approach. In the present paper, a new version of LBP is proposed originally, which is called Hybrid color local binary patterns. It can be used in many image processing applications to extract color and texture features jointly. Also, a significant point selection algorithm is proposed for the first time to select key points of images.
Tasks Texture Classification
Published 2019-06-26
URL https://arxiv.org/abs/1906.11010v1
PDF https://arxiv.org/pdf/1906.11010v1.pdf
PWC https://paperswithcode.com/paper/color-texture-classification-based-on
Repo
Framework

Electro-optical Neural Networks based on Time-stretch Method

Title Electro-optical Neural Networks based on Time-stretch Method
Authors Yubin Zang, Minghua Chen, Sigang Yang, Hongwei Chen
Abstract In this paper, a novel architecture of electro-optical neural networks based on the time-stretch method is proposed and numerically simulated. By stretching time-domain ultrashort pulses, multiplications of large scale weight matrices and vectors can be implemented on light and multiple-layer of feedforward neural network operations can be easily implemented with fiber loops. Via simulation, the performance of a three-layer electro-optical neural network is tested by the handwriting digit recognition task and the accuracy reaches 88% under considerable noise.
Tasks
Published 2019-09-13
URL https://arxiv.org/abs/1909.07788v1
PDF https://arxiv.org/pdf/1909.07788v1.pdf
PWC https://paperswithcode.com/paper/electro-optical-neural-networks-based-on-time
Repo
Framework

Applications of the Deep Galerkin Method to Solving Partial Integro-Differential and Hamilton-Jacobi-Bellman Equations

Title Applications of the Deep Galerkin Method to Solving Partial Integro-Differential and Hamilton-Jacobi-Bellman Equations
Authors Ali Al-Aradi, Adolfo Correia, Danilo de Frietas Naiff, Gabriel Jardim, Yuri Saporito
Abstract We extend the Deep Galerkin Method (DGM) introduced in Sirignano and Spiliopoulos (2018) to solve a number of partial differential equations (PDEs) that arise in the context of optimal stochastic control and mean field games. First, we consider PDEs where the function is constrained to be positive and integrate to unity, as is the case with Fokker-Planck equations. Our approach involves reparameterizing the solution as the exponential of a neural network appropriately normalized to ensure both requirements are satisfied. This then gives rise to a partial integro-differential equation (PIDE) where the integral appearing in the equation is handled using importance sampling. Secondly, we tackle a number of Hamilton-Jacobi-Bellman (HJB) equations that appear in stochastic optimal control problems. The key contribution is that these equations are approached in their unsimplified primal form which includes an optimization problem as part of the equation. We extend the DGM algorithm to solve for the value function and the optimal control simultaneously by characterizing both as deep neural networks. Training the networks is performed by taking alternating stochastic gradient descent steps for the two functions, a technique similar in spirit to policy improvement algorithms.
Tasks
Published 2019-11-30
URL https://arxiv.org/abs/1912.01455v2
PDF https://arxiv.org/pdf/1912.01455v2.pdf
PWC https://paperswithcode.com/paper/applications-of-the-deep-galerkin-method-to
Repo
Framework

Insertion Transformer: Flexible Sequence Generation via Insertion Operations

Title Insertion Transformer: Flexible Sequence Generation via Insertion Operations
Authors Mitchell Stern, William Chan, Jamie Kiros, Jakob Uszkoreit
Abstract We present the Insertion Transformer, an iterative, partially autoregressive model for sequence generation based on insertion operations. Unlike typical autoregressive models which rely on a fixed, often left-to-right ordering of the output, our approach accommodates arbitrary orderings by allowing for tokens to be inserted anywhere in the sequence during decoding. This flexibility confers a number of advantages: for instance, not only can our model be trained to follow specific orderings such as left-to-right generation or a binary tree traversal, but it can also be trained to maximize entropy over all valid insertions for robustness. In addition, our model seamlessly accommodates both fully autoregressive generation (one insertion at a time) and partially autoregressive generation (simultaneous insertions at multiple locations). We validate our approach by analyzing its performance on the WMT 2014 English-German machine translation task under various settings for training and decoding. We find that the Insertion Transformer outperforms many prior non-autoregressive approaches to translation at comparable or better levels of parallelism, and successfully recovers the performance of the original Transformer while requiring only logarithmically many iterations during decoding.
Tasks Machine Translation
Published 2019-02-08
URL http://arxiv.org/abs/1902.03249v1
PDF http://arxiv.org/pdf/1902.03249v1.pdf
PWC https://paperswithcode.com/paper/insertion-transformer-flexible-sequence
Repo
Framework

A Review of Changepoint Detection Models

Title A Review of Changepoint Detection Models
Authors Yixiao Li, Gloria Lin, Thomas Lau, Ruochen Zeng
Abstract The objective of the change-point detection is to discover the abrupt property changes lying behind the time-series data. In this paper, we firstly summarize the definition and in-depth implication of the changepoint detection. The next stage is to elaborate traditional and some alternative model-based changepoint detection algorithms. Finally, we try to go a bit further in the theory and look into future research directions.
Tasks Change Point Detection, Time Series
Published 2019-08-20
URL https://arxiv.org/abs/1908.07136v1
PDF https://arxiv.org/pdf/1908.07136v1.pdf
PWC https://paperswithcode.com/paper/a-review-of-changepoint-detection-models
Repo
Framework

Solving Empirical Risk Minimization in the Current Matrix Multiplication Time

Title Solving Empirical Risk Minimization in the Current Matrix Multiplication Time
Authors Yin Tat Lee, Zhao Song, Qiuyi Zhang
Abstract Many convex problems in machine learning and computer science share the same form: \begin{align*} \min_{x} \sum_{i} f_i( A_i x + b_i), \end{align*} where $f_i$ are convex functions on $\mathbb{R}^{n_i}$ with constant $n_i$, $A_i \in \mathbb{R}^{n_i \times d}$, $b_i \in \mathbb{R}^{n_i}$ and $\sum_i n_i = n$. This problem generalizes linear programming and includes many problems in empirical risk minimization. In this paper, we give an algorithm that runs in time \begin{align*} O^* ( ( n^{\omega} + n^{2.5 - \alpha/2} + n^{2+ 1/6} ) \log (n / \delta) ) \end{align*} where $\omega$ is the exponent of matrix multiplication, $\alpha$ is the dual exponent of matrix multiplication, and $\delta$ is the relative accuracy. Note that the runtime has only a log dependence on the condition numbers or other data dependent parameters and these are captured in $\delta$. For the current bound $\omega \sim 2.38$ [Vassilevska Williams’12, Le Gall’14] and $\alpha \sim 0.31$ [Le Gall, Urrutia’18], our runtime $O^* ( n^{\omega} \log (n / \delta))$ matches the current best for solving a dense least squares regression problem, a special case of the problem we consider. Very recently, [Alman’18] proved that all the current known techniques can not give a better $\omega$ below $2.168$ which is larger than our $2+1/6$. Our result generalizes the very recent result of solving linear programs in the current matrix multiplication time [Cohen, Lee, Song’19] to a more broad class of problems. Our algorithm proposes two concepts which are different from [Cohen, Lee, Song’19] : $\bullet$ We give a robust deterministic central path method, whereas the previous one is a stochastic central path which updates weights by a random sparse vector. $\bullet$ We propose an efficient data-structure to maintain the central path of interior point methods even when the weights update vector is dense.
Tasks
Published 2019-05-11
URL https://arxiv.org/abs/1905.04447v1
PDF https://arxiv.org/pdf/1905.04447v1.pdf
PWC https://paperswithcode.com/paper/solving-empirical-risk-minimization-in-the
Repo
Framework

Understanding the Artificial Intelligence Clinician and optimal treatment strategies for sepsis in intensive care

Title Understanding the Artificial Intelligence Clinician and optimal treatment strategies for sepsis in intensive care
Authors Matthieu Komorowski, Leo A. Celi, Omar Badawi, Anthony C. Gordon, A. Aldo Faisal
Abstract In this document, we explore in more detail our published work (Komorowski, Celi, Badawi, Gordon, & Faisal, 2018) for the benefit of the AI in Healthcare research community. In the above paper, we developed the AI Clinician system, which demonstrated how reinforcement learning could be used to make useful recommendations towards optimal treatment decisions from intensive care data. Since publication a number of authors have reviewed our work (e.g. Abbasi, 2018; Bos, Azoulay, & Martin-Loeches, 2019; Saria, 2018). Given the difference of our framework to previous work, the fact that we are bridging two very different academic communities (intensive care and machine learning) and that our work has impact on a number of other areas with more traditional computer-based approaches (biosignal processing and control, biomedical engineering), we are providing here additional details on our recent publication.
Tasks
Published 2019-03-06
URL http://arxiv.org/abs/1903.02345v1
PDF http://arxiv.org/pdf/1903.02345v1.pdf
PWC https://paperswithcode.com/paper/understanding-the-artificial-intelligence
Repo
Framework

Quantum-inspired classical sublinear-time algorithm for solving low-rank semidefinite programming via sampling approaches

Title Quantum-inspired classical sublinear-time algorithm for solving low-rank semidefinite programming via sampling approaches
Authors Nai-Hui Chia, Tongyang Li, Han-Hsuan Lin, Chunhao Wang
Abstract Semidefinite programming (SDP) is a central topic in mathematical optimization with extensive studies on its efficient solvers. Recently, quantum algorithms with superpolynomial speedups for solving SDPs have been proposed assuming access to its constraint matrices in quantum superposition. Mutually inspired by both classical and quantum SDP solvers, in this paper we present a sublinear classical algorithm for solving low-rank SDPs which is asymptotically as good as existing quantum algorithms. Specifically, given an SDP with $m$ constraint matrices, each of dimension $n$ and rank $\mathrm{poly}(\log n)$, our algorithm gives a succinct description and any entry of the solution matrix in time $O(m\cdot\mathrm{poly}(\log n,1/\varepsilon))$ given access to a sample-based low-overhead data structure of the constraint matrices, where $\varepsilon$ is the precision of the solution. In addition, we apply our algorithm to a quantum state learning task as an application. Technically, our approach aligns with both the SDP solvers based on the matrix multiplicative weight (MMW) framework and the recent studies of quantum-inspired machine learning algorithms. The cost of solving SDPs by MMW mainly comes from the exponentiation of Hermitian matrices, and we propose two new technical ingredients (compared to previous sample-based algorithms) for this task that may be of independent interest: $\bullet$ Weighted sampling: assuming sampling access to each individual constraint matrix $A_{1},\ldots,A_{\tau}$, we propose a procedure that gives a good approximation of $A=A_{1}+\cdots+A_{\tau}$. $\bullet$ Symmetric approximation: we propose a sampling procedure that gives low-rank spectral decomposition of a Hermitian matrix $A$. This improves upon previous sampling procedures that only give low-rank singular value decompositions, losing the signs of eigenvalues.
Tasks
Published 2019-01-10
URL http://arxiv.org/abs/1901.03254v1
PDF http://arxiv.org/pdf/1901.03254v1.pdf
PWC https://paperswithcode.com/paper/quantum-inspired-classical-sublinear-time
Repo
Framework

No-Regret Exploration in Goal-Oriented Reinforcement Learning

Title No-Regret Exploration in Goal-Oriented Reinforcement Learning
Authors Jean Tarbouriech, Evrard Garcelon, Michal Valko, Matteo Pirotta, Alessandro Lazaric
Abstract Many popular reinforcement learning problems (e.g., navigation in a maze, some Atari games, mountain car) are instances of the episodic setting under its stochastic shortest path (SSP) formulation, where an agent has to achieve a goal state while minimizing the cumulative cost. Despite the popularity of this setting, the exploration-exploitation dilemma has been sparsely studied in general SSP problems, with most of the theoretical literature focusing on different problems (i.e., fixed-horizon and infinite-horizon) or making the restrictive loop-free SSP assumption (i.e., no state can be visited twice during an episode). In this paper, we study the general SSP problem with no assumption on its dynamics (some policies may actually never reach the goal). We introduce UC-SSP, the first no-regret algorithm in this setting, and prove a regret bound scaling as $\displaystyle \widetilde{\mathcal{O}}( D S \sqrt{ A D K})$ after $K$ episodes for any unknown SSP with $S$ states, $A$ actions, positive costs and SSP-diameter $D$, defined as the smallest expected hitting time from any starting state to the goal. We achieve this result by crafting a novel stopping rule, such that UC-SSP may interrupt the current policy if it is taking too long to achieve the goal and switch to alternative policies that are designed to rapidly terminate the episode.
Tasks Atari Games
Published 2019-12-07
URL https://arxiv.org/abs/1912.03517v2
PDF https://arxiv.org/pdf/1912.03517v2.pdf
PWC https://paperswithcode.com/paper/no-regret-exploration-in-goal-oriented
Repo
Framework
comments powered by Disqus