{"title":"Deep Mimo Detection Using ADMM Unfolding","authors":"M. Un, Mingjie Shao, Wing-Kin Ma, P. Ching","doi":"10.1109/DSW.2019.8755566","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755566","url":null,"abstract":"This paper presents a low-complexity deep neural network (DNN)based multiple-input-multiple-output (MIMO) detector for the BPSK and QPSK constellation cases. We employ deep unfolding, whose idea is to take insight from the structure of an iterative optimization algorithm and attempt to learn a better iterative algorithm. The structure of the network is obtained from an iterative algorithm arising from the application of ADMM to the maximum-likelihood MIMO detection problem. The number of parameters to be learnt in this new design is less than that of DetNet, a recently proposed DNN-based MIMO detector. Our numerical experiments illustrate that the new method outperforms DetNet and several existing MIMO detectors in the large-scale MIMO case. In particular, we show that for a 160×160 MIMO system, our DNN design, with 40 layers, can attain nearly optimal bit-error rate performance.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126441607","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Distance-Penalized Active Learning via Markov Decision Processes","authors":"Dingyu Wang, J. Lipor, Gautam Dasarathy","doi":"10.1109/DSW.2019.8755602","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755602","url":null,"abstract":"We consider the problem of active learning in the context of spatial sampling, where the measurements are obtained by a mobile sampling unit. The goal is to localize the change point of a one-dimensional threshold classifier while minimizing the total sampling time, a function of both the cost of sampling and the distance traveled. In this paper, we present a general framework for active learning by modeling the search problem as a Markov decision process. Using this framework, we present time-optimal algorithms for the spatial sampling problem when there is a uniform prior on the change point, a known non-uniform prior on the change point, and a need to return to the origin for intermittent battery recharging. We demonstrate through simulations that our proposed algorithms significantly outperform existing methods while maintaining a low computational cost.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128760563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kejun Huang, Zhuoran Yang, Zhaoran Wang, Mingyi Hong
{"title":"Learning Partially Observable Markov Decision Processes Using Coupled Canonical Polyadic Decomposition","authors":"Kejun Huang, Zhuoran Yang, Zhaoran Wang, Mingyi Hong","doi":"10.1109/DSW.2019.8755551","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755551","url":null,"abstract":"We propose a new algorithm for learning the model parameters of a partially observable Markov decision process (POMDP) based on coupled canonical polyadic decomposition (CPD). Coupled CPD for a set of tensors is an extension to CPD for individual tensors, which has improved identifiability properties, as well as an analogous simultaneous diagonalization (SD) algorithm for uniquely recovering the latent factors efficiently. We explain how to form a set of three-way tensors from the trajectory of a POMDP under a stationary memoryless policy, so that coupled CPD can be applied afterwards to recover the model parameters, with identifiability and computational guarantees.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133418504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Estimation of State-Space Models with Gaussian Mixture Process Noise","authors":"Sina Miran, J. Simon, M. Fu, S. Marcus, B. Babadi","doi":"10.1109/DSW.2019.8755571","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755571","url":null,"abstract":"State-space models are widely used to estimate latent dynamic processes from noisy and low-dimensional observations. When applying these models to real data, it is commonly assumed that the state dynamics are governed by Gaussian statistics. However, this assumption does not hold in applications where the process noise is composed of various exogenous components with heterogeneous statistics, resulting in a multimodal distribution. In this work, we consider a state-space model with Gaussian mixture process noise to account for such multimodality. We integrate the Expectation Maximization algorithm with sequential Monte Carlo methods to jointly estimate the Gaussian mixture parameters and states from noisy and low-dimensional observations. We validate our proposed method using simulated data inspired by auditory neuroscience, which reveals significant gains in state estimation as compared to widely used techniques that assume Gaussian state dynamics.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132874818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Hierarchical Approach for Timely Cyberbullying Detection","authors":"I. Nazar, Daphney-Stavroula Zois, Mengfan Yao","doi":"10.1109/DSW.2019.8755598","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755598","url":null,"abstract":"In this paper, the problem of accurately detecting cyberbullying instances in a timely manner is addressed. In stark contrast to most prior work that attempts to detect aggressive behavior by looking at individual messages, we consider cyberbullying as a repeated aggressive behavior towards an individual and jointly examine multiple messages exchanged between users. Moreover, we are interested in reaching an accurate decision fast. To this end, we propose a novel hierarchical approach that (i) first characterizes an individual message as aggressive or not by evaluating the optimum least number of informative features extracted from this message, and (ii) uses this new knowledge to decide if it should continue reviewing messages or conclude the process and raise a cyberbullying alert. It is shown that the proposed approach is guaranteed to review the least number of messages before reaching a decision, while the optimum decision rule is shown to minimize the average Bayes risk. Evaluation on real–world Instagram data demonstrates that the proposed method is able to accurately detect cyberbullying instances by reviewing up to 59% less messages than the state–of–the–art.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"234 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122455584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Long-Range Dependence Parameter Estimation For Mixed Spectra Gaussian Processes","authors":"Lenin Arango-Castillo, G. Takahara","doi":"10.1109/DSW.2019.8755562","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755562","url":null,"abstract":"We present an algorithm to estimate the Hurst parameter, H, in processes with mixed spectra. We focus on the two most studied Gaussian Long-Range Dependent (GLRD) processes: fractional Gaussian noise, fGn(H), and fractional integrated, FI(H), with Gaussian innovations. Our method consists in the decomposition of signal and noise components of the processes under analysis: first, we use harmonic analysis for detection of line components; second, we remove the line components mixed with the GLRD using information from their amplitudes iteratively; and finally, we use a robust method for the estimation of the LRD parameter, H. We apply the method to synthetic periodic time series, and numerical results are presented. The technique is applied to stock-market trading volume time series.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132460048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Online selective training for faster neural network learning","authors":"Sara Mourad, H. Vikalo, A. Tewfik","doi":"10.1109/DSW.2019.8755604","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755604","url":null,"abstract":"Training neural networks is a computationally challenging problem that requires significant time efforts. In this paper, we propose two approaches that improve efficiency of this task by actively selecting most relevant points from a training data set. The first approach forms a batch that maximizes the reduction of the estimator’s entropy, while the second approach only trains on datapoints whose predicted probability is below a predetermined threshold. Both techniques rely on data metrics to speed up training while retaining the epoch-based neural network training framework. The results demonstrate that the proposed methods enable significant reduction of training time in experiments on the CIFAR10 dataset without compromising the accuracy.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132407790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rasoul Shafipour, Abolfazl Hashemi, G. Mateos, H. Vikalo
{"title":"Online Topology Inference from Streaming Stationary Graph Signals","authors":"Rasoul Shafipour, Abolfazl Hashemi, G. Mateos, H. Vikalo","doi":"10.1109/DSW.2019.8755560","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755560","url":null,"abstract":"We address the problem of online topology inference from streaming nodal observations of graph signals generated by linear diffusion dynamics on the sought graph. To that end, we leverage the stationarity of the signals and use the so-called graph-shift operator (GSO) as a matrix representation of the graph. Under this model, estimated covariance eigenvectors obtained from streaming independent graph signals diffused on the sought network are a valid estimator of the GSO’s spectral templates. We develop an ADMM algorithm to find a sparse and structurally admissible GSO given the eigenvectors estimate. Then, we propose an online scheme that upon sensing new diffused observations, efficiently updates eigenvectors (thus makes more accurate on expectation) and performs only one or a few iterations of the mentioned ADMM until the new data is observed. Numerical tests illustrate the effectiveness of the proposed topology inference approach in recovering large scale graphs, adapting to streaming information, and accommodating changes in the sought network.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115490787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Coke: Communication-Censored Kernel Learning Via Random Features","authors":"Ping Xu, Z. Tian, Zhe Zhang, Yue Wang","doi":"10.1109/DSW.2019.8755802","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755802","url":null,"abstract":"Distributed kernel-based methods are attractive in nonlinear learning tasks where either a dataset is too large to be processed on a single machine or the data are only locally available to geographically-located sites. For the first case, we propose to split the large dataset into multiple mini-batches and distribute them to distinct sites for parallel learning through the alternating direction method of multipliers (ADMM). For the second case, we develop a decentralized ADMM so that each site can solve the learning task collaboratively through one-hop communications. To circumvent the curse of dimensionality in kernel-based methods, we leverage the random feature approximation to map the large-volume data into a smaller feature space. This also results in a common set of decision parameters that can be exchanged among sites. Motivated by the need to conserve energy and reduce communication overheads, we apply a censoring strategy to evaluate the updated parameter at each site and decide if this update is worth transmitting. The proposed COmmunication-censored KErnel learning (COKE) algorithms are corroborated to be communication-efficient and learning-effective by simulations on both synthetic and real datasets.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127283016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Statistical Learning Using Hierarchical Modeling of Probability Tensors","authors":"Magda Amiridi, Nikos Kargas, N. Sidiropoulos","doi":"10.1109/DSW.2019.8755580","DOIUrl":"https://doi.org/10.1109/DSW.2019.8755580","url":null,"abstract":"Estimating the joint distribution of data sampled from an unknown distribution is the holy grail for modeling the structure of a dataset and deriving any desired optimal estimator. Leveraging the mere definition of conditional probability, we address the complexity of accurately estimating high-dimensional joint distributions without any assumptions on the underlying structural model by proposing a novel hierarchical learning algorithm for probability mass function (PMF) estimation through parallel local views of a probability tensor. This way the overall problem of estimating a joint distribution is divided into multiple subproblems, all of which are conquered independently by applying regional low-rank non-negative tensor models using the Canonical Polyadic Decomposition (CPD). Using conditioning, such parallelization is possible without losing sight of the full model – which can be reconstructed from the local models and the conditional probabilities. We illustrate the effectiveness and potential of our approach through judicious experiments on real datasets.","PeriodicalId":207363,"journal":{"name":"2019 IEEE Data Science Workshop (DSW)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125508445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}