October 19, 2019

3164 words 15 mins read

Paper Group ANR 283

Map matching when the map is wrong: Efficient vehicle tracking on- and off-road for map learning. Visual Diagnostics for Deep Reinforcement Learning Policy Development. IGD Indicator-based Evolutionary Algorithm for Many-objective Optimization Problems. Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitat …

Map matching when the map is wrong: Efficient vehicle tracking on- and off-road for map learning


Title	Map matching when the map is wrong: Efficient vehicle tracking on- and off-road for map learning
Authors	James Murphy, Yuanyuan Pao, Albert Yuen
Abstract	Given a sequence of possibly sparse and noisy GPS traces and a map of the road network, map matching algorithms can infer the most accurate trajectory on the road network. However, if the road network is wrong (for example due to missing or incorrectly mapped roads, missing parking lots, misdirected turn restrictions or misdirected one-way streets) standard map matching algorithms fail to reconstruct the correct trajectory. In this paper, an algorithm to tracking vehicles able to move both on and off the known road network is formulated. It efficiently unifies existing hidden Markov model (HMM) approaches for map matching and standard free-space tracking methods (e.g. Kalman smoothing) in a principled way. The algorithm is a form of interacting multiple model (IMM) filter subject to an additional assumption on the type of model interaction permitted, termed here as semi-interacting multiple model (sIMM) filter. A forward filter (suitable for real-time tracking) and backward MAP sampling step (suitable for MAP trajectory inference and map matching) are described. The framework set out here is agnostic to the specific tracking models used, and makes clear how to replace these components with others of a similar type. In addition to avoiding generating misleading map matching trajectories, this algorithm can be applied to learn map features by detecting unmapped or incorrectly mapped roads and parking lots, incorrectly mapped turn restrictions and road directions.
Tasks
Published	2018-09-25
URL	https://arxiv.org/abs/1809.09755v2
PDF	https://arxiv.org/pdf/1809.09755v2.pdf
PWC	https://paperswithcode.com/paper/map-matching-when-the-map-is-wrong-efficient
Repo
Framework

Visual Diagnostics for Deep Reinforcement Learning Policy Development


Title	Visual Diagnostics for Deep Reinforcement Learning Policy Development
Authors	Jieliang Luo, Sam Green, Peter Feghali, George Legrady, Çetin Kaya Koç
Abstract	Modern vision-based reinforcement learning techniques often use convolutional neural networks (CNN) as universal function approximators to choose which action to take for a given visual input. Until recently, CNNs have been treated like black-box functions, but this mindset is especially dangerous when used for control in safety-critical settings. In this paper, we present our extensions of CNN visualization algorithms to the domain of vision-based reinforcement learning. We use a simulated drone environment as an example scenario. These visualization algorithms are an important tool for behavior introspection and provide insight into the qualities and flaws of trained policies when interacting with the physical world. A video may be seen at https://sites.google.com/view/drlvisual .
Tasks
Published	2018-09-14
URL	http://arxiv.org/abs/1809.06781v2
PDF	http://arxiv.org/pdf/1809.06781v2.pdf
PWC	https://paperswithcode.com/paper/visual-diagnostics-for-deep-reinforcement
Repo
Framework

IGD Indicator-based Evolutionary Algorithm for Many-objective Optimization Problems


Title	IGD Indicator-based Evolutionary Algorithm for Many-objective Optimization Problems
Authors	Yanan Sun, Gary G. Yen, Zhang Yi
Abstract	Inverted Generational Distance (IGD) has been widely considered as a reliable performance indicator to concurrently quantify the convergence and diversity of multi- and many-objective evolutionary algorithms. In this paper, an IGD indicator-based evolutionary algorithm for solving many-objective optimization problems (MaOPs) has been proposed. Specifically, the IGD indicator is employed in each generation to select the solutions with favorable convergence and diversity. In addition, a computationally efficient dominance comparison method is designed to assign the rank values of solutions along with three newly proposed proximity distance assignments. Based on these two designs, the solutions are selected from a global view by linear assignment mechanism to concern the convergence and diversity simultaneously. In order to facilitate the accuracy of the sampled reference points for the calculation of IGD indicator, we also propose an efficient decomposition-based nadir point estimation method for constructing the Utopian Pareto front which is regarded as the best approximate Pareto front for real-world MaOPs at the early stage of the evolution. To evaluate the performance, a series of experiments is performed on the proposed algorithm against a group of selected state-of-the-art many-objective optimization algorithms over optimization problems with $8$-, $15$-, and $20$-objective. Experimental results measured by the chosen performance metrics indicate that the proposed algorithm is very competitive in addressing MaOPs.
Tasks
Published	2018-02-24
URL	http://arxiv.org/abs/1802.08792v1
PDF	http://arxiv.org/pdf/1802.08792v1.pdf
PWC	https://paperswithcode.com/paper/igd-indicator-based-evolutionary-algorithm
Repo
Framework

Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning


Title	Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning
Authors	Ilya Kostrikov, Kumar Krishna Agrawal, Debidatta Dwibedi, Sergey Levine, Jonathan Tompson
Abstract	We identify two issues with the family of algorithms based on the Adversarial Imitation Learning framework. The first problem is implicit bias present in the reward functions used in these algorithms. While these biases might work well for some environments, they can also lead to sub-optimal behavior in others. Secondly, even though these algorithms can learn from few expert demonstrations, they require a prohibitively large number of interactions with the environment in order to imitate the expert for many real-world applications. In order to address these issues, we propose a new algorithm called Discriminator-Actor-Critic that uses off-policy Reinforcement Learning to reduce policy-environment interaction sample complexity by an average factor of 10. Furthermore, since our reward function is designed to be unbiased, we can apply our algorithm to many problems without making any task-specific adjustments.
Tasks	Imitation Learning
Published	2018-09-09
URL	http://arxiv.org/abs/1809.02925v2
PDF	http://arxiv.org/pdf/1809.02925v2.pdf
PWC	https://paperswithcode.com/paper/discriminator-actor-critic-addressing-sample
Repo
Framework

Deep Echo State Networks with Uncertainty Quantification for Spatio-Temporal Forecasting


Title	Deep Echo State Networks with Uncertainty Quantification for Spatio-Temporal Forecasting
Authors	Patrick L. McDermott, Christopher K. Wikle
Abstract	Long-lead forecasting for spatio-temporal systems can often entail complex nonlinear dynamics that are difficult to specify it a priori. Current statistical methodologies for modeling these processes are often highly parameterized and thus, challenging to implement from a computational perspective. One potential parsimonious solution to this problem is a method from the dynamical systems and engineering literature referred to as an echo state network (ESN). ESN models use so-called {\it reservoir computing} to efficiently compute recurrent neural network (RNN) forecasts. Moreover, so-called “deep” models have recently been shown to be successful at predicting high-dimensional complex nonlinear processes, particularly those with multiple spatial and temporal scales of variability (such as we often find in spatio-temporal environmental data). Here we introduce a deep ensemble ESN (D-EESN) model. We present two versions of this model for spatio-temporal processes that both produce forecasts and associated measures of uncertainty. The first approach utilizes a bootstrap ensemble framework and the second is developed within a hierarchical Bayesian framework (BD-EESN). This more general hierarchical Bayesian framework naturally accommodates non-Gaussian data types and multiple levels of uncertainties. The methodology is first applied to a data set simulated from a novel non-Gaussian multiscale Lorenz-96 dynamical system simulation model and then to a long-lead United States (U.S.) soil moisture forecasting application.
Tasks	Spatio-Temporal Forecasting
Published	2018-06-28
URL	http://arxiv.org/abs/1806.10728v2
PDF	http://arxiv.org/pdf/1806.10728v2.pdf
PWC	https://paperswithcode.com/paper/deep-echo-state-networks-with-uncertainty
Repo
Framework

Convolutional Neural Network Approach for EEG-based Emotion Recognition using Brain Connectivity and its Spatial Information


Title	Convolutional Neural Network Approach for EEG-based Emotion Recognition using Brain Connectivity and its Spatial Information
Authors	Seong-Eun Moon, Soobeom Jang, Jong-Seok Lee
Abstract	Emotion recognition based on electroencephalography (EEG) has received attention as a way to implement human-centric services. However, there is still much room for improvement, particularly in terms of the recognition accuracy. In this paper, we propose a novel deep learning approach using convolutional neural networks (CNNs) for EEG-based emotion recognition. In particular, we employ brain connectivity features that have not been used with deep learning models in previous studies, which can account for synchronous activations of different brain regions. In addition, we develop a method to effectively capture asymmetric brain activity patterns that are important for emotion recognition. Experimental results confirm the effectiveness of our approach.
Tasks	EEG, Emotion Recognition
Published	2018-09-12
URL	http://arxiv.org/abs/1809.04208v1
PDF	http://arxiv.org/pdf/1809.04208v1.pdf
PWC	https://paperswithcode.com/paper/convolutional-neural-network-approach-for-eeg
Repo
Framework

Automated detection of block falls in the north polar region of Mars


Title	Automated detection of block falls in the north polar region of Mars
Authors	L. Fanara, K. Gwinner, E. Hauber, J. Oberst
Abstract	We developed a change detection method for the identification of ice block falls using NASA’s HiRISE images of the north polar scarps on Mars. Our method is based on a Support Vector Machine (SVM), trained using Histograms of Oriented Gradients (HOG), and on blob detection. The SVM detects potential new blocks between a set of images; the blob detection, then, confirms the identification of a block inside the area indicated by the SVM and derives the shape of the block. The results from the automatic analysis were compared with block statistics from visual inspection. We tested our method in 6 areas consisting of 1000x1000 pixels, where several hundreds of blocks were identified. The results for the given test areas produced a true positive rate of ~75% for blocks with sizes larger than 0.7 m (i.e., approx. 3 times the available ground pixel size) and a false discovery rate of ~8.5%. Using blob detection we also recover the size of each block within 3 pixels of their actual size.
Tasks
Published	2018-12-20
URL	http://arxiv.org/abs/1812.08624v1
PDF	http://arxiv.org/pdf/1812.08624v1.pdf
PWC	https://paperswithcode.com/paper/automated-detection-of-block-falls-in-the
Repo
Framework

A Fast and Robust Matching Framework for Multimodal Remote Sensing Image Registration


Title	A Fast and Robust Matching Framework for Multimodal Remote Sensing Image Registration
Authors	Yuanxin Ye, Lorenzo Bruzzone, Jie Shan, Francesca Bovolo, Qing Zhu
Abstract	While image registration has been studied in remote sensing community for decades, registering multimodal data [e.g., optical, light detection and ranging (LiDAR), synthetic aperture radar (SAR), and map] remains a challenging problem because of significant nonlinear intensity differences between such data. To address this problem, we present a novel fast and robust matching framework integrating local descriptors for multimodal registration. In the proposed framework, a local descriptor (such as Histogram of Oriented Gradient (HOG), Local Self-Similarity or Speeded-Up Robust Feature) is first extracted at each pixel to form a pixel-wise feature representation of an image. Then we define a similarity measure based on the feature representation in frequency domain using the Fast Fourier Transform (FFT) technique, followed by a template matching scheme to detect control points between images. We also propose a novel pixel-wise feature representation using orientated gradients of images, which is named channel features of orientated gradients (CFOG). This novel feature is an extension of the pixel-wise HOG descriptor, and outperforms that both in matching performance and computational efficiency. The major advantages of the proposed framework include (1) structural similarity representation using the pixel-wise feature description and (2) high computational efficiency due to the use of FFT. Moreover, we design an automatic registration system for very large-size multimodal images based on the proposed framework. Experimental results obtained on many different types of multimodal images show the superior matching performance of the proposed framework with respect to the state-of-the-art methods and the effectiveness of the designed system, which show very good potential large-size image registration in real applications.
Tasks	Image Registration
Published	2018-08-19
URL	http://arxiv.org/abs/1808.06194v4
PDF	http://arxiv.org/pdf/1808.06194v4.pdf
PWC	https://paperswithcode.com/paper/a-fast-and-robust-matching-framework-for
Repo
Framework

OCTen: Online Compression-based Tensor Decomposition


Title	OCTen: Online Compression-based Tensor Decomposition
Authors	Ekta Gujral, Ravdeep Pasricha, Tianxiong Yang, Evangelos E. Papalexakis
Abstract	Tensor decompositions are powerful tools for large data analytics as they jointly model multiple aspects of data into one framework and enable the discovery of the latent structures and higher-order correlations within the data. One of the most widely studied and used decompositions, especially in data mining and machine learning, is the Canonical Polyadic or CP decomposition. However, today’s datasets are not static and these datasets often dynamically growing and changing with time. To operate on such large data, we present OCTen the first ever compression-based online parallel implementation for the CP decomposition. We conduct an extensive empirical analysis of the algorithms in terms of fitness, memory used and CPU time, and in order to demonstrate the compression and scalability of the method, we apply OCTen to big tensor data. Indicatively, OCTen performs on-par or better than state-of-the-art online and online methods in terms of decomposition accuracy and efficiency, while saving up to 40-200 % memory space.
Tasks
Published	2018-07-03
URL	http://arxiv.org/abs/1807.01350v1
PDF	http://arxiv.org/pdf/1807.01350v1.pdf
PWC	https://paperswithcode.com/paper/octen-online-compression-based-tensor
Repo
Framework

Building Safer AGI by introducing Artificial Stupidity


Title	Building Safer AGI by introducing Artificial Stupidity
Authors	Michaël Trazzi, Roman V. Yampolskiy
Abstract	Artificial Intelligence (AI) achieved super-human performance in a broad variety of domains. We say that an AI is made Artificially Stupid on a task when some limitations are deliberately introduced to match a human’s ability to do the task. An Artificial General Intelligence (AGI) can be made safer by limiting its computing power and memory, or by introducing Artificial Stupidity on certain tasks. We survey human intellectual limits and give recommendations for which limits to implement in order to build a safe AGI.
Tasks
Published	2018-08-11
URL	http://arxiv.org/abs/1808.03644v1
PDF	http://arxiv.org/pdf/1808.03644v1.pdf
PWC	https://paperswithcode.com/paper/building-safer-agi-by-introducing-artificial
Repo
Framework

Amortized Inference Regularization


Title	Amortized Inference Regularization
Authors	Rui Shu, Hung H. Bui, Shengjia Zhao, Mykel J. Kochenderfer, Stefano Ermon
Abstract	The variational autoencoder (VAE) is a popular model for density estimation and representation learning. Canonically, the variational principle suggests to prefer an expressive inference model so that the variational approximation is accurate. However, it is often overlooked that an overly-expressive inference model can be detrimental to the test set performance of both the amortized posterior approximator and, more importantly, the generative density estimator. In this paper, we leverage the fact that VAEs rely on amortized inference and propose techniques for amortized inference regularization (AIR) that control the smoothness of the inference model. We demonstrate that, by applying AIR, it is possible to improve VAE generalization on both inference and generative performance. Our paper challenges the belief that amortized inference is simply a mechanism for approximating maximum likelihood training and illustrates that regularization of the amortization family provides a new direction for understanding and improving generalization in VAEs.
Tasks	Density Estimation, Representation Learning
Published	2018-05-23
URL	http://arxiv.org/abs/1805.08913v2
PDF	http://arxiv.org/pdf/1805.08913v2.pdf
PWC	https://paperswithcode.com/paper/amortized-inference-regularization
Repo
Framework

Between hard and soft thresholding: optimal iterative thresholding algorithms


Title	Between hard and soft thresholding: optimal iterative thresholding algorithms
Authors	Haoyang Liu, Rina Foygel Barber
Abstract	Iterative thresholding algorithms seek to optimize a differentiable objective function over a sparsity or rank constraint by alternating between gradient steps that reduce the objective, and thresholding steps that enforce the constraint. This work examines the choice of the thresholding operator, and asks whether it is possible to achieve stronger guarantees than what is possible with hard thresholding. We develop the notion of relative concavity of a thresholding operator, a quantity that characterizes the worst-case convergence performance of any thresholding operator on the target optimization problem. Surprisingly, we find that commonly used thresholding operators, such as hard thresholding and soft thresholding, are suboptimal in terms of worst-case convergence guarantees. Instead, a general class of thresholding operators, lying between hard thresholding and soft thresholding, is shown to be optimal with the strongest possible convergence guarantee among all thresholding operators. Examples of this general class includes $\ell_q$ thresholding with appropriate choices of $q$, and a newly defined {\em reciprocal thresholding} operator. We also investigate the implications of the improved optimization guarantee in the statistical setting of sparse linear regression, and show that this new class of thresholding operators attain the optimal rate for computationally efficient estimators, matching the Lasso.
Tasks
Published	2018-04-24
URL	https://arxiv.org/abs/1804.08841v4
PDF	https://arxiv.org/pdf/1804.08841v4.pdf
PWC	https://paperswithcode.com/paper/between-hard-and-soft-thresholding-optimal
Repo
Framework

VerIDeep: Verifying Integrity of Deep Neural Networks through Sensitive-Sample Fingerprinting


Title	VerIDeep: Verifying Integrity of Deep Neural Networks through Sensitive-Sample Fingerprinting
Authors	Zecheng He, Tianwei Zhang, Ruby B. Lee
Abstract	Deep learning has become popular, and numerous cloud-based services are provided to help customers develop and deploy deep learning applications. Meanwhile, various attack techniques have also been discovered to stealthily compromise the model’s integrity. When a cloud customer deploys a deep learning model in the cloud and serves it to end-users, it is important for him to be able to verify that the deployed model has not been tampered with, and the model’s integrity is protected. We propose a new low-cost and self-served methodology for customers to verify that the model deployed in the cloud is intact, while having only black-box access (e.g., via APIs) to the deployed model. Customers can detect arbitrary changes to their deep learning models. Specifically, we define \texttt{Sensitive-Sample} fingerprints, which are a small set of transformed inputs that make the model outputs sensitive to the model’s parameters. Even small weight changes can be clearly reflected in the model outputs, and observed by the customer. Our experiments on different types of model integrity attacks show that we can detect model integrity breaches with high accuracy ($>$99%) and low overhead ($<$10 black-box model accesses).
Tasks
Published	2018-08-09
URL	http://arxiv.org/abs/1808.03277v2
PDF	http://arxiv.org/pdf/1808.03277v2.pdf
PWC	https://paperswithcode.com/paper/verideep-verifying-integrity-of-deep-neural
Repo
Framework

Exploring Textual and Speech information in Dialogue Act Classification with Speaker Domain Adaptation


Title	Exploring Textual and Speech information in Dialogue Act Classification with Speaker Domain Adaptation
Authors	Xuanli He, Quan Hung Tran, William Havard, Laurent Besacier, Ingrid Zukerman, Gholamreza Haffari
Abstract	In spite of the recent success of Dialogue Act (DA) classification, the majority of prior works focus on text-based classification with oracle transcriptions, i.e. human transcriptions, instead of Automatic Speech Recognition (ASR)‘s transcriptions. In spoken dialog systems, however, the agent would only have access to noisy ASR transcriptions, which may further suffer performance degradation due to domain shift. In this paper, we explore the effectiveness of using both acoustic and textual signals, either oracle or ASR transcriptions, and investigate speaker domain adaptation for DA classification. Our multimodal model proves to be superior to the unimodal models, particularly when the oracle transcriptions are not available. We also propose an effective method for speaker domain adaptation, which achieves competitive results.
Tasks	Dialogue Act Classification, Domain Adaptation, Speech Recognition
Published	2018-10-17
URL	http://arxiv.org/abs/1810.07455v1
PDF	http://arxiv.org/pdf/1810.07455v1.pdf
PWC	https://paperswithcode.com/paper/exploring-textual-and-speech-information-in
Repo
Framework

Efficient Learning of Restricted Boltzmann Machines Using Covariance Estimates


Title	Efficient Learning of Restricted Boltzmann Machines Using Covariance Estimates
Authors	Vidyadhar Upadhya, P. S. Sastry
Abstract	Learning RBMs using standard algorithms such as CD(k) involves gradient descent on the negative log-likelihood. One of the terms in the gradient, which involves expectation w.r.t. the model distribution, is intractable and is obtained through an MCMC estimate. In this work we show that the Hessian of the log-likelihood can be written in terms of covariances of hidden and visible units and hence, all elements of the Hessian can also be estimated using the same MCMC samples with small extra computational costs. Since inverting the Hessian may be computationally expensive, we propose an algorithm that uses inverse of the diagonal approximation of the Hessian, instead. This essentially results in parameter-specific adaptive learning rates for the gradient descent process and improves the efficiency of learning RBMs compared to the standard methods. Specifically we show that using the inverse of diagonal approximation of Hessian in the stochastic DC (difference of convex functions) program approach results in very efficient learning of RBMs.
Tasks
Published	2018-10-25
URL	https://arxiv.org/abs/1810.10777v2
PDF	https://arxiv.org/pdf/1810.10777v2.pdf
PWC	https://paperswithcode.com/paper/efficient-learning-of-restricted-boltzmann
Repo
Framework