October 16, 2019

2986 words 15 mins read

Paper Group ANR 1079

Interpretable Charge Predictions for Criminal Cases: Learning to Generate Court Views from Fact Descriptions. End-to-end driving simulation via angle branched network. Understanding the Loss Surface of Neural Networks for Binary Classification. Leveraging Gaussian Process and Voting-Empowered Many-Objective Evaluation for Fault Identification. From …

Interpretable Charge Predictions for Criminal Cases: Learning to Generate Court Views from Fact Descriptions


Title	Interpretable Charge Predictions for Criminal Cases: Learning to Generate Court Views from Fact Descriptions
Authors	Hai Ye, Xin Jiang, Zhunchen Luo, Wenhan Chao
Abstract	In this paper, we propose to study the problem of COURT VIEW GENeration from the fact description in a criminal case. The task aims to improve the interpretability of charge prediction systems and help automatic legal document generation. We formulate this task as a text-to-text natural language generation (NLG) problem. Sequenceto-sequence model has achieved cutting-edge performances in many NLG tasks. However, due to the non-distinctions of fact descriptions, it is hard for Seq2Seq model to generate charge-discriminative court views. In this work, we explore charge labels to tackle this issue. We propose a label-conditioned Seq2Seq model with attention for this problem, to decode court views conditioned on encoded charge labels. Experimental results show the effectiveness of our method.
Tasks	Text Generation
Published	2018-02-23
URL	http://arxiv.org/abs/1802.08504v1
PDF	http://arxiv.org/pdf/1802.08504v1.pdf
PWC	https://paperswithcode.com/paper/interpretable-charge-predictions-for-criminal
Repo
Framework

End-to-end driving simulation via angle branched network


Title	End-to-end driving simulation via angle branched network
Authors	Qing Wang, Long Chen, Wei Tian
Abstract	Imitation learning for end-to-end autonomous driving has drawn attention from academic communities. Current methods either only use images as the input which is ambiguous when a car approaches an intersection, or use additional command information to navigate the vehicle but not automated enough. Focusing on making the vehicle drive along the given path, we propose a new navigation command that does not require human’s participation and a novel model architecture called angle branched network. Both the new navigation command and the angle branched network are easy to understand and effective. Besides, we find that not only segmentation information but also depth information can boost the performance of the driving model. We conduct experiments in a 3D urban simulator and both qualitative and quantitative evaluation results show the effectiveness of our model.
Tasks	Autonomous Driving, Imitation Learning
Published	2018-05-19
URL	http://arxiv.org/abs/1805.07545v1
PDF	http://arxiv.org/pdf/1805.07545v1.pdf
PWC	https://paperswithcode.com/paper/end-to-end-driving-simulation-via-angle
Repo
Framework

Understanding the Loss Surface of Neural Networks for Binary Classification


Title	Understanding the Loss Surface of Neural Networks for Binary Classification
Authors	Shiyu Liang, Ruoyu Sun, Yixuan Li, R. Srikant
Abstract	It is widely conjectured that the reason that training algorithms for neural networks are successful because all local minima lead to similar performance, for example, see (LeCun et al., 2015, Choromanska et al., 2015, Dauphin et al., 2014). Performance is typically measured in terms of two metrics: training performance and generalization performance. Here we focus on the training performance of single-layered neural networks for binary classification, and provide conditions under which the training error is zero at all local minima of a smooth hinge loss function. Our conditions are roughly in the following form: the neurons have to be strictly convex and the surrogate loss function should be a smooth version of hinge loss. We also provide counterexamples to show that when the loss function is replaced with quadratic loss or logistic loss, the result may not hold.
Tasks
Published	2018-02-19
URL	http://arxiv.org/abs/1803.00909v2
PDF	http://arxiv.org/pdf/1803.00909v2.pdf
PWC	https://paperswithcode.com/paper/understanding-the-loss-surface-of-neural
Repo
Framework

Leveraging Gaussian Process and Voting-Empowered Many-Objective Evaluation for Fault Identification


Title	Leveraging Gaussian Process and Voting-Empowered Many-Objective Evaluation for Fault Identification
Authors	Pei Cao, Qi Shuai, Jiong Tang
Abstract	Using piezoelectric impedance/admittance sensing for structural health monitoring is promising, owing to the simplicity in circuitry design as well as the high-frequency interrogation capability. The actual identification of fault location and severity using impedance/admittance measurements, nevertheless, remains to be an extremely challenging task. A first-principle based structural model using finite element discretization requires high dimensionality to characterize the high-frequency response. As such, direct inversion using the sensitivity matrix usually yields an under-determined problem. Alternatively, the identification problem may be cast into an optimization framework in which fault parameters are identified through repeated forward finite element analysis which however is oftentimes computationally prohibitive. This paper presents an efficient data-assisted optimization approach for fault identification without using finite element model iteratively. We formulate a many-objective optimization problem to identify fault parameters, where response surfaces of impedance measurements are constructed through Gaussian process-based calibration. To balance between solution diversity and convergence, an -dominance enabled many-objective simulated annealing algorithm is established. As multiple solutions are expected, a voting score calculation procedure is developed to further identify those solutions that yield better implications regarding structural health condition. The effectiveness of the proposed approach is demonstrated by systematic numerical and experimental case studies.
Tasks	Calibration
Published	2018-10-29
URL	http://arxiv.org/abs/1810.12228v1
PDF	http://arxiv.org/pdf/1810.12228v1.pdf
PWC	https://paperswithcode.com/paper/leveraging-gaussian-process-and-voting
Repo
Framework

From Self-ception to Image Self-ception: A method to represent an image with its own approximations


Title	From Self-ception to Image Self-ception: A method to represent an image with its own approximations
Authors	Hamed Shah-Hosseini
Abstract	A concept of defining images based on its own approximate ones is proposed here, which is called ‘Self-ception’. In this regard, an algorithm is proposed to implement the self-ception for images, which we call it ‘Image Self-ception’ since we use it for images. We can control the accuracy of this self-ception representation by deciding how many segments or regions we want to use for the representation. Some self-ception images are included in the paper. The video versions of the proposed image self-ception algorithm in action are shown in a YouTube channel (find it by Googling image self-ception).
Tasks
Published	2018-06-14
URL	http://arxiv.org/abs/1806.05610v1
PDF	http://arxiv.org/pdf/1806.05610v1.pdf
PWC	https://paperswithcode.com/paper/from-self-ception-to-image-self-ception-a
Repo
Framework

Inverse reinforcement learning for video games


Title	Inverse reinforcement learning for video games
Authors	Aaron Tucker, Adam Gleave, Stuart Russell
Abstract	Deep reinforcement learning achieves superhuman performance in a range of video game environments, but requires that a designer manually specify a reward function. It is often easier to provide demonstrations of a target behavior than to design a reward function describing that behavior. Inverse reinforcement learning (IRL) algorithms can infer a reward from demonstrations in low-dimensional continuous control environments, but there has been little work on applying IRL to high-dimensional video games. In our CNN-AIRL baseline, we modify the state-of-the-art adversarial IRL (AIRL) algorithm to use CNNs for the generator and discriminator. To stabilize training, we normalize the reward and increase the size of the discriminator training dataset. We additionally learn a low-dimensional state representation using a novel autoencoder architecture tuned for video game environments. This embedding is used as input to the reward network, improving the sample efficiency of expert demonstrations. Our method achieves high-level performance on the simple Catcher video game, substantially outperforming the CNN-AIRL baseline. We also score points on the Enduro Atari racing game, but do not match expert performance, highlighting the need for further work.
Tasks	Continuous Control
Published	2018-10-24
URL	http://arxiv.org/abs/1810.10593v1
PDF	http://arxiv.org/pdf/1810.10593v1.pdf
PWC	https://paperswithcode.com/paper/inverse-reinforcement-learning-for-video
Repo
Framework

Semantic Classification of 3D Point Clouds with Multiscale Spherical Neighborhoods


Title	Semantic Classification of 3D Point Clouds with Multiscale Spherical Neighborhoods
Authors	Hugues Thomas, Jean-Emmanuel Deschaud, Beatriz Marcotegui, François Goulette, Yann Le Gall
Abstract	This paper introduces a new definition of multiscale neighborhoods in 3D point clouds. This definition, based on spherical neighborhoods and proportional subsampling, allows the computation of features with a consistent geometrical meaning, which is not the case when using k-nearest neighbors. With an appropriate learning strategy, the proposed features can be used in a random forest to classify 3D points. In this semantic classification task, we show that our multiscale features outperform state-of-the-art features using the same experimental conditions. Furthermore, their classification power competes with more elaborate classification approaches including Deep Learning methods.
Tasks	Semantic Segmentation
Published	2018-08-01
URL	http://arxiv.org/abs/1808.00495v1
PDF	http://arxiv.org/pdf/1808.00495v1.pdf
PWC	https://paperswithcode.com/paper/semantic-classification-of-3d-point-clouds
Repo
Framework

Penalizing Top Performers: Conservative Loss for Semantic Segmentation Adaptation


Title	Penalizing Top Performers: Conservative Loss for Semantic Segmentation Adaptation
Authors	Xinge Zhu, Hui Zhou, Ceyuan Yang, Jianping Shi, Dahua Lin
Abstract	Due to the expensive and time-consuming annotations (e.g., segmentation) for real-world images, recent works in computer vision resort to synthetic data. However, the performance on the real image often drops significantly because of the domain shift between the synthetic data and the real images. In this setting, domain adaptation brings an appealing option. The effective approaches of domain adaptation shape the representations that (1) are discriminative for the main task and (2) have good generalization capability for domain shift. To this end, we propose a novel loss function, i.e., Conservative Loss, which penalizes the extreme good and bad cases while encouraging the moderate examples. More specifically, it enables the network to learn features that are discriminative by gradient descent and are invariant to the change of domains via gradient ascend method. Extensive experiments on synthetic to real segmentation adaptation show our proposed method achieves state of the art results. Ablation studies give more insights into properties of the Conservative Loss. Exploratory experiments and discussion demonstrate that our Conservative Loss has good flexibility rather than restricting an exact form.
Tasks	Domain Adaptation, Semantic Segmentation
Published	2018-09-04
URL	http://arxiv.org/abs/1809.00903v2
PDF	http://arxiv.org/pdf/1809.00903v2.pdf
PWC	https://paperswithcode.com/paper/penalizing-top-performers-conservative-loss
Repo
Framework

RGB-Depth SLAM Review


Title	RGB-Depth SLAM Review
Authors	Redhwan Jamiruddin, Ali Osman Sari, Jahanzaib Shabbir, Tarique Anwer
Abstract	Simultaneous Localization and Mapping (SLAM) have made the real-time dense reconstruction possible increasing the prospects of navigation, tracking, and augmented reality problems. Some breakthroughs have been achieved in this regard during past few decades and more remarkable works are still going on. This paper presents an overview of SLAM approaches that have been developed till now. Kinect Fusion algorithm, its variants, and further developed approaches are discussed in detailed. The algorithms and approaches are compared for their effectiveness in tracking and mapping based on Root Mean Square error over online available datasets.
Tasks	Simultaneous Localization and Mapping
Published	2018-05-20
URL	http://arxiv.org/abs/1805.07696v1
PDF	http://arxiv.org/pdf/1805.07696v1.pdf
PWC	https://paperswithcode.com/paper/rgb-depth-slam-review
Repo
Framework

DeepConsensus: using the consensus of features from multiple layers to attain robust image classification


Title	DeepConsensus: using the consensus of features from multiple layers to attain robust image classification
Authors	Yuchen Li, Safwan Hossain, Kiarash Jamali, Frank Rudzicz
Abstract	We consider a classifier whose test set is exposed to various perturbations that are not present in the training set. These test samples still contain enough features to map them to the same class as their unperturbed counterpart. Current architectures exhibit rapid degradation of accuracy when trained on standard datasets but then used to classify perturbed samples of that data. To address this, we present a novel architecture named DeepConsensus that significantly improves generalization to these test-time perturbations. Our key insight is that deep neural networks should directly consider summaries of low and high level features when making classifications. Existing convolutional neural networks can be augmented with DeepConsensus, leading to improved resistance against large and small perturbations on MNIST, EMNIST, FashionMNIST, CIFAR10 and SVHN datasets.
Tasks	Image Classification
Published	2018-11-18
URL	http://arxiv.org/abs/1811.07266v3
PDF	http://arxiv.org/pdf/1811.07266v3.pdf
PWC	https://paperswithcode.com/paper/deepconsensus-using-the-consensus-of-features
Repo
Framework

Dynamical Isometry and a Mean Field Theory of RNNs: Gating Enables Signal Propagation in Recurrent Neural Networks


Title	Dynamical Isometry and a Mean Field Theory of RNNs: Gating Enables Signal Propagation in Recurrent Neural Networks
Authors	Minmin Chen, Jeffrey Pennington, Samuel S. Schoenholz
Abstract	Recurrent neural networks have gained widespread use in modeling sequence data across various domains. While many successful recurrent architectures employ a notion of gating, the exact mechanism that enables such remarkable performance is not well understood. We develop a theory for signal propagation in recurrent networks after random initialization using a combination of mean field theory and random matrix theory. To simplify our discussion, we introduce a new RNN cell with a simple gating mechanism that we call the minimalRNN and compare it with vanilla RNNs. Our theory allows us to define a maximum timescale over which RNNs can remember an input. We show that this theory predicts trainability for both recurrent architectures. We show that gated recurrent networks feature a much broader, more robust, trainable region than vanilla RNNs, which corroborates recent experimental findings. Finally, we develop a closed-form critical initialization scheme that achieves dynamical isometry in both vanilla RNNs and minimalRNNs. We show that this results in significantly improvement in training dynamics. Finally, we demonstrate that the minimalRNN achieves comparable performance to its more complex counterparts, such as LSTMs or GRUs, on a language modeling task.
Tasks	Language Modelling
Published	2018-06-14
URL	http://arxiv.org/abs/1806.05394v2
PDF	http://arxiv.org/pdf/1806.05394v2.pdf
PWC	https://paperswithcode.com/paper/dynamical-isometry-and-a-mean-field-theory-of
Repo
Framework

apk2vec: Semi-supervised multi-view representation learning for profiling Android applications


Title	apk2vec: Semi-supervised multi-view representation learning for profiling Android applications
Authors	Annamalai Narayanan, Charlie Soh, Lihui Chen, Yang Liu, Lipo Wang
Abstract	Building behavior profiles of Android applications (apps) with holistic, rich and multi-view information (e.g., incorporating several semantic views of an app such as API sequences, system calls, etc.) would help catering downstream analytics tasks such as app categorization, recommendation and malware analysis significantly better. Towards this goal, we design a semi-supervised Representation Learning (RL) framework named apk2vec to automatically generate a compact representation (aka profile/embedding) for a given app. More specifically, apk2vec has the three following unique characteristics which make it an excellent choice for largescale app profiling: (1) it encompasses information from multiple semantic views such as API sequences, permissions, etc., (2) being a semi-supervised embedding technique, it can make use of labels associated with apps (e.g., malware family or app category labels) to build high quality app profiles, and (3) it combines RL and feature hashing which allows it to efficiently build profiles of apps that stream over time (i.e., online learning). The resulting semi-supervised multi-view hash embeddings of apps could then be used for a wide variety of downstream tasks such as the ones mentioned above. Our extensive evaluations with more than 42,000 apps demonstrate that apk2vec’s app profiles could significantly outperform state-of-the-art techniques in four app analytics tasks namely, malware detection, familial clustering, app clone detection and app recommendation.
Tasks	Malware Detection, Representation Learning
Published	2018-09-15
URL	http://arxiv.org/abs/1809.05693v1
PDF	http://arxiv.org/pdf/1809.05693v1.pdf
PWC	https://paperswithcode.com/paper/apk2vec-semi-supervised-multi-view
Repo
Framework

A Block-wise, Asynchronous and Distributed ADMM Algorithm for General Form Consensus Optimization


Title	A Block-wise, Asynchronous and Distributed ADMM Algorithm for General Form Consensus Optimization
Authors	Rui Zhu, Di Niu, Zongpeng Li
Abstract	Many machine learning models, including those with non-smooth regularizers, can be formulated as consensus optimization problems, which can be solved by the alternating direction method of multipliers (ADMM). Many recent efforts have been made to develop asynchronous distributed ADMM to handle large amounts of training data. However, all existing asynchronous distributed ADMM methods are based on full model updates and require locking all global model parameters to handle concurrency, which essentially serializes the updates from different workers. In this paper, we present a novel block-wise, asynchronous and distributed ADMM algorithm, which allows different blocks of model parameters to be updated in parallel. The lock-free block-wise algorithm may greatly speedup sparse optimization problems, a common scenario in reality, in which most model updates only modify a subset of all decision variables. We theoretically prove the convergence of our proposed algorithm to stationary points for non-convex general form consensus problems with possibly non-smooth regularizers. We implement the proposed ADMM algorithm on the Parameter Server framework and demonstrate its convergence and near-linear speedup performance as the number of workers increases.
Tasks
Published	2018-02-24
URL	http://arxiv.org/abs/1802.08882v1
PDF	http://arxiv.org/pdf/1802.08882v1.pdf
PWC	https://paperswithcode.com/paper/a-block-wise-asynchronous-and-distributed
Repo
Framework

Robust Subspace Clustering with Compressed Data


Title	Robust Subspace Clustering with Compressed Data
Authors	Guangcan Liu, Zhao Zhang, Qingshan Liu, Kongkai Xiong
Abstract	Dimension reduction is widely regarded as an effective way for decreasing the computation, storage and communication loads of data-driven intelligent systems, leading to a growing demand for statistical methods that allow analysis (e.g., clustering) of compressed data. We therefore study in this paper a novel problem called compressive robust subspace clustering, which is to perform robust subspace clustering with the compressed data, and which is generated by projecting the original high-dimensional data onto a lower-dimensional subspace chosen at random. Given only the compressed data and sensing matrix, the proposed method, row space pursuit (RSP), recovers the authentic row space that gives correct clustering results under certain conditions. Extensive experiments show that RSP is distinctly better than the competing methods, in terms of both clustering accuracy and computational efficiency.
Tasks	Dimensionality Reduction
Published	2018-03-30
URL	https://arxiv.org/abs/1803.11305v6
PDF	https://arxiv.org/pdf/1803.11305v6.pdf
PWC	https://paperswithcode.com/paper/robust-subspace-clustering-with-compressed
Repo
Framework


Title	From Eliza to XiaoIce: Challenges and Opportunities with Social Chatbots
Authors	Heung-Yeung Shum, Xiaodong He, Di Li
Abstract	Conversational systems have come a long way since their inception in the 1960s. After decades of research and development, we’ve seen progress from Eliza and Parry in the 60’s and 70’s, to task-completion systems as in the DARPA Communicator program in the 2000s, to intelligent personal assistants such as Siri in the 2010s, to today’s social chatbots like XiaoIce. Social chatbots’ appeal lies not only in their ability to respond to users’ diverse requests, but also in being able to establish an emotional connection with users. The latter is done by satisfying users’ need for communication, affection, as well as social belonging. To further the advancement and adoption of social chatbots, their design must focus on user engagement and take both intellectual quotient (IQ) and emotional quotient (EQ) into account. Users should want to engage with a social chatbot; as such, we define the success metric for social chatbots as conversation-turns per session (CPS). Using XiaoIce as an illustrative example, we discuss key technologies in building social chatbots from core chat to visual awareness to skills. We also show how XiaoIce can dynamically recognize emotion and engage the user throughout long conversations with appropriate interpersonal responses. As we become the first generation of humans ever living with AI, we have a responsibility to design social chatbots to be both useful and empathetic, so they will become ubiquitous and help society as a whole.
Tasks	Chatbot
Published	2018-01-06
URL	http://arxiv.org/abs/1801.01957v2
PDF	http://arxiv.org/pdf/1801.01957v2.pdf
PWC	https://paperswithcode.com/paper/from-eliza-to-xiaoice-challenges-and
Repo
Framework