January 26, 2020

3257 words 16 mins read

Paper Group ANR 1444

Low-Cost Transfer Learning of Face Tasks. Checking Chase Termination over Ontologies of Existential Rules with Equality. Distilling the Knowledge of BERT for Text Generation. Human-Like Autonomous Car-Following Model with Deep Reinforcement Learning. Value of structural health monitoring quantification in partially observable stochastic environment …

Low-Cost Transfer Learning of Face Tasks


Title	Low-Cost Transfer Learning of Face Tasks
Authors	Thrupthi Ann John, Isha Dua, Vineeth N Balasubramanian, C. V. Jawahar
Abstract	Do we know what the different filters of a face network represent? Can we use this filter information to train other tasks without transfer learning? For instance, can age, head pose, emotion and other face related tasks be learned from face recognition network without transfer learning? Understanding the role of these filters allows us to transfer knowledge across tasks and take advantage of large data sets in related tasks. Given a pretrained network, we can infer which tasks the network generalizes for and the best way to transfer the information to a new task.
Tasks	Face Recognition, Transfer Learning
Published	2019-01-09
URL	http://arxiv.org/abs/1901.02675v1
PDF	http://arxiv.org/pdf/1901.02675v1.pdf
PWC	https://paperswithcode.com/paper/low-cost-transfer-learning-of-face-tasks
Repo
Framework

Checking Chase Termination over Ontologies of Existential Rules with Equality


Title	Checking Chase Termination over Ontologies of Existential Rules with Equality
Authors	David Carral, Jacopo Urbani
Abstract	The chase is a sound and complete algorithm for conjunctive query answering over ontologies of existential rules with equality. To enable its effective use, we can apply acyclicity notions; that is, sufficient conditions that guarantee chase termination. Unfortunately, most of these notions have only been defined for existential rule sets without equality. A proposed solution to circumvent this issue is to treat equality as an ordinary predicate with an explicit axiomatisation. We empirically show that this solution is not efficient in practice and propose an alternative approach. More precisely, we show that, if the chase terminates for any equality axiomatisation of an ontology, then it terminates for the original ontology (which may contain equality). Therefore, one can apply existing acyclicity notions to check chase termination over an axiomatisation of an ontology and then use the original ontology for reasoning. We show that, in practice, doing so results in a more efficient reasoning procedure. Furthermore, we present equality model-faithful acyclicity, a general acyclicity notion that can be directly applied to ontologies with equality.
Tasks
Published	2019-11-25
URL	https://arxiv.org/abs/1911.10981v1
PDF	https://arxiv.org/pdf/1911.10981v1.pdf
PWC	https://paperswithcode.com/paper/checking-chase-termination-over-ontologies-of
Repo
Framework

Distilling the Knowledge of BERT for Text Generation


Title	Distilling the Knowledge of BERT for Text Generation
Authors	Yen-Chun Chen, Zhe Gan, Yu Cheng, Jingzhou Liu, Jingjing Liu
Abstract	Large-scale pre-trained language model, such as BERT, has recently achieved great success in a wide range of language understanding tasks. However, it remains an open question how to utilize BERT for text generation tasks. In this paper, we present a novel approach to addressing this challenge in a generic sequence-to-sequence (Seq2Seq) setting. We first propose a new task, Conditional Masked Language Modeling (C-MLM), to enable fine-tuning of BERT on target text-generation dataset. The fine-tuned BERT (i.e., teacher) is then exploited as extra supervision to improve conventional Seq2Seq models (i.e., student) for text generation. By leveraging BERT’s idiosyncratic bidirectional nature, distilling the knowledge learned from BERT can encourage auto-regressive Seq2Seq models to plan ahead, imposing global sequence-level supervision for coherent text generation. Experiments show that the proposed approach significantly outperforms strong baselines of Transformer on multiple text generation tasks, including machine translation (MT) and text summarization. Our proposed model also achieves new state-of-the-art results on the IWSLT German-English and English-Vietnamese MT datasets.
Tasks	Language Modelling, Machine Translation, Text Generation, Text Summarization
Published	2019-11-10
URL	https://arxiv.org/abs/1911.03829v1
PDF	https://arxiv.org/pdf/1911.03829v1.pdf
PWC	https://paperswithcode.com/paper/distilling-the-knowledge-of-bert-for-text-1
Repo
Framework

Human-Like Autonomous Car-Following Model with Deep Reinforcement Learning


Title	Human-Like Autonomous Car-Following Model with Deep Reinforcement Learning
Authors	Meixin Zhu, Xuesong Wang, Yinhai Wang
Abstract	This study proposes a framework for human-like autonomous car-following planning based on deep reinforcement learning (deep RL). Historical driving data are fed into a simulation environment where an RL agent learns from trial and error interactions based on a reward function that signals how much the agent deviates from the empirical data. Through these interactions, an optimal policy, or car-following model that maps in a human-like way from speed, relative speed between a lead and following vehicle, and inter-vehicle spacing to acceleration of a following vehicle is finally obtained. The model can be continuously updated when more data are fed in. Two thousand car-following periods extracted from the 2015 Shanghai Naturalistic Driving Study were used to train the model and compare its performance with that of traditional and recent data-driven car-following models. As shown by this study results, a deep deterministic policy gradient car-following model that uses disparity between simulated and observed speed as the reward function and considers a reaction delay of 1s, denoted as DDPGvRT, can reproduce human-like car-following behavior with higher accuracy than traditional and recent data-driven car-following models. Specifically, the DDPGvRT model has a spacing validation error of 18% and speed validation error of 5%, which are less than those of other models, including the intelligent driver model, models based on locally weighted regression, and conventional neural network-based models. Moreover, the DDPGvRT demonstrates good capability of generalization to various driving situations and can adapt to different drivers by continuously learning. This study demonstrates that reinforcement learning methodology can offer insight into driver behavior and can contribute to the development of human-like autonomous driving algorithms and traffic-flow models.
Tasks	Autonomous Driving
Published	2019-01-03
URL	http://arxiv.org/abs/1901.00569v1
PDF	http://arxiv.org/pdf/1901.00569v1.pdf
PWC	https://paperswithcode.com/paper/human-like-autonomous-car-following-model
Repo
Framework

Value of structural health monitoring quantification in partially observable stochastic environments


Title	Value of structural health monitoring quantification in partially observable stochastic environments
Authors	C. P. Andriotis, K. G. Papakonstantinou, E. N. Chatzi
Abstract	Sequential decision-making under uncertainty for optimal life-cycle control of deteriorating engineering systems and infrastructure entails two fundamental classes of decisions. The first class pertains to the various structural interventions, which can directly modify the existing properties of the system, while the second class refers to prescribing appropriate inspection and monitoring schemes, which are essential for updating our existing knowledge about the system states. The latter have to rely on quantifiable measures of efficiency, determined on the basis of objective criteria that, among others, consider the Value of Information (VoI) of different observational strategies, and the Value of Structural Health Monitoring (VoSHM) over the entire system life-cycle. In this work, we present general solutions for quantifying the VoI and VoSHM in partially observable stochastic domains, and although our definitions and methodology are general, we are particularly emphasizing and describing the role of Partially Observable Markov Decision Processes (POMDPs) in solving this problem, due to their advantageous theoretical and practical attributes in estimating arbitrarily well globally optimal policies. POMDP formulations are articulated for different structural environments having shared intervention actions but diversified inspection and monitoring options, thus enabling VoI and VoSHM estimation through their differentiated stochastic optimal control policies. POMDP solutions are derived using point-based solvers, which can efficiently approximate the POMDP value functions through Bellman backups at selected reachable points of the belief space. The suggested methodology is applied on stationary and non-stationary deteriorating environments, with both infinite and finite planning horizons, featuring single- or multi-component engineering systems.
Tasks	Decision Making, Decision Making Under Uncertainty
Published	2019-12-28
URL	https://arxiv.org/abs/1912.12534v1
PDF	https://arxiv.org/pdf/1912.12534v1.pdf
PWC	https://paperswithcode.com/paper/value-of-structural-health-monitoring
Repo
Framework

Uplift Modeling for Multiple Treatments with Cost Optimization


Title	Uplift Modeling for Multiple Treatments with Cost Optimization
Authors	Zhenyu Zhao, Totte Harinen
Abstract	Uplift modeling is an emerging machine learning approach for estimating the treatment effect at an individual or subgroup level. It can be used for optimizing the performance of interventions such as marketing campaigns and product designs. Uplift modeling can be used to estimate which users are likely to benefit from a treatment and then prioritize delivering or promoting the preferred experience to those users. An important but so far neglected use case for uplift modeling is an experiment with multiple treatment groups that have different costs, such as for example when different communication channels and promotion types are tested simultaneously. In this paper, we extend standard uplift models to support multiple treatment groups with different costs. We evaluate the performance of the proposed models using both synthetic and real data. We also describe a production implementation of the approach.
Tasks
Published	2019-08-14
URL	https://arxiv.org/abs/1908.05372v3
PDF	https://arxiv.org/pdf/1908.05372v3.pdf
PWC	https://paperswithcode.com/paper/uplift-modeling-for-multiple-treatments-with
Repo
Framework

Contextual Relabelling of Detected Objects


Title	Contextual Relabelling of Detected Objects
Authors	Faisal Alamri, Nicolas Pugeault
Abstract	Contextual information, such as the co-occurrence of objects and the spatial and relative size among objects provides deep and complex information about scenes. It also can play an important role in improving object detection. In this work, we present two contextual models (rescoring and re-labeling models) that leverage contextual information (16 contextual relationships are applied in this paper) to enhance the state-of-the-art RCNN-based object detection (Faster RCNN). We experimentally demonstrate that our models lead to enhancement in detection performance using the most common dataset used in this field (MSCOCO).
Tasks	Object Detection
Published	2019-06-06
URL	https://arxiv.org/abs/1906.02534v1
PDF	https://arxiv.org/pdf/1906.02534v1.pdf
PWC	https://paperswithcode.com/paper/contextual-relabelling-of-detected-objects
Repo
Framework

Topological Stability: a New Algorithm for Selecting The Nearest Neighbors in Non-Linear Dimensionality Reduction Techniques


Title	Topological Stability: a New Algorithm for Selecting The Nearest Neighbors in Non-Linear Dimensionality Reduction Techniques
Authors	Mohammed Elhenawy, Mahmoud Masoud, Sebastian Glaser, Andry Rakotonirainy
Abstract	In the machine learning field, dimensionality reduction is an important task. It mitigates the undesired properties of high-dimensional spaces to facilitate classification, compression, and visualization of high-dimensional data. During the last decade, researchers proposed many new (non-linear) techniques for dimensionality reduction. Most of these techniques are based on the intuition that data lies on or near a complex low-dimensional manifold that is embedded in the high-dimensional space. New techniques for dimensionality reduction aim at identifying and extracting the manifold from the high-dimensional space. Isomap is one of widely-used low-dimensional embedding methods, where geodesic distances on a weighted graph are incorporated with the classical scaling (metric multidimensional scaling). The Isomap chooses the nearest neighbours based on the distance only which causes bridges and topological instability. In this paper, we propose a new algorithm to choose the nearest neighbours to reduce the number of short-circuit errors and hence improves the topological stability. Because at any point on the manifold, that point and its nearest neighbours form a vector subspace and the orthogonal to that subspace is orthogonal to all vectors spans the vector subspace. The prposed algorithmuses the point itself and its two nearest neighbours to find the bases of the subspace and the orthogonal to that subspace which belongs to the orthogonal complementary subspace. The proposed algorithm then adds new points to the two nearest neighbours based on the distance and the angle between each new point and the orthogonal to the subspace. The superior performance of the new algorithm in choosing the nearest neighbours is confirmed through experimental work with several datasets.
Tasks	Dimensionality Reduction
Published	2019-11-13
URL	https://arxiv.org/abs/1911.05312v2
PDF	https://arxiv.org/pdf/1911.05312v2.pdf
PWC	https://paperswithcode.com/paper/topological-stability-guided-determination-of
Repo
Framework

Machine Learning for Paper Grammage Prediction Based on Sensor Measurements in Paper Mills


Title	Machine Learning for Paper Grammage Prediction Based on Sensor Measurements in Paper Mills
Authors	Hosny Abbas
Abstract	Automation is at the core of modern industry. It aims to increase production rates, decrease production costs, and reduce human intervention in order to avoid human mistakes and time delays during manufacturing. On the other hand, human assistance is usually required to customize products and reconfigure control systems through a special process interface called Human Machine Interface (HMI). Machine Learning (ML) algorithms can effectively be used to resolve this tradeoff between full automation and human assistance.This paper provides an example of the industrial application of ML algorithms to help human operators save their mental effort and avoid time delays and unintended mistakes for the sake of high production rates. Based on real-time sensor measurements, several ML algorithms have been tried to classify paper rolls according to paper grammage in a white paper mill. The performance evaluation shows that the AdaBoost algorithm is the best ML algorithm for this application with classification accuracy (CA), precision, and recall of 97.1%. The generalization of the proposed approach for achieving cost-effective mills construction will be the subject of our future research.
Tasks
Published	2019-09-25
URL	https://arxiv.org/abs/1910.06908v1
PDF	https://arxiv.org/pdf/1910.06908v1.pdf
PWC	https://paperswithcode.com/paper/machine-learning-for-paper-grammage
Repo
Framework

Progressive Stochastic Greedy Sparse Reconstruction and Support Selection


Title	Progressive Stochastic Greedy Sparse Reconstruction and Support Selection
Authors	Abolfazl Hashemi, Haris Vikalo, Gustavo de Veciana
Abstract	Sparse reconstruction and sparse support selection, i.e., the tasks of inferring an arbitrary $m$-dimensional sparse vector $\mathbf{x}$ having $k$ nonzero entries from $n$ measurements of linear combinations of its components, are often encountered in machine learning, computer vision, and signal processing. Existing greedy-based algorithms achieve optimal $n = \mathcal{O}(k\log\frac{m}{k})$ sampling complexity with computational complexity that is linear in the size of the data $m$ and cardinality constraint $k$. However, the ${\mathcal{O}}(mk)$ computational complexity is still prohibitive for large-scale datasets. In this paper, we present the first sparse support selection algorithm for arbitrary sparse vectors that achieves exact identification of the optimal subset from $n = \mathcal{O}(k\log\frac{m}{k})$ measurements with $\tilde{\mathcal{O}}(m)$ computational complexity. The proposed scheme utilizes the idea of randomly restricting search space of the greedy method in a progressive manner to reduce the computational cost while maintaining the same order of sampling complexity as the existing greedy schemes. Simulation results including an application of the algorithm to the task of column subset selection demonstrate efficacy of the proposed algorithm.
Tasks
Published	2019-07-22
URL	https://arxiv.org/abs/1907.09064v2
PDF	https://arxiv.org/pdf/1907.09064v2.pdf
PWC	https://paperswithcode.com/paper/stochastic-greedy-closing-the-optimality-gap
Repo
Framework

High-dimensional Bayesian optimization using low-dimensional feature spaces


Title	High-dimensional Bayesian optimization using low-dimensional feature spaces
Authors	Riccardo Moriconi, Marc P. Deisenroth, K. S. Sesh Kumar
Abstract	Bayesian optimization (BO) is a powerful approach for seeking the global optimum of expensive black-box functions and has proven successful for fine tuning hyper-parameters of machine learning models. However, in practice, BO is typically limited to optimizing 10–20 parameters. To scale BO to high dimensions, we normally make structural assumptions on the decomposition of the objective and/or exploit the intrinsic lower dimensionality of the problem, e.g. by using linear projections. The limitation of aforementioned approaches is the assumption of a linear subspace. We could achieve a higher compression rate with nonlinear projections, but learning these nonlinear embeddings typically requires much data. This contradicts the BO objective of a relatively small evaluation budget. To address this challenge, we propose to learn a low-dimensional feature space jointly with (a) the response surface and (b) a reconstruction mapping. Our approach allows for optimization of the acquisition function in the lower-dimensional subspace. We reconstruct the original parameter space from the lower-dimensional subspace for evaluating the black-box function. For meaningful exploration, we solve a constrained optimization problem.
Tasks	Dimensionality Reduction
Published	2019-02-27
URL	https://arxiv.org/abs/1902.10675v4
PDF	https://arxiv.org/pdf/1902.10675v4.pdf
PWC	https://paperswithcode.com/paper/high-dimensional-bayesian-optimization-with-1
Repo
Framework

Heterogeneous Relational Kernel Learning


Title	Heterogeneous Relational Kernel Learning
Authors	Andre T. Nguyen, Edward Raff
Abstract	Recent work has developed Bayesian methods for the automatic statistical analysis and description of single time series as well as of homogeneous sets of time series data. We extend prior work to create an interpretable kernel embedding for heterogeneous time series. Our method adds practically no computational cost compared to prior results by leveraging previously discarded intermediate results. We show the practical utility of our method by leveraging the learned embeddings for clustering, pattern discovery, and anomaly detection. These applications are beyond the ability of prior relational kernel learning approaches.
Tasks	Anomaly Detection, Time Series
Published	2019-08-24
URL	https://arxiv.org/abs/1908.09219v1
PDF	https://arxiv.org/pdf/1908.09219v1.pdf
PWC	https://paperswithcode.com/paper/heterogeneous-relational-kernel-learning
Repo
Framework

A Persona-based Multi-turn Conversation Model in an Adversarial Learning Framework


Title	A Persona-based Multi-turn Conversation Model in an Adversarial Learning Framework
Authors	Oluwatobi O. Olabiyi, Anish Khazane, Erik T. Mueller
Abstract	In this paper, we extend the persona-based sequence-to-sequence (Seq2Seq) neural network conversation model to multi-turn dialogue by modifying the state-of-the-art hredGAN architecture. To achieve this, we introduce an additional input modality into the encoder and decoder of hredGAN to capture other attributes such as speaker identity, location, sub-topics, and other external attributes that might be available from the corpus of human-to-human interactions. The resulting persona hredGAN ($phredGAN$) shows better performance than both the existing persona-based Seq2Seq and hredGAN models when those external attributes are available in a multi-turn dialogue corpus. This superiority is demonstrated on TV drama series with character consistency (such as Big Bang Theory and Friends) and customer service interaction datasets such as Ubuntu dialogue corpus in terms of perplexity, BLEU, ROUGE, and Distinct n-gram scores.
Tasks
Published	2019-04-29
URL	http://arxiv.org/abs/1905.01998v1
PDF	http://arxiv.org/pdf/1905.01998v1.pdf
PWC	https://paperswithcode.com/paper/190501998
Repo
Framework

A Comparison of Action Spaces for Learning Manipulation Tasks


Title	A Comparison of Action Spaces for Learning Manipulation Tasks
Authors	Patrick Varin, Lev Grossman, Scott Kuindersma
Abstract	Designing reinforcement learning (RL) problems that can produce delicate and precise manipulation policies requires careful choice of the reward function, state, and action spaces. Much prior work on applying RL to manipulation tasks has defined the action space in terms of direct joint torques or reference positions for a joint-space proportional derivative (PD) controller. In practice, it is often possible to add additional structure by taking advantage of model-based controllers that support both accurate positioning and control of the dynamic response of the manipulator. In this paper, we evaluate how the choice of action space for dynamic manipulation tasks affects the sample complexity as well as the final quality of learned policies. We compare learning performance across three tasks (peg insertion, hammering, and pushing), four action spaces (torque, joint PD, inverse dynamics, and impedance control), and using two modern reinforcement learning algorithms (Proximal Policy Optimization and Soft Actor-Critic). Our results lend support to the hypothesis that learning references for a task-space impedance controller significantly reduces the number of samples needed to achieve good performance across all tasks and algorithms.
Tasks
Published	2019-08-23
URL	https://arxiv.org/abs/1908.08659v1
PDF	https://arxiv.org/pdf/1908.08659v1.pdf
PWC	https://paperswithcode.com/paper/a-comparison-of-action-spaces-for-learning
Repo
Framework

DerainCycleGAN: An Attention-guided Unsupervised Benchmark for Single Image Deraining and Rainmaking


Title	DerainCycleGAN: An Attention-guided Unsupervised Benchmark for Single Image Deraining and Rainmaking
Authors	Yanyan Wei, Zhao Zhang, Jicong Fan, Yang Wang, Shuicheng Yan, Meng Wang
Abstract	Single image deraining (SID) is an important and challenging topic in emerging vision applications, and most of emerged deraining methods are supervised relying on the ground truth (i.e., paired images) in recent years. However, in practice it is rather common to have no un-paired images in real deraining task, in such cases how to remove the rain streaks in an unsupervised way will be a very challenging task due to lack of constraints between images and hence suffering from low-quality recovery results. In this paper, we explore the unsupervised SID task using unpaired data and propose a novel net called Attention-guided Deraining by Constrained CycleGAN (or shortly, DerainCycleGAN), which can fully utilize the constrained transfer learning abilitiy and circulatory structure of CycleGAN. Specifically, we design an unsu-pervised attention guided rain streak extractor (U-ARSE) that utilizes a memory to extract the rain streak masks with two constrained cycle-consistency branches jointly by paying attention to both the rainy and rain-free image domains. As a by-product, we also contribute a new paired rain image dataset called Rain200A, which is constructed by our network automatically. Compared with existing synthesis datasets, the rainy streaks in Rain200A contains more obvious and diverse shapes and directions. As a result, existing supervised methods trained on Rain200A can perform much better for processing real rainy images. Extensive experiments on synthesis and real datasets show that our net is superior to existing unsupervised deraining networks, and is also very competitive to other related supervised networks.
Tasks	Rain Removal, Single Image Deraining, Transfer Learning
Published	2019-12-15
URL	https://arxiv.org/abs/1912.07015v2
PDF	https://arxiv.org/pdf/1912.07015v2.pdf
PWC	https://paperswithcode.com/paper/deraincyclegan-an-attention-guided
Repo
Framework