May 5, 2019

2965 words 14 mins read

Paper Group ANR 567

dMath: A Scalable Linear Algebra and Math Library for Heterogeneous GP-GPU Architectures. Improved Dynamic Regret for Non-degenerate Functions. Grammar rules for the isiZulu complex verb. End-to-End Subtitle Detection and Recognition for Videos in East Asian Languages via CNN Ensemble with Near-Human-Level Performance. Spatio-Colour Asplünd ‘s Metr …

dMath: A Scalable Linear Algebra and Math Library for Heterogeneous GP-GPU Architectures


Title	dMath: A Scalable Linear Algebra and Math Library for Heterogeneous GP-GPU Architectures
Authors	Steven Eliuk, Cameron Upright, Anthony Skjellum
Abstract	A new scalable parallel math library, dMath, is presented in this paper that demonstrates leading scaling when using intranode, or internode, hybrid-parallelism for deep-learning. dMath provides easy-to-use distributed base primitives and a variety of domain-specific algorithms. These include matrix multiplication, convolutions, and others allowing for rapid development of highly scalable applications, including Deep Neural Networks (DNN), whereas previously one was restricted to libraries that provided effective primitives for only a single GPU, like Nvidia cublas and cudnn or DNN primitives from Nervana neon framework. Development of HPC software is difficult, labor-intensive work, requiring a unique skill set. dMath allows a wide range of developers to utilize parallel and distributed hardware easily. One contribution of this approach is that data is stored persistently on the GPU hardware, avoiding costly transfers between host and device. Advanced memory management techniques are utilized, including caching of transferred data and memory reuse through pooling. A key contribution of dMath is that it delivers performance, portability, and productivity to its specific domain of support. It enables algorithm and application programmers to quickly solve problems without managing the significant complexity associated with multi-level parallelism.
Tasks
Published	2016-04-05
URL	http://arxiv.org/abs/1604.01416v1
PDF	http://arxiv.org/pdf/1604.01416v1.pdf
PWC	https://paperswithcode.com/paper/dmath-a-scalable-linear-algebra-and-math
Repo
Framework

Improved Dynamic Regret for Non-degenerate Functions


Title	Improved Dynamic Regret for Non-degenerate Functions
Authors	Lijun Zhang, Tianbao Yang, Jinfeng Yi, Rong Jin, Zhi-Hua Zhou
Abstract	Recently, there has been a growing research interest in the analysis of dynamic regret, which measures the performance of an online learner against a sequence of local minimizers. By exploiting the strong convexity, previous studies have shown that the dynamic regret can be upper bounded by the path-length of the comparator sequence. In this paper, we illustrate that the dynamic regret can be further improved by allowing the learner to query the gradient of the function multiple times, and meanwhile the strong convexity can be weakened to other non-degenerate conditions. Specifically, we introduce the squared path-length, which could be much smaller than the path-length, as a new regularity of the comparator sequence. When multiple gradients are accessible to the learner, we first demonstrate that the dynamic regret of strongly convex functions can be upper bounded by the minimum of the path-length and the squared path-length. We then extend our theoretical guarantee to functions that are semi-strongly convex or self-concordant. To the best of our knowledge, this is the first time that semi-strong convexity and self-concordance are utilized to tighten the dynamic regret.
Tasks
Published	2016-08-13
URL	http://arxiv.org/abs/1608.03933v3
PDF	http://arxiv.org/pdf/1608.03933v3.pdf
PWC	https://paperswithcode.com/paper/improved-dynamic-regret-for-non-degenerate
Repo
Framework

Grammar rules for the isiZulu complex verb


Title	Grammar rules for the isiZulu complex verb
Authors	C. Maria Keet, Langa Khumalo
Abstract	The isiZulu verb is known for its morphological complexity, which is a subject for on-going linguistics research, as well as for prospects of computational use, such as controlled natural language interfaces, machine translation, and spellcheckers. To this end, we seek to answer the question as to what the precise grammar rules for the isiZulu complex verb are (and, by extension, the Bantu verb morphology). To this end, we iteratively specify the grammar as a Context Free Grammar, and evaluate it computationally. The grammar presented in this paper covers the subject and object concords, negation, present tense, aspect, mood, and the causative, applicative, stative, and the reciprocal verbal extensions, politeness, the wh-question modifiers, and aspect doubling, ensuring their correct order as they appear in verbs. The grammar conforms to specification.
Tasks	Machine Translation
Published	2016-12-20
URL	http://arxiv.org/abs/1612.06581v1
PDF	http://arxiv.org/pdf/1612.06581v1.pdf
PWC	https://paperswithcode.com/paper/grammar-rules-for-the-isizulu-complex-verb
Repo
Framework

End-to-End Subtitle Detection and Recognition for Videos in East Asian Languages via CNN Ensemble with Near-Human-Level Performance


Title	End-to-End Subtitle Detection and Recognition for Videos in East Asian Languages via CNN Ensemble with Near-Human-Level Performance
Authors	Yan Xu, Siyuan Shan, Ziming Qiu, Zhipeng Jia, Zhengyang Shen, Yipei Wang, Mengfei Shi, Eric I-Chao Chang
Abstract	In this paper, we propose an innovative end-to-end subtitle detection and recognition system for videos in East Asian languages. Our end-to-end system consists of multiple stages. Subtitles are firstly detected by a novel image operator based on the sequence information of consecutive video frames. Then, an ensemble of Convolutional Neural Networks (CNNs) trained on synthetic data is adopted for detecting and recognizing East Asian characters. Finally, a dynamic programming approach leveraging language models is applied to constitute results of the entire body of text lines. The proposed system achieves average end-to-end accuracies of 98.2% and 98.3% on 40 videos in Simplified Chinese and 40 videos in Traditional Chinese respectively, which is a significant outperformance of other existing methods. The near-perfect accuracy of our system dramatically narrows the gap between human cognitive ability and state-of-the-art algorithms used for such a task.
Tasks
Published	2016-11-18
URL	http://arxiv.org/abs/1611.06159v1
PDF	http://arxiv.org/pdf/1611.06159v1.pdf
PWC	https://paperswithcode.com/paper/end-to-end-subtitle-detection-and-recognition
Repo
Framework

Spatio-Colour Asplünd ‘s Metric and Logarithmic Image Processing for Colour Images (LIPC)


Title	Spatio-Colour Asplünd ‘s Metric and Logarithmic Image Processing for Colour Images (LIPC)
Authors	Guillaume Noyel, Michel Jourlin
Abstract	Aspl"und ‘s metric, which is useful for pattern matching, consists in a double-sided probing, i.e. the over-graph and the sub-graph of a function are probed jointly. This paper extends the Aspl"und ‘s metric we previously defined for colour and multivariate images using a marginal approach (i.e. component by component) to the first spatio-colour Aspl"und ‘s metric based on the vectorial colour LIP model (LIPC). LIPC is a non-linear model with operations between colour images which are consistent with the human visual system. The defined colour metric is insensitive to lighting variations and a variant which is robust to noise is used for colour pattern matching.
Tasks
Published	2016-08-31
URL	http://arxiv.org/abs/1608.08831v2
PDF	http://arxiv.org/pdf/1608.08831v2.pdf
PWC	https://paperswithcode.com/paper/spatio-colour-asplund-s-metric-and
Repo
Framework

Laplacian regularized low rank subspace clustering


Title	Laplacian regularized low rank subspace clustering
Authors	Yu Song, Yiquan Wu
Abstract	The problem of fitting a union of subspaces to a collection of data points drawn from multiple subspaces is considered in this paper. In the traditional low rank representation model, the dictionary used to represent the data points is chosen as the data points themselves and thus the dictionary is corrupted with noise. This problem is solved in the low rank subspace clustering model which decomposes the corrupted data matrix as the sum of a clean and self-expressive dictionary plus a matrix of noise and gross errors. Also, the clustering results of the low rank representation model can be enhanced by using a graph of data similarity. This model is called Laplacian regularized low rank representation model with a graph regularization term added to the objective function. Inspired from the above two ideas, in this paper a Laplacian regularized low rank subspace clustering model is proposed. This model uses a clean dictionary to represent the data points and a graph regularization term is also incorporated in the objective function. Experimental results show that, compared with the traditional low rank representation model, low rank subspace clustering model and several other state-of-the-art subspace clustering model, the model proposed in this paper can get better subspace clustering results with lower clustering error.
Tasks
Published	2016-10-24
URL	http://arxiv.org/abs/1610.07488v2
PDF	http://arxiv.org/pdf/1610.07488v2.pdf
PWC	https://paperswithcode.com/paper/laplacian-regularized-low-rank-subspace
Repo
Framework

Detection and Tracking of Liquids with Fully Convolutional Networks


Title	Detection and Tracking of Liquids with Fully Convolutional Networks
Authors	Connor Schenck, Dieter Fox
Abstract	Recent advances in AI and robotics have claimed many incredible results with deep learning, yet no work to date has applied deep learning to the problem of liquid perception and reasoning. In this paper, we apply fully-convolutional deep neural networks to the tasks of detecting and tracking liquids. We evaluate three models: a single-frame network, multi-frame network, and a LSTM recurrent network. Our results show that the best liquid detection results are achieved when aggregating data over multiple frames, in contrast to standard image segmentation. They also show that the LSTM network outperforms the other two in both tasks. This suggests that LSTM-based neural networks have the potential to be a key component for enabling robots to handle liquids using robust, closed-loop controllers.
Tasks	Semantic Segmentation
Published	2016-06-20
URL	http://arxiv.org/abs/1606.06266v1
PDF	http://arxiv.org/pdf/1606.06266v1.pdf
PWC	https://paperswithcode.com/paper/detection-and-tracking-of-liquids-with-fully
Repo
Framework

Heuristic with elements of tabu search for Truck and Trailer Routing Problem


Title	Heuristic with elements of tabu search for Truck and Trailer Routing Problem
Authors	Ivan S. Grechikhin
Abstract	Vehicle Routing Problem is a well-known problem in logistics and transportation, and the variety of such problems is explained by the fact that it occurs in many real-life situations. It is an NP-hard combinatorial optimization problem and finding an exact optimal solution is practically impossible. In this work, Site-Dependent Truck and Trailer Routing Problem with hard and soft Time Windows and Split Deliveries is considered (SDTTRPTWSD). In this article, we develop a heuristic with the elements of Tabu Search for solving SDTTRPTWSD. The heuristic uses the concept of neighborhoods and visits infeasible solutions during the search. A greedy heuristic is applied to construct an initial solution.
Tasks	Combinatorial Optimization
Published	2016-09-29
URL	http://arxiv.org/abs/1609.09253v1
PDF	http://arxiv.org/pdf/1609.09253v1.pdf
PWC	https://paperswithcode.com/paper/heuristic-with-elements-of-tabu-search-for
Repo
Framework

Shuffle and Learn: Unsupervised Learning using Temporal Order Verification


Title	Shuffle and Learn: Unsupervised Learning using Temporal Order Verification
Authors	Ishan Misra, C. Lawrence Zitnick, Martial Hebert
Abstract	In this paper, we present an approach for learning a visual representation from the raw spatiotemporal signals in videos. Our representation is learned without supervision from semantic labels. We formulate our method as an unsupervised sequential verification task, i.e., we determine whether a sequence of frames from a video is in the correct temporal order. With this simple task and no semantic labels, we learn a powerful visual representation using a Convolutional Neural Network (CNN). The representation contains complementary information to that learned from supervised image datasets like ImageNet. Qualitative results show that our method captures information that is temporally varying, such as human pose. When used as pre-training for action recognition, our method gives significant gains over learning without external data on benchmark datasets like UCF101 and HMDB51. To demonstrate its sensitivity to human pose, we show results for pose estimation on the FLIC and MPII datasets that are competitive, or better than approaches using significantly more supervision. Our method can be combined with supervised representations to provide an additional boost in accuracy.
Tasks	Pose Estimation, Temporal Action Localization
Published	2016-03-28
URL	http://arxiv.org/abs/1603.08561v2
PDF	http://arxiv.org/pdf/1603.08561v2.pdf
PWC	https://paperswithcode.com/paper/shuffle-and-learn-unsupervised-learning-using
Repo
Framework

A Generic Bet-and-run Strategy for Speeding Up Traveling Salesperson and Minimum Vertex Cover


Title	A Generic Bet-and-run Strategy for Speeding Up Traveling Salesperson and Minimum Vertex Cover
Authors	Tobias Friedrich, Timo Kötzing, Markus Wagner
Abstract	A common strategy for improving optimization algorithms is to restart the algorithm when it is believed to be trapped in an inferior part of the search space. However, while specific restart strategies have been developed for specific problems (and specific algorithms), restarts are typically not regarded as a general tool to speed up an optimization algorithm. In fact, many optimization algorithms do not employ restarts at all. Recently, “bet-and-run” was introduced in the context of mixed-integer programming, where first a number of short runs with randomized initial conditions is made, and then the most promising run of these is continued. In this article, we consider two classical NP-complete combinatorial optimization problems, traveling salesperson and minimum vertex cover, and study the effectiveness of different bet-and-run strategies. In particular, our restart strategies do not take any problem knowledge into account, nor are tailored to the optimization algorithm. Therefore, they can be used off-the-shelf. We observe that state-of-the-art solvers for these problems can benefit significantly from restarts on standard benchmark instances.
Tasks	Combinatorial Optimization
Published	2016-09-13
URL	http://arxiv.org/abs/1609.03993v1
PDF	http://arxiv.org/pdf/1609.03993v1.pdf
PWC	https://paperswithcode.com/paper/a-generic-bet-and-run-strategy-for-speeding
Repo
Framework

A Large Dataset of Object Scans


Title	A Large Dataset of Object Scans
Authors	Sungjoon Choi, Qian-Yi Zhou, Stephen Miller, Vladlen Koltun
Abstract	We have created a dataset of more than ten thousand 3D scans of real objects. To create the dataset, we recruited 70 operators, equipped them with consumer-grade mobile 3D scanning setups, and paid them to scan objects in their environments. The operators scanned objects of their choosing, outside the laboratory and without direct supervision by computer vision professionals. The result is a large and diverse collection of object scans: from shoes, mugs, and toys to grand pianos, construction vehicles, and large outdoor sculptures. We worked with an attorney to ensure that data acquisition did not violate privacy constraints. The acquired data was irrevocably placed in the public domain and is available freely at http://redwood-data.org/3dscan .
Tasks
Published	2016-02-08
URL	http://arxiv.org/abs/1602.02481v3
PDF	http://arxiv.org/pdf/1602.02481v3.pdf
PWC	https://paperswithcode.com/paper/a-large-dataset-of-object-scans
Repo
Framework

A Bayesian Nonparametric Approach for Estimating Individualized Treatment-Response Curves


Title	A Bayesian Nonparametric Approach for Estimating Individualized Treatment-Response Curves
Authors	Yanbo Xu, Yanxun Xu, Suchi Saria
Abstract	We study the problem of estimating the continuous response over time to interventions using observational time series—a retrospective dataset where the policy by which the data are generated is unknown to the learner. We are motivated by applications where response varies by individuals and therefore, estimating responses at the individual-level is valuable for personalizing decision-making. We refer to this as the problem of estimating individualized treatment response (ITR) curves. In statistics, G-computation formula (Robins, 1986) has been commonly used for estimating treatment responses from observational data containing sequential treatment assignments. However, past studies have focused predominantly on obtaining point-in-time estimates at the population level. We leverage the G-computation formula and develop a novel Bayesian nonparametric (BNP) method that can flexibly model functional data and provide posterior inference over the treatment response curves at both the individual and population level. On a challenging dataset containing time series from patients admitted to a hospital, we estimate responses to treatments used in managing kidney function and show that the resulting fits are more accurate than alternative approaches. Accurate methods for obtaining ITRs from observational data can dramatically accelerate the pace at which personalized treatment plans become possible.
Tasks	Decision Making, Time Series
Published	2016-08-18
URL	http://arxiv.org/abs/1608.05182v2
PDF	http://arxiv.org/pdf/1608.05182v2.pdf
PWC	https://paperswithcode.com/paper/a-bayesian-nonparametric-approach-for
Repo
Framework

Death and Suicide in Universal Artificial Intelligence


Title	Death and Suicide in Universal Artificial Intelligence
Authors	Jarryd Martin, Tom Everitt, Marcus Hutter
Abstract	Reinforcement learning (RL) is a general paradigm for studying intelligent behaviour, with applications ranging from artificial intelligence to psychology and economics. AIXI is a universal solution to the RL problem; it can learn any computable environment. A technical subtlety of AIXI is that it is defined using a mixture over semimeasures that need not sum to 1, rather than over proper probability measures. In this work we argue that the shortfall of a semimeasure can naturally be interpreted as the agent’s estimate of the probability of its death. We formally define death for generally intelligent agents like AIXI, and prove a number of related theorems about their behaviour. Notable discoveries include that agent behaviour can change radically under positive linear transformations of the reward signal (from suicidal to dogmatically self-preserving), and that the agent’s posterior belief that it will survive increases over time.
Tasks
Published	2016-06-02
URL	http://arxiv.org/abs/1606.00652v1
PDF	http://arxiv.org/pdf/1606.00652v1.pdf
PWC	https://paperswithcode.com/paper/death-and-suicide-in-universal-artificial
Repo
Framework

Beyond Fine Tuning: A Modular Approach to Learning on Small Data


Title	Beyond Fine Tuning: A Modular Approach to Learning on Small Data
Authors	Ark Anderson, Kyle Shaffer, Artem Yankov, Court D. Corley, Nathan O. Hodas
Abstract	In this paper we present a technique to train neural network models on small amounts of data. Current methods for training neural networks on small amounts of rich data typically rely on strategies such as fine-tuning a pre-trained neural network or the use of domain-specific hand-engineered features. Here we take the approach of treating network layers, or entire networks, as modules and combine pre-trained modules with untrained modules, to learn the shift in distributions between data sets. The central impact of using a modular approach comes from adding new representations to a network, as opposed to replacing representations via fine-tuning. Using this technique, we are able surpass results using standard fine-tuning transfer learning approaches, and we are also able to significantly increase performance over such approaches when using smaller amounts of data.
Tasks	Transfer Learning
Published	2016-11-06
URL	http://arxiv.org/abs/1611.01714v1
PDF	http://arxiv.org/pdf/1611.01714v1.pdf
PWC	https://paperswithcode.com/paper/beyond-fine-tuning-a-modular-approach-to
Repo
Framework

Darknet and Deepnet Mining for Proactive Cybersecurity Threat Intelligence


Title	Darknet and Deepnet Mining for Proactive Cybersecurity Threat Intelligence
Authors	Eric Nunes, Ahmad Diab, Andrew Gunn, Ericsson Marin, Vineet Mishra, Vivin Paliath, John Robertson, Jana Shakarian, Amanda Thart, Paulo Shakarian
Abstract	In this paper, we present an operational system for cyber threat intelligence gathering from various social platforms on the Internet particularly sites on the darknet and deepnet. We focus our attention to collecting information from hacker forum discussions and marketplaces offering products and services focusing on malicious hacking. We have developed an operational system for obtaining information from these sites for the purposes of identifying emerging cyber threats. Currently, this system collects on average 305 high-quality cyber threat warnings each week. These threat warnings include information on newly developed malware and exploits that have not yet been deployed in a cyber-attack. This provides a significant service to cyber-defenders. The system is significantly augmented through the use of various data mining and machine learning techniques. With the use of machine learning models, we are able to recall 92% of products in marketplaces and 80% of discussions on forums relating to malicious hacking with high precision. We perform preliminary analysis on the data collected, demonstrating its application to aid a security expert for better threat analysis.
Tasks
Published	2016-07-28
URL	http://arxiv.org/abs/1607.08583v1
PDF	http://arxiv.org/pdf/1607.08583v1.pdf
PWC	https://paperswithcode.com/paper/darknet-and-deepnet-mining-for-proactive
Repo
Framework