检索结果-内蒙古大学图书馆

Pose networks Unveiled: Bridging the Gap for Monocular Depth Perception

学校读者我要写书评

暂无评论

Pose Networks Unveiled: Bridging the Gap for Monocular Depth...

IEEE International Symposium on Mixed and Augmented Reality Workshops (ISMARW)

作者： Yazan Dayoub Andrey V. Savchenko Ilya Makarov HSE University Laboratory of Algorithms and Technologies for Network Analysis Sber AI Lab HSE University ISP RAS Research Center for Trusted Artificial Intelligence Artificial Intelligence Research Institute (AIRI) HSE University

ISBN: (数字)9798331506919

ISBN: (纸本)9798331506926

Depth estimation is essential in Augmented Reality applications, enabling realistic object placement, scene understanding, spatial mapping, interaction, and environment awareness. This paper proposes a method to enhance depth model performance without increasing inference costs by improving the pose network in a self-supervised learning setup. In particular, we enrich spatial information in the pose network by incorporating features from different scales and normalized coordinates. It is experimentally shown on the KITTI dataset that our approach achieves a 2-7% improvement in the abs_rel metric when compared to baseline techniques.

关键词： Accuracy Costs Computational modeling Pose estimation Self-supervised learning Cameras Encoding Augmented reality Load modeling

On a numerically stable algorithm for the analysis of Generalized Volterra lattice

学校读者我要写书评

暂无评论

On a numerically stable algorithm for the analysis of Genera...

作者： Kalyagin, Valery Sokolov, Maxim Laboratory of Algorithms and Technologies for Network Analysis Department of Applied Mathematics and Informatics The National Research University Higher School of Economics 136 Rodionova Street Nizhny Novgorod Russian Federation 603093 Russia MERA Ltd and Department of Applied Mathematics and Informatics National Research University Higher School of Economics Bol'shaya Pechërskaya Street 25/12 Nizhny Novgorod Russian Federation 603155 Russia

ISBN: (纸本)9781461455738

Volterra or Langmuir lattice is the dynamical model where the interaction of particle with the nearest neighbors is taken into account. It is known since J. Moser that the analysis of the Volterra lattice is related with isospectral deformation of a tridiagonal Jacobi operator. The main numerical problem in this setting is the inverse spectral problem for this Jacobi operator. Generalized Volterra lattice is a dynamical model where the interaction of particle with some fixed number of neighbors is taken into account. This model is a particular case of the discrete KP equation. The analysis of discrete KP equation is related with a class of Hessenberg operators. In this chapter we propose and study a stable algorithm for the numerical solution of the inverse spectral problem for the band Hessenberg operator with application to the analysis of generalized Volterra lattice. © Springer Science+Business Media New York 2013.

关键词： Inverse problems

Hyperplanes Avoiding Problem and Integer Points Counting in Polyhedra

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Dakhno, Grigorii Gribanov, Dmitry Kasianov, Nikita Kats, Anastasiia Kupavskii, Andrey Kuz’min, Nikita Laboratory of Discrete and Combinatorial Optimization Moscow Institute of Physics and Technology Institutsky lane 9 Moscow region Dolgoprudny141700 Russia Laboratory of Algorithms and Technologies for Network Analysis HSE University 136 Rodionova Ulitsa Nizhny Novgorod603093 Russia

In our work, we consider the problem of computing a vector 6 x ∈ n of minimum k·kp-norm such that ax 6= a0, for any vector (a, a0) from a given finite set A ⊆ n. In other words, we search for a vector of minimum norm that avoids a given finite set of hyperplanes, which is natural to call as the Hyperplanes Avoiding Problem. This problem naturally appears as a subproblem in Barvinok-type algorithms for counting integer points in polyhedra. More precisely, it appears when one needs to evaluate certain rational generating functions in an avoidable critical point. We show that: 1. With respect to k·k1, the problem admits a feasible solution x with kxk1 ≤ (m + n)/2, where m = |A|, and show that such solution can be constructed by a deterministic polynomial-time algorithm with O(n · m) operations. Moreover, this inequality is the best possible. This is a significant improvement over the previous randomized algorithm, which computes x with a guaranty kxk1 ≤ n · m. The original approach of A. Barvinok can guarantee only kxk1 = O((n · m)n). To prove this result, we use a newly established algorithmic variant of the Combinatorial Nullstellensatz;2. The problem is NP-hard with respect to any norm k·kp, for p ∈ (R≥1 ∪{∞});3. As an application, we show that the problem to count integer points in a polytope P = {x ∈ Rn : Ax ≤ b}, for given A ∈ m×n and b ∈ Qm, can be solved by an algorithm with O(ν2·n3·∆3) operations, where ν is the maximum size of a normal fan triangulation of P, and ∆ is the maximum value of rank-order subdeterminants of A. It refines the previous state-of-the-art O(ν2 · n4 · ∆3)-time algorithm. As a further application, it provides a refined complexity bound for the counting problem in polyhedra of bounded codimension. More specifically, it improves the computational complexity bound for counting the number of solutions in the Unbounded Subset-Sum problem. Copyright © 2024, The Authors. All rights reserved.

关键词： Rational functions

Neural networks compression for language modeling

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Grachev, Artem M. Ignatov, Dmitry I. Savchenko, Andrey V. Samsung RandD Institute Rus Moscow Russia National Research University Higher School of Economics Moscow Russia National Research University Higher School of Economics Laboratory of Algorithms and Technologies for Network Analysis Nizhny Novgorod Russia

In this paper, we consider several compression techniques for the language modeling problem based on recurrent neural networks (RNNs). It is known that conventional RNNs, e.g, LSTM-based networks in language modeling, are characterized with either high space complex- ity or substantial inference time. This problem is especially crucial for mobile applications, in which the constant interaction with the remote server is inappropriate. By using the Penn Treebank (PTB) dataset we compare pruning, quantization, low-rank factorization, tensor train decomposition for LSTM networks in terms of model size and suitability for fast inference.62M45, 68T50 Copyright © 2017, The Authors. All rights reserved.

关键词： Factorization

Optimal statistical decision for Gaussian graphical model selection

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Kalyagin, Valery A. Koldanov, Alexander P. Koldanov, Petr A. Pardalos, Panos M. Laboratory of Algorithms and Technologies for Network Analysis National Research University Higher School of Economics Bolshaya Pecherskaya 25/12 Nizhny Novgorod603155 Russia University of Florida ISE Department GainesvilleFL32611 United States

Gaussian graphical model is a graphical representation of the dependence structure for a Gaussian random vector. It is recognized as a powerful tool in different applied fields such as bioinformatics, error-control codes, speech language, information retrieval and others. Gaussian graphical model selection is a statistical problem to identify the Gaussian graphical model from a sample of a given size. Different approaches for Gaussian graphical model selection are suggested in the literature. One of them is based on considering the family of individual conditional independence tests. The application of this approach leads to the construction of a variety of multiple testing statistical procedures for Gaussian graphical model selection. An important characteristic of these procedures is its error rate for a given sample size. In existing literature great attention is paid to the control of error rates for incorrect edge inclusion (Type I error). However, in graphical model selection it is also important to take into account error rates for incorrect edge exclusion (Type II error). To deal with this issue we consider the graphical model selection problem in the framework of the multiple decision theory. The quality of statistical procedures is measured by a risk function with additive losses. Additive losses allow both types of errors to be taken into account. We construct the tests of a Neyman structure for individual hypotheses and combine them to obtain a multiple decision statistical procedure. We show that the obtained procedure is optimal in the sense that it minimizes the linear combination of expected numbers of Type I and Type II errors in the class of unbiased multiple decision procedures.62-09, 62H15, 62C05 Copyright © 2017, The Authors. All rights reserved.

关键词： Decision making

Sequential analysis with Specified Confidence Level and Adaptive Convolutional Neural networks in Image Recognition

学校读者我要写书评

暂无评论

Sequential Analysis with Specified Confidence Level and Adap...

International Joint Conference on Neural networks (IJCNN)

作者： Andrey Savchenko St. Petersburg Department of Steklov Institute of Mathematics Samsung-PDMI Joint AI Center Nizhny Novgorod Russia Laboratory of Algorithms and Technologies for Network Analysis National Research University Higher School of Economics Nizhny Novgorod Russia

ISBN: (数字)9781728169262

ISBN: (纸本)9781728169279

In this paper the problem of high computational complexity of deep convolutional nets in image recognition is considered. An existing framework of adaptive neural networks is extended by appending the separate classifier to intermediate layers. The hierarchical representations of the input image are sequentially analyzed. If the first classifier returns rather high confidence score, the inference process will be terminated. Otherwise, the inference to the next intermediate layer with attached classifier is continued until the reliable solution is obtained or the penultimate layer is reached. The thresholds for classifier scores at each layer are automatically chosen based on the Benjamini-Hochberg multiple comparisons for a specified confidence level. Experimental study for both pre-trained and fine-tuned deep convolutional neural networks demonstrates that the proposed approach reduces the running time by up to 1.7 times without significant accuracy degradation. Moreover, the larger is the training sample, the more noticeable is the gain in performance.

关键词： Training Image recognition Adaptive systems Feature extraction Computer architecture Neural networks Adaptation models

Compression of recurrent neural networks for efficient language modeling

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Grachev, Artem M. Ignatov, Dmitry I. Savchenko, Andrey V. Samsung R&D Institute Moscow Russia National Research University Higher School of Economics Moscow Russia National Research University Higher School of Economics Laboratory of Algorithms and Technologies for Network Analysis Nizhny Novgorod Russia

Recurrent neural networks have proved to be an effective method for statistical language modeling. However, in practice their memory and run-time complexity are usually too large to be implemented in real-time offline mobile applications. In this paper we consider several compression techniques for recurrent neural networks including Long-Short Term Memory models. We make particular attention to the high-dimensional output problem caused by the very large vocabulary size. We focus on effective compression methods in the context of their exploitation on devices: pruning, quantization, and matrix decomposition approaches (low-rank factorization and tensor train decomposition, in particular). For each model we investigate the trade-off between its size, suitability for fast inference and perplexity. We propose a general pipeline for applying the most suitable methods to compress recurrent neural networks for language modeling. It has been shown in the experimental study with the Penn Treebank (PTB) dataset that the most efficient results in terms of speed and compression-perplexity balance are obtained by matrix decomposition techniques. Copyright © 2019, The Authors. All rights reserved.

关键词： Factorization

Event Recognition with Automatic Album Detection based on Sequential Processing, Neural Attention and Image Captioning

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Savchenko, Andrey V. National Research University Higher School of Economics Laboratory of Algorithms and Technologies for Network Analysis Nizhny Novgorod Russia St. Petersburg Department of Steklov Institute of Mathematics SamsungPDMI Joint AI Center St. Petersburg Russia

In this paper a new formulation of event recognition task is examined: it is required to predict event categories in a gallery of images, for which albums (groups of photos corresponding to a single event) are unknown. We propose the novel two-stage approach. At first, features are extracted in each photo using the pre-trained convolutional neural network. These features are classified individually. The scores of the classifier are used to group sequential photos into several clusters. Finally, the features of photos in each group are aggregated into a single descriptor using neural attention mechanism. This algorithm is optionally extended to improve the accuracy for classification of each image in an album. In contrast to conventional fine-tuning of convolutional neural networks (CNN) we proposed to use image captioning, i.e., generative model that converts images to textual descriptions. They are one-hot encoded and summarized into sparse feature vector suitable for learning of arbitrary classifier. Experimental study with Photo Event Collection and Multi-Label Cu-ration of Flickr Events Dataset demonstrates that our approach is 9-20% more accurate than event recognition on single photos. Moreover, proposed method has 13-16% lower error rate than classification of groups of photos obtained with hierarchical clustering. It is experimentally shown that the image captions trained on Conceptual Captions dataset can be classified more accurately than the features from object detector, though they both are obviously not as rich as the CNN-based features. However, it is possible to combine our approach with conventional CNNs in an ensemble to provide the state-of-the-art results for several event *** Codes 68T10 (Primary) Copyright © 2019, The Authors. All rights reserved.

关键词： Object detection

Event Recognition with Automatic Album Detection based on Sequential Grouping of Confidence Scores and Neural Attention

学校读者我要写书评

暂无评论

Event Recognition with Automatic Album Detection based on Se...

International Joint Conference on Neural networks (IJCNN)

ISBN: (数字)9781728169262

ISBN: (纸本)9781728169279

In this paper a new formulation of event recognition task is examined: it is required to predict event categories given a gallery of images, for which albums (groups of photos corresponding to a single event) are unknown. The novel two-stage approach is proposed. At first, features are extracted in each photo using the pre-trained convolutional neural network (CNN). These features are classified individually. The normalized scores of the classifier are used to group sequential photos into several clusters. Finally, the features of photos in each group are aggregated into a single descriptor using neural attention mechanism. This algorithm is implemented in Android mobile application. Experimental study with features extracted by contemporary convolutional neural networks including EfficientNets for Photo Event Collection and Multi-Label Curation of Flickr Events Dataset demonstrates that the proposed approach is 9-23% more accurate than conventional event recognition on single photos. Moreover, proposed method has 13-16% lower error rate when compared to classification of groups of photos obtained with hierarchical clustering of CNN-based embeddings.

关键词： Feature extraction Task analysis Image recognition Neural networks Hidden Markov models Mobile handsets