Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userA First Step to Convolutive Sparse Representation
132
0
0.0
(
0
)
Added by
Hamed Firouzi
Publication date
2008
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
In this paper an extension of the sparse decomposition problem is considered and an algorithm for solving it is presented. In this extension, it is known that one of the shift
rate research
Read More
Freight carriers rely on tactical plans to satisfy demand in a cost-effective way. For computational tractability in real large-scale settings, such plans are typically computed by solving deterministic and cyclic formulations. An important input is the periodic demand, i.e., the demand that is expected to repeat in each period of the planning horizon. Motivated by the discrepancy between time series forecasts of demand in each period and the periodic demand, Laage et al. (2021) recently introduced the Periodic Demand Estimation (PDE) problem and showed that it has a high value. However, they made strong assumptions on the solution space so that the problem could be solved by enumeration. In this paper we significantly extend their work. We propose a new PDE formulation that relaxes the strong assumptions on the solution space. We solve large instances of this formulation with a two-step heuristic. The first step reduces the dimension of the feasible space by performing clustering of commodities based on instance-specific information about demand and supply interactions. The formulation along with the first step allow to solve the problem in a second step by either metaheuristics or the state-of-the-art black-box optimization solver NOMAD. In an extensive empirical study using real data from the Canadian National Railway Company, we show that our methodology produces high quality solutions and outperforms existing ones.
Cover song identification represents a challenging task in the field of Music Information Retrieval (MIR) due to complex musical variations between query tracks and cov
Studies of sparse representation of deterministic signals have been well developed. Amongst there exists one called adaptive Fourier decomposition (AFD) established through adaptive selections of the parameters defining a Takenaka-Malmquist system in one-complex variable. The AFD type algorithms give rise to sparse representations of signals of finite energy. The multivariate generalization of AFD is one called pre-orthogonal AFD (POAFD), the latter being established with the context Hilbert space possessing a dictionary. The purpose of the present study is to generalize both AFD and POAFD to random signals. We work on two types of random signals. One is those expressible as the sum of a deterministic signal with an error term such as a white noise; and the other is, in general, as mixture of several classes of random signals obeying certain distributive law. In the first part of the paper we develop an AFD type sparse representation for one-dimensional random signals by making use analysis of one complex variable. In the second part, without complex analysis, we treat multivariate random signals in the context of stochastic Hilbert space with a dictionary. Like in the deterministic signal case the established random sparse representations are powerful tools in practical signal analysis.
Enabling users to interactively navigate through different viewpoints of a static scene is a new interesting functionality in 3D streaming systems. While it opens exciting perspectives towards rich multimedia applications, it requires the design of novel representations and coding techniques in order to solve the new challenges imposed by interactive navigation. Interactivity clearly brings new design constraints: the encoder is unaware of the exact decoding process, while the decoder has to reconstruct information from incomplete subsets of data since the server can generally not transmit images for all possible viewpoints due to resource constrains. In this paper, we propose a novel multiview data representation that permits to satisfy bandwidth and storage constraints in an interactive multiview streaming system. In particular, we partition the multiview navigation domain into segments, each of which is described by a reference image and some auxiliary information. The auxiliary information enables the client to recreate any viewpoint in the navigation segment via view synthesis. The decoder is then able to navigate freely in the segment without further data request to the server; it requests additional data only when it moves to a different segment. We discuss the benefits of this novel representation in interactive navigation systems and further propose a method to optimize the partitioning of the navigation domain into independent segments, under bandwidth and storage constraints. Experimental results confirm the potential of the proposed representation; namely, our system leads to similar compression performance as classical inter-view coding, while it provides the high level of flexibility that is required for interactive streaming. Hence, our new framework represents a promising solution for 3D data representation in novel interactive multimedia services.
As an important component of multimedia analysis tasks, audio classification aims to discriminate between different audio signal types and has received intensive attention due to its wide applications. Generally speaking, the raw signal can be transformed into various representations (such as Short Time Fourier Transform and Mel Frequency Cepstral Coefficients), and information implied in different representations can be complementary. Ensembling the models trained on different representations can greatly boost the classification performance, however, making inference using a large number of models is cumbersome and computationally expensive. In this paper, we propose a novel end-to-end collaborative learning framework for the audio classification task. The framework takes multiple representations as the input to train the models in parallel. The complementary information provided by different representations is shared by knowledge distillation. Consequently, the performance of each model can be significantly promoted without increasing the computational overhead in the inference stage. Extensive experimental results demonstrate that the proposed approach can improve the classification performance and achieve state-of-the-art results on both acoustic scene classification tasks and general audio tagging tasks.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
