Filters








25,339 Hits in 6.0 sec

Deep Explicit Duration Switching Models for Time Series [article]

Abdul Fatir Ansari, Konstantinos Benidis, Richard Kurle, Ali Caner Turkmen, Harold Soh, Alexander J. Smola, Yuyang Wang, Tim Januschowski
2021 arXiv   pre-print
We propose the Recurrent Explicit Duration Switching Dynamical System (RED-SDS), a flexible model that is capable of identifying both state- and time-dependent switching dynamics.  ...  State-dependent switching is enabled by a recurrent state-to-switch connection and an explicit duration count variable is used to improve the time-dependent switching behavior.  ...  explicit duration modeling for switches in a single non-linear model.  ... 
arXiv:2110.13878v1 fatcat:lbnouvsrlrdhje3mlxz7ynr4iu

Ensemble Deep Learning for Biomedical Time Series Classification

Lin-peng Jin, Jun Dong
2016 Computational Intelligence and Neuroscience  
for biomedical time series classification.  ...  Then, a new deep neural network-based ensemble method that integrates filtering views, local views, distorted views, explicit training, implicit training, subview prediction, and Simple Average is proposed  ...  biomedical time series classification.  ... 
doi:10.1155/2016/6212684 pmid:27725828 pmcid:PMC5048093 fatcat:x73wl3jayva2bnccfpnglnz5n4

Seasonal changes in basking shark vertical space use in the north-east Atlantic

P. D. Doherty, J. M. Baxter, B. J. Godley, R. T. Graham, G. Hall, J. Hall, L. A. Hawkes, S. M. Henderson, L. Johnson, C. Speedie, M. J. Witt
2019 Marine Biology  
The satellite tags provided depth and temperature data for a cumulative 4489 days (mean 140 ± 97 days per shark, range 10-292 days) to describe vertical space use and thermal range of basking sharks in  ...  These time-series data also revealed a conspicuous switch in vertical movements from a relatively uniform use of 50-250 m depths during the winter, to deep, rapid and repeated 'bounce' or 'yo-yo' dive  ...  Five of the 12 archival tags with attachment durations greater than 165 days were physically recovered, allowing for high-resolution time-series (depth and temperature data recorded at 10-15 s intervals  ... 
doi:10.1007/s00227-019-3565-6 fatcat:utzdf2nnergx7hqyxucgptii4y

Unified Treatment of Hidden Markov Switching Models [article]

Silvia Chiappa
2011 arXiv   pre-print
Many real-world problems encountered in several disciplines deal with the modeling of time-series containing different underlying dynamical regimes, for which probabilistic approaches are very often employed  ...  among models that were not observed in the past.  ...  Acknowledgments The author would like to thank the European Community for supporting her research through a Marie Curie Intra European Fellowship.  ... 
arXiv:1104.1992v1 fatcat:lycu63qlmbc7dkrfab346roys4

Asset Market Linkages in a Regime Switching Environment: Evidence from Commodity and Stock Markets in India

Shelly Singhal
2016 Business and Economics Research Journal  
model proposed by Hamilton (2005) which has the capability of capturing temporal asymmetries and nonlinear dynamics of time series.  ...  Time series models investigating the linkages between various asset markets (Commodity and Equity) in India have assumed a symmetric and linear relationship between them.  ...  The major advantage of Regime Switching model is that it does not necessitate the explicit specification of the time when the shift in economic states occurs.  ... 
doi:10.20409/berj.2016422336 fatcat:amzlojmu4zdr5coqhkewfrpbu4

PRRS Outbreak Prediction via Deep Switching Auto-Regressive Factorization Modeling [article]

Mohammadsadegh Shamsabardeh, Bahar Azari, Beatriz Martínez-López
2021 arXiv   pre-print
per farm time series prediction.  ...  We develop a hierarchical factorized deep generative model that approximates high dimensional data by a product between time-dependent weights and spatially dependent low dimensional factors to perform  ...  This dynamic produces a spatio-temporal time series from all farms over time that can be used for predication modeling in the next section (Sec. 3).  ... 
arXiv:2110.03147v1 fatcat:qmx3gr3qo5h6xayyzvmfiykhve

The Effect of Explicit Structure Encoding of Deep Neural Networks for Symbolic Music Generation [article]

Ke Chen, Weilin Zhang, Shlomo Dubnov, Gus Xia, Wei Li
2019 arXiv   pre-print
With recent breakthroughs in artificial neural networks, deep generative models have become one of the leading techniques for computational creativity.  ...  In particular, we explore the effect of explicit architectural encoding of musical structure via comparing two sequential generative models: LSTM (a type of RNN) and WaveNet (dilated temporal-CNN).  ...  The unit selection method [32] took a series of measures in music as a unit and used a deep structured semantic model (DSSM) with LSTM to predict future units, instead of directly generate essential  ... 
arXiv:1811.08380v3 fatcat:pjvkqwkzvfgytaenpszkur3v2y

Finite element analysis of impact between cricket ball and cantilever beam

Toh Yen Pang, Aleksandar Subic, Monir Takla
2011 Procedia Engineering  
It is envisaged that the model will be used for design customization and optimization of cricket helmets and other types of protective helmets.  ...  The friction effects between the cricket ball and the cantilever beam have been considered, and a friction coefficient of 0.2 was adopted for the model.  ...  The authors express their deep appreciation to technical staff of School of Aerospace, Mechanical and Manufacturing Engineering, RMIT University for assistance with test development and set-up.  ... 
doi:10.1016/j.proeng.2011.05.082 fatcat:f7uyswlihja5zm76yq4njb2oxq

Mechanisms of Thermohaline Mode Switching with Application to Warm Equable Climates

Rong Zhang, Michael Follows, John Marshall
2002 Journal of Climate  
By including convective adjustment modified to represent the localized nature of deep convection, the box model shows that a steady haline mode circulation is unstable.  ...  For certain ranges of freshwater forcing/ vertical diffusivity, a self-sustained oscillatory circulation is found in which haline-thermal mode switching occurs with a period of centuries to millennia.  ...  We thank Jochem Marotzke for helpful discussion on the box model. We also would like to thank the constructive comments from two reviewers of the paper.  ... 
doi:10.1175/1520-0442(2002)015<2056:motmsw>2.0.co;2 fatcat:nuet4ivgcjbetpqblgmfylvqva

Topological Analysis of Contradictions in Text

Xiangcheng Wu, Xi Niu, Ruhani Rahman
2022 Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval  
This study presents a topological approach to enhancing deep learning models in detecting contradictions in text.  ...  Following that, the topologically enhanced models are evaluated with different contradictions types, as well as different text genres.  ...  Each hole at any dimension has the characteristics of birth time, death time, and persistence duration.  ... 
doi:10.1145/3477495.3531881 fatcat:fw2yn2zicjecpm75euxzngbrx4

Generative Adversarial Networks (GAN) for the simulation of central-place foraging trajectories [article]

Amédée Roy, Sophie Lanco Bertrand, Ronan Fablet
2021 bioRxiv   pre-print
such as state-switching Hidden Markov Models (HMM). 3.  ...  Existing approaches to animal movement modeling mainly addressed the first objective and they are yet soon limited when used for simulation.  ...  We thank the Brazilian Ministry of Environment and Fernando de Noronha's firemen for the authorization and technical support to capture seabirds in Brazil.  ... 
doi:10.1101/2021.09.27.461940 fatcat:3qv6tfcaczccjcbri4nasmytz4

An application of upscaled optimal foraging theory using hidden Markov modelling: year-round behavioural variation in a large arctic herbivore

Larissa T. Beumer, Jennifer Pohle, Niels M. Schmidt, Marianna Chimienti, Jean-Pierre Desforges, Lars H. Hansen, Roland Langrock, Stine Højlund Pedersen, Mikkel Stelvig, Floris M. van Beest
2020 Movement Ecology  
The overall reduction in activity likely reflects higher time requirements for rumination in response to the reduction of forage quality (supporting an energy intake maximisation strategy).  ...  To relate behavioural variation to environmental conditions, we considered a wide range of spatially and/or temporally explicit covariates in the HMMs.  ...  Acknowledgements We thank the Greenland Ecosystem Monitoring Programme for access to ecosystem data, and Aarhus University, Denmark, for providing access to and logistics at Zackenberg.  ... 
doi:10.1186/s40462-020-00213-x pmid:32518653 pmcid:PMC7275509 fatcat:pb62tsc5z5bbxaj2afx6uaj7py

The costs of taking it slowly: Fast and slow movement timing in older age

Ralf Th. Krampe, Mihalis Doumas, Ann Lavrysen, Michael Rapp
2010 Psychology and Aging  
Dual-task costs for both cognitive and timing performances were pronounced at slower tapping tempos, an effect exacerbated in older adults.  ...  Our findings implicate attention and working memory processes as critical components of slow movement timing and sources of specific challenges thereof for older adults.  ...  Such models consider timing as a property emerging from biomechanical constraints (Schöner, 2002) without assuming explicit cognitive representations for target durations.  ... 
doi:10.1037/a0020090 pmid:21186918 fatcat:mxff2szy7zf43h66dgm4uvq2w4

Deep Learning for Time Series Forecasting: Tutorial and Literature Survey

Konstantinos Benidis, Syama Sundar Rangapuram, Valentin Flunkert, Yuyang Wang, Danielle Maddix, Caner Turkmen, Jan Gasthaus, Michael Bohlke-Schneider, David Salinas, Lorenzo Stella, François-Xavier Aubet, Laurent Callot (+1 others)
2022 ACM Computing Surveys  
Deep learning based forecasting methods have become the methods of choice in many applications of time series prediction or forecasting often outperforming other approaches.  ...  In this article we provide an introduction and overview of the field: We present important building blocks for deep forecasting in some depth; using these building blocks, we then survey the breadth of  ...  The recurrent explicit duration switching dynamical system (RED-SDS) is a lexible model that is capable of identifying both state-and time-dependent switching dynamics of a time series.  ... 
doi:10.1145/3533382 fatcat:l46f34dbp5fdpawbpnoiippf6q

Connecting spatial and temporal scales of tropical precipitation in observations and the MetUM-GA6

Gill M. Martin, Nicholas P. Klingaman, Aurel F. Moise
2016 Geoscientific Model Development Discussions  
In contrast, switching off the deep convection parametrization, albeit at an unrealistic resolution for resolving tropical convection, results in very persistent (for limited periods), but very sporadic  ...  We find that the behaviour of the deep convection parametrization in this model on the native grid and time-step is largely independent of the grid-box size and time-step length over which it operates.  ...  The authors are grateful to the model development teams at the Met Office, who ran the MetUM-GA6 simulations as part of the Global Atmosphere 6.0 development process, and to Alison Stirling for helpful  ... 
doi:10.5194/gmd-2016-202 fatcat:t5ivb3lfkfhlvn5o6t2splnumq
« Previous Showing results 1 — 15 out of 25,339 results