Sökning: id:"swepub:oai:DiVA.org:kth-316354" >
Monte Carlo Filteri...
Monte Carlo Filtering Objectives
-
- Chen, Shuangshuang, 1992- (författare)
- KTH,Robotik, perception och lärande, RPL,AI Lab, Volvo Car Corporation,Kungliga Tekniska Högskolan (KTH),Royal Institute of Technology (KTH),Volvo
-
- Ding, Sihao (författare)
- AI Lab, Volvo Car Corporation,Volvo
-
- Karayiannidis, Yiannis, 1980 (författare)
- Chalmers University of Technology, Gothenburg, Sweden,Chalmers tekniska högskola
-
visa fler...
-
- Björkman, Mårten, 1970- (författare)
- KTH,Robotik, perception och lärande, RPL,Kungliga Tekniska Högskolan (KTH),Royal Institute of Technology (KTH)
-
visa färre...
-
(creator_code:org_t)
- International Joint Conferences on Artificial Intelligence, 2021
- 2021
- Engelska.
-
Ingår i: IJCAI International Joint Conference on Artificial Intelligence. - : International Joint Conferences on Artificial Intelligence. - 1045-0823. ; , s. 2256-2262
- Relaterad länk:
-
https://urn.kb.se/re...
-
visa fler...
-
https://research.cha...
-
visa färre...
Abstract
Ämnesord
Stäng
- Learning generative models and inferring latent trajectories have shown to be challenging for time series due to the intractable marginal likelihoods of flexible generative models. It can be addressed by surrogate objectives for optimization. We propose Monte Carlo filtering objectives (MCFOs), a family of variational objectives for jointly learning parametric generative models and amortized adaptive importance proposals of time series. MCFOs extend the choices of likelihood estimators beyond Sequential Monte Carlo in state-of-the-art objectives, possess important properties revealing the factors for the tightness of objectives, and allow for less biased and variant gradient estimates. We demonstrate that the proposed MCFOs and gradient estimations lead to efficient and stable model learning, and learned generative models well explain data and importance proposals are more sample efficient on various kinds of time series data.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Bioinformatik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Bioinformatics (hsv//eng)
- NATURVETENSKAP -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Probability Theory and Statistics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
Nyckelord
- Artificial intelligence
- Learning systems
- Monte Carlo methods
- Generative model
- Gradient estimates
- Marginal likelihood
- Monte Carlo filtering
- Objective estimations
- Optimisations
- Property
- Sequential Monte Carlo
- State of the art
- Times series
- Time series
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)
Hitta via bibliotek
Till lärosätets databas