1,714 Hits in 6.9 sec

Dual Online Stein Variational Inference for Control and Dynamics [article]

Lucas Barcelos, Alexander Lambert, Rafael Oliveira, Paulo Borges, Byron Boots, Fabio Ramos
2021 arXiv   pre-print
Model predictive control (MPC) schemes have a proven track record for delivering aggressive and robust performance in many challenging control tasks, coping with nonlinear system dynamics, constraints,  ...  In this paper, we devise an implicit variational inference algorithm able to estimate distributions over model parameters and control inputs on-the-fly.  ...  METHOD In this section, we present our approach for joint inference over control and model parameters for MPC. We call this method Dual Stein Variational Inference MPC, or DuSt-MPC for conciseness.  ... 
arXiv:2103.12890v1 fatcat:jmixv6hzjrellovglp53khtmqi

Dual Online Stein Variational Inference for Control and Dynamics

Lucas Barcelos, Alexander Lambert, Rafael Oliveira, Paulo Borges, Byron Boots, Fabio Ramos
2021 Robotics: Science and Systems XVII   unpublished
Model predictive control (MPC) schemes have a proven track record for delivering aggressive and robust performance in many challenging control tasks, coping with nonlinear system dynamics, constraints,  ...  In this paper, we devise an implicit variational inference algorithm able to estimate distributions over model parameters and control inputs on-the-fly.  ...  The resulting framework, Stein-Variational Model Predictive Control (SVMPC), adapts the particle distribution in an online fashion.  ... 
doi:10.15607/rss.2021.xvii.068 fatcat:f7akjt7ct5ep3djocfprdamh54

Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models [article]

Fan Bao, Kun Xu, Chongxuan Li, Lanqing Hong, Jun Zhu, Bo Zhang
2021 arXiv   pre-print
With a minimal model assumption, VaES and VaGES can be applied to the kernelized Stein discrepancy (KSD) and score matching (SM)-based methods to learn EBLVMs.  ...  This paper presents variational estimates of the score function and its gradient with respect to the model parameters in a general EBLVM, referred to as VaES and VaGES respectively.  ...  ., Doubly Dual Embedding (Dai et al., 2019a), Adversarial Dynam- ics Embedding (Dai et al., 2019b) and Fenchel Mini-Max Learning (Tao et al., 2019), and extending theses methods for learning EBLVMs  ... 
arXiv:2010.08258v3 fatcat:2dtbg7eimrc3nfdwcacemmu6wq

Robust Control Under Uncertainty via Bounded Rationality and Differential Privacy [article]

Vincent Pacelli, Anirudha Majumdar
2021 arXiv   pre-print
of state information used for control (i.e., to impose bounded rationality).  ...  This paper addresses these challenges using the theory of differential privacy, which allows us to (i) design controllers with bounded sensitivity to errors in state estimates, and (ii) bound the amount  ...  methods based on importance sampling, Stein-variational gradient descent, and Q-learning.  ... 
arXiv:2109.08262v1 fatcat:s5br47bgnjbjdjd6sbhwinbzhe

Robot Learning from Randomized Simulations: A Review [article]

Fabio Muratore, Fabio Ramos, Greg Turk, Wenhao Yu, Michael Gienger, Jan Peters
2021 arXiv   pre-print
This raises the question of how simulators can be modified to facilitate learning robot control policies and overcome the mismatch between simulation and reality, often called the 'reality gap'.  ...  We provide a comprehensive review of sim-to-real research for robotics, focusing on a technique named 'domain randomization' which is a method for learning from randomized simulations.  ...  Taking this idea one step further, we could additionally include the domain parameters for inference, and thereby establish connections to dual control (Feldbaum, 1960; Wittenmark, 1995) .  ... 
arXiv:2111.00956v1 fatcat:khywklph2jfkbd2at3fvnszf2e

Table of Contents

2020 2020 IEEE Symposium Series on Computational Intelligence (SSCI)  
Time Series Forecasting with Temporal Attention Convolutional Neural Networks Leonardos Pantiskas, Kees Verstoep and Henri Bal .......... 1687 Online System Identification for Nonlinear Uncertain Dynamical  ...  : Learning from Neural Architecture Search Bas van Stein, Hao Wang and Thomas Back .......... 1341 CIDUE2: Learning in Non-Stationary and Uncertain Environments/Dynamic Single and Multi-Objective Optimization  ... 
doi:10.1109/ssci47803.2020.9308155 fatcat:hyargfnk4vevpnooatlovxm4li

Deciphering microbial interactions in synthetic human gut microbiome communities

Ophelia S Venturelli, Alex C Carr, Garth Fisher, Ryan H Hsu, Rebecca Lau, Benjamin P Bowen, Susan Hromada, Trent Northen, Adam P Arkin
2018 Molecular Systems Biology  
The inferred ecological network exhibits a high proportion of negative and frequent positive interactions. Ecological drivers and responsive recipient species were discovered in the network.  ...  We show that pairwise interactions are major drivers of multispecies community dynamics, as opposed to higher-order interactions.  ...  Acknowledgements We would like to thank Nicholas Justice, Michael Fischbach, and Justin Sonnenburg for helpful discussions. We would like to thank Suzanne Kosina for help with metabolomics methods.  ... 
doi:10.15252/msb.20178157 pmid:29930200 pmcid:PMC6011841 fatcat:qaj3hds5fvbgfdworkyzzxhk3y

Trust-Region Variational Inference with Gaussian Mixture Models [article]

Oleg Arenz, Mingjun Zhong, Gerhard Neumann
2020 arXiv   pre-print
Variational inference approximates such distributions by tractable models that can be subsequently used for approximate inference.  ...  We demonstrate on several domains that we can learn approximations of complex, multimodal distributions with a quality that is unmet by previous variational inference methods, and that the GMM approximation  ...  Stein variational gradient descent (SVGD) (Liu and Wang, 2016 ) is a sampling method that closely relates to variational inference.  ... 
arXiv:1907.04710v2 fatcat:iue475fldvdc7l4gxsqgmkpasu

Epigenetic Regulation of Gene Expression in Cancer: Techniques, Resources, and Analysis [article]

Luciane Kagohara, Genevieve Stein-O'Brien, Dylan Kelley, Emily Flam, Heather Wick, Ludmila Danilova, Hariharan Easwaran, Alexander Favorov, Jiang Qian, Daria Gaykalova, Elana Fertig
2017 bioRxiv   pre-print
Therefore, this review describes these high-throughput measurement technologies, public domain databases for high-throughput epigenetic data in tumors and model systems, and bioinformatics algorithms for  ...  This deeper understanding is essential to future studies that will precisely infer patients prognosis and select patients who will be responsive to emerging epigenetic therapies.  ...  Robust standards for quality control and preprocessing were adopted by ENCODE as gold standards for all chromatin based analyses [113] .  ... 
doi:10.1101/114025 fatcat:gbju3lqdxzho3ckdbwtxw5vizy

Dual control of fault intersections on stop-start rupture in the 2016 Central Italy seismic sequence

R.J. Walters, L.C. Gregory, L.N.J. Wedmore, T.J. Craig, K. McCaffrey, M. Wilkinson, J. Chen, Z. Li, J.R. Elliott, H. Goodall, F. Iezzi, F. Livio (+3 others)
2018 Earth and Planetary Science Letters  
We thank Simon Matthias, Stefan Nielsen for helpful discussions and comments which improved this manuscript, as well as Andy Nicol and three anonymous reviewers for their helpful and constructive reviews  ...  We thank Lauro Chiaraluce for sharing his relocated aftershock catalogue, INGV and Nicola D'Agostino for sharing the GNSS coseismic displacement data, and Anna Maria Blumetti, Luca Guerrieri, Pio di Manna  ...  we exploit seis-mological and field observations, as well as geodetic data, to image the kinematics of the sequence, and to understand structural and dynamic controls on its evolution.  ... 
doi:10.1016/j.epsl.2018.07.043 fatcat:epe7vzdhy5ee7cpi2dycc6lepy

Sequential, Bayesian Geostatistics: A Principled Method for Large Data Sets

Dan Cornford, Lehel Csato, Manfred Opper
2005 Geographical Analysis  
In this paper, we present a Bayesian method for estimating the posterior mean and covariance structures of a Gaussian random field using a sequential estimation algorithm.  ...  We show the application of the sequential, sparse Bayesian estimation in Gaussian random field models and discuss its merits and draw-backs.  ...  We are grateful to the conference organisers for inviting submission of this paper, and to the reviewers for their helpful comments and suggestions which have greatly improved the readability of the paper  ... 
doi:10.1111/j.1538-4632.2005.00635.x fatcat:fc3o6nhj5nfqnlghgdaw4kkt4q

Planning under Uncertainty to Goal Distributions [article]

Adam Conkey, Tucker Hermans
2022 arXiv   pre-print
We build on previous results in the literature by formally framing our approach as an instance of planning as inference.  ...  Goals for planning problems are typically conceived of as subsets of the state space.  ...  APPENDIX A VARIATIONAL INFERENCE FOR PLANNING AS INFERENCE We described at a high-level in Sec. V-A how variational inference techniques are often used to solve planning as inference problems.  ... 
arXiv:2011.04782v2 fatcat:eb3jvsc4hbachjb57ijovhf7sy

Assessment of Brain Monoaminergic Signaling Through Mathematical Modeling of Skin Conductance Response [chapter]

Saa Brankovi
2012 Neuroscience - Dealing With Frontiers  
The twist is that we firstly examine signal processing and control system aspects of a relatively defined neurobiological mechanism.  ...  Then, informed with the system and the signals' properties we are able to infer about the underlying neurochemical activity of the involved neurotransmitter systems.  ...  allow inferring about tonic and phasic functions of the brain monoaminergic signaling and about the central neural events -neural input for the SCR system. 1 Mathematical model of the SCR process: A  ... 
doi:10.5772/34164 fatcat:i5urjg3xsvf5nfsofr45jvkuxq

Bayesian Inference for Brain Activity from Functional Magnetic Resonance Imaging Collected at Two Spatial Resolutions [article]

Andrew S. Whiteman, Andreas J. Bartsch, Jian Kang, Timothy D. Johnson
2021 arXiv   pre-print
Neuroradiologists and neurosurgeons increasingly opt to use functional magnetic resonance imaging (fMRI) to map functionally relevant brain regions for noninvasive presurgical planning and intraoperative  ...  In practice, fMRI scans can be collected at multiple spatial resolutions, and it is of interest to make more accurate inference on brain activity by combining data with different resolutions.  ...  This work was partially supported by NIH R01 DA048993 (Kang and Johnson).  ... 
arXiv:2103.13131v1 fatcat:54dg2y5t3nd3tm2ifeaxrr2khq

Capitalizing on Content: Information Adoption in Two Online communities

Stephanie Watts, Wei Zhang
2008 Journal of the AIS  
It also suggests numerous opportunities for future research and potential ways that online communities might improve their information sharing.  ...  and the extent to which information-seeking members are actively searching for on-topic information to satisfy their specific information needs.  ...  As we will discuss, these inferences suggest important and interesting directions for future studies.  ... 
doi:10.17705/1jais.00149 fatcat:6ej4qcgouzcdjlesvqiixsgpwy
« Previous Showing results 1 — 15 out of 1,714 results