Filters








138 Hits in 6.1 sec

Simplified decision making in the belief space using belief sparsification [article]

Khen Elimelech, Vadim Indelman
2021 arXiv   pre-print
In this work, we introduce a new and efficient solution approach for the problem of decision making under uncertainty, which can be formulated as decision making in a belief space, over a possibly high-dimensional  ...  state space.  ...  Andrej Kitanov from the Faculty of Aerospace Engineering at the Technion -Israel Institute of Technology, for insightful discussions concerning Section 3.3.2, and his assistance with implementing the simulation  ... 
arXiv:1909.00885v4 fatcat:gur4uqitdng7zimopzenivv5re

Combinatorial Bayesian Optimization using the Graph Cartesian Product [article]

Changyong Oh and Jakub M. Tomczak and Efstratios Gavves and Max Welling
2019 arXiv   pre-print
Moreover, using the Horseshoe prior for the scale parameter in the ARD diffusion kernel results in an effective variable selection procedure, making COMBO suitable for high dimensional problems.  ...  COMBO outperforms consistently the latest state-of-the-art while maintaining computational and statistical efficiency.  ...  The Horseshoe prior encourages sparsity, and, thus, enables variable selection, which, in turn, makes COMBO statistically scalable to high dimensional problems.  ... 
arXiv:1902.00448v2 fatcat:ogxvc42acja3zgxja62p6xnayu

Topological Information-Theoretic Belief Space Planning with Optimality Guarantees [article]

Andrej Kitanov, Vadim Indelman
2019 arXiv   pre-print
Determining a globally optimal solution of belief space planning (BSP) in high-dimensional state spaces is computationally expensive, as it involves belief propagation and objective function evaluation  ...  In this paper we contribute to this body of work a novel method for efficiently determining error bounds of t-bsp, thereby providing global optimality guarantees or uncertainty margin of its solution.  ...  CONCLUSIONS This paper provides theoretical foundations for action consistent topological belief space planning (t-BSP). t-BSP enables efficient decision making in high dimensional state spaces by considering  ... 
arXiv:1903.00927v1 fatcat:22eioiahwng7rn7kib7clc2smy

Adaptive Planning for Markov Decision Processes with Uncertain Transition Models via Incremental Feature Dependency Discovery [chapter]

N. Kemal Ure, Alborz Geramifard, Girish Chowdhary, Jonathan P. How
2012 Lecture Notes in Computer Science  
Solving large scale sequential decision making problems without prior knowledge of the state transition model is a key problem in the planning literature.  ...  Theoretical analysis and numerical simulations in domains with state space sizes varying from thousands to millions are used to illustrate the benefit of using iFDD for incrementally building transition  ...  Problem Definition The problem of sequential decision making under uncertainty is formulated as an MDP, which is defined as the tuple M = S, A, P a ss , R a ss , γ , where S is the discrete state space  ... 
doi:10.1007/978-3-642-33486-3_7 fatcat:i7qkpdix3fh6vaxwvxr2gq4xcu

A-optimal design of experiments for infinite-dimensional Bayesian linear inverse problems with regularized ℓ_0-sparsification [article]

Alen Alexanderian, Noemi Petra, Georg Stadler, Omar Ghattas
2014 arXiv   pre-print
We present an efficient method for computing A-optimal experimental designs for infinite-dimensional Bayesian linear inverse problems governed by partial differential equations (PDEs).  ...  Computing optimal experimental designs (OEDs) is particularly challenging for inverse problems governed by computationally expensive PDE models with infinite-dimensional (or, after discretization, high-dimensional  ...  We would like to thank James Martin for providing us with an implementation of the algorithm in [11] , which was used to compute the application of the inverse square root of the mass matrix.  ... 
arXiv:1308.4084v2 fatcat:pzllmxlnn5aunmy6m3cvqkuhqu

Deep Bayesian Quadrature Policy Optimization [article]

Akella Ravi Tej, Kamyar Azizzadenesheli, Mohammad Ghavamzadeh, Anima Anandkumar, Yisong Yue
2020 arXiv   pre-print
In this work, we propose deep Bayesian quadrature policy gradient (DBQPG), a computationally efficient high-dimensional generalization of Bayesian quadrature, for policy gradient estimation.  ...  for several deep policy gradient algorithms, and, (iii) the uncertainty in gradient estimation that can be incorporated to further improve the performance.  ...  Azizzadenesheli is supported in part by Raytheon and Amazon Web Service. A.  ... 
arXiv:2006.15637v3 fatcat:udecffhidrgdlfk36x2zbopyxq

Sparsifying to optimize over multiple information sources: an augmented Gaussian process based algorithm

Antonio Candelieri, Francesco Archetti
2021 Structural And Multidisciplinary Optimization  
This approximation is estimated through a model discrepancy measure and the prediction uncertainty of the GPs.  ...  challenging task known as multi-information source optimization (MISO), where each source has a different cost and the level of approximation (aka fidelity) of each source can change over the search space  ...  Acknowledgements We greatly acknowledge the DEMS Data Science Lab, Department of Economics Management and Statistics (DEMS), University of Milano-Bicocca, for supporting this work by providing computational  ... 
doi:10.1007/s00158-021-02882-7 fatcat:mr2cfcz4h5cgthohk4o6hwojsm

Combining Model and Parameter Uncertainty in Bayesian Neural Networks [article]

Aliaksandr Hubin, Geir Storvik
2019 arXiv   pre-print
In this paper we introduce the concept of model uncertainty in BNNs and hence make inference in the joint space of models and parameters.  ...  However so far there have been no scalable techniques capable of combining both model (structural) and parameter uncertainty.  ...  high dimensional problems.  ... 
arXiv:1903.07594v3 fatcat:ecnc5lcsuvai3o4l5fg34iv3ja

Compressive Information Extraction: A Dynamical Systems Approach1

Mario Sznaier
2012 IFAC Proceedings Volumes  
time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information.  ...  THIS PAGE Unclassified REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188 The public reporting burden for this collection of information is estimated to average 1 hour per response, including the  ...  Notably, merely postulating the existence of a dynamically sparse underlying model led to efficient, scalable algorithms for information extraction.  ... 
doi:10.3182/20120711-3-be-2027.00430 fatcat:4xxsi2orf5bdlpddghh3fpl65e

Bayesian Optimization of Combinatorial Structures [article]

Ricardo Baptista, Matthias Poloczek
2018 arXiv   pre-print
The combinatorial explosion of the search space and costly evaluations pose challenges for current techniques in discrete optimization and machine learning, and critically require new algorithmic ideas  ...  Our acquisition function pioneers the use of semidefinite programming to achieve efficiency and scalability.  ...  ACKNOWLEDGEMENT This work has been supported in part by the Air Force Office of Scientific Research (AFOSR) MURI on "Managing multiple information sources of multi-physics systems," Program Officer Dr.  ... 
arXiv:1806.08838v2 fatcat:m55pha7xtfflnnsgpx7yvjxape

Challenges and Opportunities in Approximate Bayesian Deep Learning for Intelligent IoT Systems [article]

Meet P. Vadera, Benjamin M. Marlin
2021 arXiv   pre-print
However, the computational requirements of existing approximate Bayesian inference methods can make them ill-suited for deployment in intelligent IoT systems that include lower-powered edge devices.  ...  Approximate Bayesian deep learning methods hold significant promise for addressing several issues that occur when deploying deep learning components in intelligent systems, including mitigating the occurrence  ...  Acknowledgement This work was partially supported by the US Army Research Laboratory under cooperative agreement W911NF-17-2-0196.  ... 
arXiv:2112.01675v1 fatcat:okknsw5gifhl7ghzt4cnuchuje

Automated Machine Learning on Big Data using Stochastic Algorithm Tuning [article]

Thomas Nickson, Michael A Osborne, Steven Reece, Stephen J Roberts
2014 arXiv   pre-print
We provide a comprehensive benchmarking of possible sparsification strategies for Bayesian optimisation, concluding that a Nystrom approximation offers the best scaling and performance for real tasks.  ...  Against this background, Bayesian optimisation is finding increasing use in automating parameter tuning, making ML algorithms accessible even to non-experts.  ...  For very high dimensional spaces it would seem impractical to use a length-scale much below the size of the space.  ... 
arXiv:1407.7969v1 fatcat:nf723w7cfbbwffpimt2jvv5hku

Randomized Algorithms for Scientific Computing (RASC) [article]

Aydin Buluc, Tamara G. Kolda, Stefan M. Wild, Mihai Anitescu, Anthony DeGennaro, John Jakeman, Chandrika Kamath, Ramakrishnan Kannan, Miles E. Lopes, Per-Gunnar Martinsson, Kary Myers, Jelani Nelson (+7 others)
2021 arXiv   pre-print
Randomized algorithms have propelled advances in artificial intelligence and represent a foundational research area in advancing AI for Science.  ...  challenges of complexity, robustness, and scalability.  ...  This capability is critical in contexts characterized by high-complexity dynamics and high-dimensional decision spaces.  ... 
arXiv:2104.11079v2 fatcat:qwwowtufzvbfjaiotx733eexxe

Communication-Efficient and Distributed Learning Over Wireless Networks: Principles and Applications [article]

Jihong Park, Sumudu Samarakoon, Anis Elgabli, Joongheon Kim, Mehdi Bennis, Seong-Lyun Kim, Mérouane Debbah
2020 arXiv   pre-print
To achieve this goal, it is essential to cater for high ML inference accuracy at scale under time-varying channel and network dynamics, by continuously exchanging fresh data and ML model updates in a distributed  ...  By imbuing intelligence into the network edge, edge nodes can proactively carry out decision-making, and thereby react to local environmental changes and disturbances while experiencing zero communication  ...  CONCLUDING REMARKS Imbuing intelligence into edge devices enables low-latency and scalable decision-making at the network edge in 5G communication systems and beyond.  ... 
arXiv:2008.02608v1 fatcat:luuo5pja5zfihhpybger6tuqrq

Deep Learning on Computational-Resource-Limited Platforms: A Survey

Chunlei Chen, Peng Zhang, Huixiang Zhang, Jiangyan Dai, Yugen Yi, Huihui Zhang, Yonghui Zhang
2020 Mobile Information Systems  
Subsequently, we explore the underlying reasons for the high computational overhead of DL through reviewing the fundamental concepts including capacity, generalization, and backpropagation of a neural  ...  In pursuant to these solutions, we identify challenges to be addressed.  ...  Tsushiya et al. design a solid-state ionic device to address decision-making issues like the multiarmed bandit problem (MBPs). is device opens a way to achieve decision-making through motion of ions, which  ... 
doi:10.1155/2020/8454327 fatcat:pocvmihd7jcw7ig544s2yduovu
« Previous Showing results 1 — 15 out of 138 results