Filters








5,789 Hits in 6.0 sec

Bayesian Optimization for Policy Search in High-Dimensional Systems via Automatic Domain Selection [article]

Lukas P. Fröhlich, Edgar D. Klenske, Christian G. Daniel, Melanie N. Zeilinger
2020 arXiv   pre-print
In this paper, we propose to leverage results from optimal control to scale BO to higher dimensional control tasks and to reduce the need for manually selecting the optimization domain.  ...  Bayesian Optimization (BO) is an effective method for optimizing expensive-to-evaluate black-box functions with a wide range of applications for example in robotics, system design and parameter optimization  ...  BO based policy search for higher-dimensional systems.  ... 
arXiv:2001.07394v1 fatcat:mhdal3cgvrfv5af2cxcicqfrbm

Bayesian Optimization for Policy Search in High-Dimensional Systems via Automatic Domain Selection

Lukas P. Frohlich, Edgar D. Klenske, Christian G. Daniel, Melanie N. Zeilinger
2019 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
In this paper, we propose to leverage results from optimal control to scale BO to higher dimensional control tasks and to reduce the need for manually selecting the optimization domain.  ...  Bayesian Optimization (BO) is an effective method for optimizing expensive-to-evaluate black-box functions with a wide range of applications for example in robotics, system design and parameter optimization  ...  BO based policy search for higher-dimensional systems.  ... 
doi:10.1109/iros40897.2019.8967736 dblp:conf/iros/FrohlichKDZ19 fatcat:a6yyuapvdrgmlmrglomzjo6aue

Automated Machine Learning for Deep Recommender Systems: A Survey [article]

Bo Chen, Xiangyu Zhao, Yejing Wang, Wenqi Fan, Huifeng Guo, Ruiming Tang
2022 arXiv   pre-print
Then we discuss the state-of-the-art AutoML approaches that automate the feature selection, feature embeddings, feature interactions, and system design in DRS.  ...  Deep recommender systems (DRS) are critical for current commercial online service providers, which address the issue of information overload by recommending items that are tailored to the user's interests  ...  Works in this domain mainly optimize the recommender systems from a framework perspective.  ... 
arXiv:2204.01390v1 fatcat:ybiang7gajdkrljsrhbq6ih62m

Learning Optimal Data Augmentation Policies via Bayesian Optimization for Image Classification Tasks [article]

Chunxu Zhang and Jiaxu Cui and Bo Yang
2019 arXiv   pre-print
We propose a method named BO-Aug for automating the process by finding the optimal DA policies using the Bayesian optimization approach.  ...  Our method can find the optimal policies at a relatively low search cost, and the searched policies based on a specific dataset are transferable across different neural network architectures or even different  ...  Bayesian optimization can sample a promising DA policy in each iteration from policies search space.  ... 
arXiv:1905.02610v2 fatcat:3x3wqpnmmbbgvpxiu6fiwuso4y

Deep Kernels for Optimizing Locomotion Controllers [article]

Rika Antonova, Akshara Rai, Christopher G. Atkeson
2017 arXiv   pre-print
To address this, prior work has proposed using domain expertise for constructing custom distance metrics for locomotion. In this work we show how to learn such a distance metric automatically.  ...  Bayesian Optimization, a sample-efficient optimization framework, has recently been widely applied to address this problem, but further improvements in sample efficiency are needed for practical applicability  ...  Acknowledgments This research was supported in part by National Science Foundation grant IIS-1563807, the Max-Planck-Society, & the Knut and Alice Wallenberg Foundation.  ... 
arXiv:1707.09062v2 fatcat:nbmemaqzszclpmejcqyzkvlwxi

Scalable Constrained Bayesian Optimization [article]

David Eriksson, Matthias Poloczek
2021 arXiv   pre-print
The global optimization of a high-dimensional black-box function under black-box constraints is a pervasive task in machine learning, control, and engineering.  ...  In particular, these characteristics dramatically impact the performance of Bayesian optimization methods, that otherwise have become the de facto standard for sample-efficient optimization in unconstrained  ...  Efficient high dimensional Bayesian optimization with additivity and quadrature Fourier features. In Advances in Neural Information Processing Systems, pp. 9005-9016, 2018.  ... 
arXiv:2002.08526v3 fatcat:3652qjzfqnbbbl5v6pc6ponab4

Taking the Human Out of the Loop: A Review of Bayesian Optimization

Bobak Shahriari, Kevin Swersky, Ziyu Wang, Ryan P. Adams, Nando de Freitas
2016 Proceedings of the IEEE  
Bayesian optimization is a powerful tool for the joint optimization of design choices that is gaining great popularity in recent years.  ...  If optimized jointly, these parameters can result in significant improvements.  ...  This idea enables us to perform Bayesian optimization in a low-dimensional space to optimize a high-dimensional function with low intrinsic dimensionality.  ... 
doi:10.1109/jproc.2015.2494218 fatcat:dcdmezhogrd45ippmdaslddlxa

Bayesian Optimization for Optimizing Retrieval Systems

Dan Li, Evangelos Kanoulas
2018 Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining - WSDM '18  
Grid and random search, the dominant methods to search for the optimal system configuration, lack a search strategy that can guide them in the hyperparameter space.  ...  In this paper, we propose to use Bayesian Optimization to jointly search and optimize over the hyperparameter space.  ...  Random search is comparable with Bayesian Optimization in low (2) dimensional space, but it fails in high (18) dimensional space.  ... 
doi:10.1145/3159652.3159665 dblp:conf/wsdm/LiK18 fatcat:q3bsdk4fdnba7izcm6tijv4tnu

Model-based Bayesian Reinforcement Learning for Dialogue Management [article]

Pierre Lison
2013 arXiv   pre-print
The results illustrate in particular the benefits of capturing prior domain knowledge with high-level rules.  ...  In this paper, we investigate an alternative strategy grounded in model-based Bayesian reinforcement learning.  ...  5, 6, 7, 8] to automatically optimise the dialogue policy.  ... 
arXiv:1304.1819v1 fatcat:2j7gywskjvcm3pmok5eaqwjh2a

Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree Search [article]

Linnan Wang, Rodrigo Fonseca, Yuandong Tian
2022 arXiv   pre-print
benchmarks, in particular for high-dimensional problems.  ...  Given a set of samples {_i, y_i}, building a global model (like Bayesian Optimization (BO)) suffers from the curse of dimensionality in the high-dimensional search space, while a greedy search may lead  ...  Sampling via Bayesian Optimizations: select finds a path from the root to leaf, and SVMs on the path collectively intersects a region for sampling (e.g. Ω E in Fig. 2(c) ).  ... 
arXiv:2007.00708v2 fatcat:l7wgsqtzxff6ngd4jhomov7tve

Recent Advances in Bayesian Optimization [article]

Xilu Wang, Yaochu Jin, Sebastian Schmitt, Markus Olhofer
2022 arXiv   pre-print
systems.  ...  Hence, this paper attempts to provide a comprehensive and updated survey of recent advances in Bayesian optimization and identify interesting open problems.  ...  Therefore, we are interested in scalable Bayesian optimization algorithms for tackling high dimensionality, rather than construction of high-dimensional GPs only.  ... 
arXiv:2206.03301v1 fatcat:d4mlbxwdjvad5jbzmsskp44dxq

Bayesian optimization for automated model selection

Gustavo Malkomes, Chip Schaff, Roman Garnett
2016 Neural Information Processing Systems  
Our proposed search method is based on Bayesian optimization in model space, where we reason about model evidence as a function to be maximized.  ...  We present a sophisticated method for automatically searching for an appropriate kernel from an infinite space of potential choices.  ...  Additionally, GM acknowledges support from the Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES).  ... 
dblp:conf/nips/MalkomesSG16 fatcat:xcpfjeo5dnddhjobuikfcko6sq

Bayesian Optimization for Whole-Body Control of High Degrees of Freedom Robots through Reduction of Dimensionality

Kai Yuan, Iordanis Chatzinikolaidis, Zhibin Li
2019 IEEE Robotics and Automation Letters  
However, for high dimensional problems, BO is often infeasible in realistic settings as we studied in this paper.  ...  This paper aims to achieve automatic tuning of optimal parameters for whole-body control algorithms to achieve the best performance of high-DoF robots.  ...  This research is supported by the EPSRC CDT in Robotics and Autonomous Systems (EP/L016834/1), Future AI and Robotics for Space (EP/R026092/1), and Offshore Robotics for Certification of Assets (EP/R026173  ... 
doi:10.1109/lra.2019.2901308 fatcat:4ua55c44xvasrorllyxif4rvme

A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning [article]

Eric Brochu and Vlad M. Cora and Nando de Freitas
2010 arXiv   pre-print
This permits a utility-based selection of the next observation to make on the objective function, which must take into account both exploration (sampling from areas of high uncertainty) and exploitation  ...  of Bayesian optimization based on our experiences.  ...  et al. [2009] also applied Bayesian optimization to policy search.  ... 
arXiv:1012.2599v1 fatcat:tl62s6djn5hkpi2d4ctw3cnini

AutoML to Date and Beyond: Challenges and Opportunities [article]

Shubhra Kanti Karmaker Santu, Md. Mahadi Hassan, Micah J. Smith, Lei Xu, ChengXiang Zhai, Kalyan Veeramachaneni
2021 arXiv   pre-print
These steps often require a prolonged back-and-forth that makes this process inefficient for domain experts and data scientists alike, and keeps so-called AutoML systems from being truly automatic.  ...  In this review article, we introduce a new classification system for AutoML systems, using a seven-tiered schematic to distinguish these systems based on their level of autonomy.  ...  MLBox can process, clean and format row data in distributed fashion. They use robust methods for feature selection and optimize the hyperparameters in high-dimensional space.  ... 
arXiv:2010.10777v4 fatcat:arixmky6erdvhnmboe2sfgbb7a
« Previous Showing results 1 — 15 out of 5,789 results