Bayesian optimisation of functions on graphs
The increasing availability of graph-structured data motivates the task of optimising over functions defined on the node set of graphs. Traditional graph search algorithms can be applied in this case, but they may be sample-inefficient and do not make use of information about the function values; on...
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
Neural Information Processing Systems Foundation
2024
|
| _version_ | 1811140294304858112 |
|---|---|
| author | Wan, X Osselin, P Kenlay, H Ru, B Osborne, MA Dong, X |
| author_facet | Wan, X Osselin, P Kenlay, H Ru, B Osborne, MA Dong, X |
| author_sort | Wan, X |
| collection | OXFORD |
| description | The increasing availability of graph-structured data motivates the task of optimising over functions defined on the node set of graphs. Traditional graph search algorithms can be applied in this case, but they may be sample-inefficient and do not make use of information about the function values; on the other hand, Bayesian optimisation is a class of promising black-box solvers with superior sample efficiency, but it has scarcely been applied to such novel setups. To fill this gap, we propose a novel Bayesian optimisation framework that optimises over functions defined on generic, large-scale and potentially unknown graphs. Through the learning of suitable kernels on graphs, our framework has the advantage of adapting to the behaviour of the target function. The local modelling approach further guarantees the efficiency of our method. Extensive experiments on both synthetic and real-world graphs demonstrate the effectiveness of the proposed optimisation framework. |
| first_indexed | 2024-09-25T04:19:42Z |
| format | Conference item |
| id | oxford-uuid:f51445f4-a776-454a-84cd-82b2a7f1b4f7 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:19:42Z |
| publishDate | 2024 |
| publisher | Neural Information Processing Systems Foundation |
| record_format | dspace |
| spelling | oxford-uuid:f51445f4-a776-454a-84cd-82b2a7f1b4f72024-07-29T16:46:23ZBayesian optimisation of functions on graphsConference itemhttp://purl.org/coar/resource_type/c_5794uuid:f51445f4-a776-454a-84cd-82b2a7f1b4f7EnglishSymplectic Elements Neural Information Processing Systems Foundation2024Wan, XOsselin, PKenlay, HRu, BOsborne, MADong, XThe increasing availability of graph-structured data motivates the task of optimising over functions defined on the node set of graphs. Traditional graph search algorithms can be applied in this case, but they may be sample-inefficient and do not make use of information about the function values; on the other hand, Bayesian optimisation is a class of promising black-box solvers with superior sample efficiency, but it has scarcely been applied to such novel setups. To fill this gap, we propose a novel Bayesian optimisation framework that optimises over functions defined on generic, large-scale and potentially unknown graphs. Through the learning of suitable kernels on graphs, our framework has the advantage of adapting to the behaviour of the target function. The local modelling approach further guarantees the efficiency of our method. Extensive experiments on both synthetic and real-world graphs demonstrate the effectiveness of the proposed optimisation framework. |
| spellingShingle | Wan, X Osselin, P Kenlay, H Ru, B Osborne, MA Dong, X Bayesian optimisation of functions on graphs |
| title | Bayesian optimisation of functions on graphs |
| title_full | Bayesian optimisation of functions on graphs |
| title_fullStr | Bayesian optimisation of functions on graphs |
| title_full_unstemmed | Bayesian optimisation of functions on graphs |
| title_short | Bayesian optimisation of functions on graphs |
| title_sort | bayesian optimisation of functions on graphs |
| work_keys_str_mv | AT wanx bayesianoptimisationoffunctionsongraphs AT osselinp bayesianoptimisationoffunctionsongraphs AT kenlayh bayesianoptimisationoffunctionsongraphs AT rub bayesianoptimisationoffunctionsongraphs AT osbornema bayesianoptimisationoffunctionsongraphs AT dongx bayesianoptimisationoffunctionsongraphs |