SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2412.07186 · NeurIPS · 2024

Monte Carlo Tree Search based Space Transfer for Black-box Optimization

Lei Song, Ke Xue, Xiaobin Huang, Chao Qian, Shukuan Wang

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 16 functions out of this paper's own repositories and ran 13 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
lamda-bbo/mcts-transfer canonical 13 of 16
FunctionStatusWhere it lives
Classifier Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("8b3cb92357c3a2e9")
KL_distance Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("5f9da55b32debcde")
Kendall_coefficient Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("be8819f9e1077ed8")
from_unit_cube Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("3ba380d745085a84")
get_N_best Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("9fc71b89b88775ad")
get_N_best_mean Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("b8e4666b2b1acccf")
get_data_in_node Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("dcb259fded886307")
get_num_best Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("a9fef6c1f8daf75c")
latin_hypercube Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("37d629517c8ac0d5")
manhattan_distance Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("3e34043a45f43b60")
manhattan_distance_N Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("65c707986a8cb46a")
minmax Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("a89769c364b31a64")
standardization Ran lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("3760c50860f8cbc1")
MCTS Not yet run lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("a09bc86bb469556d")
Node Not yet run lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("d1c4bc7ac1ec6710")
kl_divergence Not yet run lamda-bbo/mcts-transfer/mcts/MCTS.py
pointer only (licence: NONE) · get_code("583e40f60b83cc8b")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Bayesian optimization (BO) is a popular method for computationally expensive black-box optimization. However, traditional BO methods need to solve new problems from scratch, leading to slow convergence. Recent studies try to extend BO to a transfer learning setup to speed up the optimization, where search space transfer is one of the most promising approaches and has shown impressive performance on many tasks. However, existing search space transfer methods either lack an adaptive mechanism or are not flexible enough, making it difficult to efficiently identify promising search space during the optimization process. In this paper, we propose a search space transfer learning method based on Monte Carlo tree search (MCTS), called MCTS-transfer, to iteratively divide, select, and optimize in a learned subspace. MCTS-transfer can not only provide a well-performing search space for warm-start but also adaptively identify and leverage the information of similar source tasks to reconstruct the search space during the optimization process. Experiments on synthetic functions, real-world problems, Design-Bench and hyperparameter optimization show that MCTS-transfer can demonstrate superior performance compared to other search space transfer methods under different settings. Our code is available at https://github.com/lamda-bbo/mcts-transfer.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2412.07186")
get_code_for_paper("2412.07186")
have("2412.07186")

Connect an agent — have() is free.