Lei Song, Ke Xue, Xiaobin Huang, Chao Qian, Shukuan Wang
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.
| Repository | Role | Ran |
|---|---|---|
| lamda-bbo/mcts-transfer | canonical | 13 of 16 |
| Function | Status | Where 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") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
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.
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.