Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
Multi-fidelity Bayesian Optimization with Max-value Entropy Search and its Parallelization
Authors: Shion Takeno, Hitoshi Fukuoka, Yuhki Tsukada, Toshiyuki Koyama, Motoki Shiga, Ichiro Takeuchi, Masayuki Karasuyama
ICML 2020 | Venue PDF | LLM Run Details | Input Tokens: 19,435 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 6,141 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate effectiveness of our approach by using benchmark datasets and a real-world application to materials science data. |
| Researcher Affiliation | Academia | 1Department of Computer Science, Nagoya Institute of Technology, Aichi, Japan 2Center for Advanced Intelligence Project, RIKEN, Tokyo, Japan 3Department of Materials Design Innovation Engineering, Nagoya University, Aichi, Japan 4PRESTO, Japan Science and Technology Agency, Saitama, Japan 5Department of Electrical, Electronic and Computer Engineering, Gifu University, Gifu, Japan 6Center for Materials Research by Information Integration, National Institute for Material Science, Ibaraki, Japan. |
| Pseudocode | Yes | Algorithm 1 shows the procedure of MF-MES for sequential querying. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. |
| Open Datasets | Yes | We used a synthetic function generated by MF-GPR, two benchmark functions, and a real-world dataset from materials science. The details of functions are described as follows. (...) Benchmark Functions: We used two benchmark functions called Styblinski-Tang (d = 2, M = 2), and Hart Mann6 (d = 6, M = 3). (...) Material Data: As an example of practical applications, we applied our method to the parameter optimization of a simulation model in materials science. The task is to optimize d = 2 material parameters in the simulation model (Tsukada et al., 2014). |
| Dataset Splits | No | The paper defines metrics like Simple Regret and Inference Regret but does not specify dataset splits (e.g., percentages or counts) for training, validation, and testing as commonly done in supervised learning. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'GPytorch library' and 'Adam optimizer' but does not specify their version numbers, which is required for reproducible software dependencies. |
| Experiment Setup | Yes | For each evaluation, we ๏ฌt the hyperparameters of the kernel and the noise parameter by maximizing the marginal likelihood with 1000 iterations of Adam optimizer (Kingma & Ba, 2014). For optimization of the highest ๏ฌdelity f and x , we ran L-BFGS-B 10 times with random initialization from the observed data for 200 iterations each time. (...) For the sampling of f in MES and MF-MES, we employed the RFM-based approach described in Section 4, and sampled 10 f s at every iteration. |