Convergence Rate of a Simulated Annealing Algorithm with Noisy Observations
Authors: Clément Bouttier, Ioana Gavra
JMLR 2019 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | This work is completed with a set of numerical experimentations and assesses the practical performance both on benchmark test cases and on real world examples. |
| Researcher Affiliation | Collaboration | Clement Bouttier EMAIL Aircraft Performance Departement Airbus Operations 316 route de Blagnac, 31300 Toulouse, France Ioana Gavra EMAIL Institut de Math ematiques de Toulouse Universit e Toulouse III 118 route de Narbonne, 31062 Toulouse Cedex 9, France |
| Pseudocode | Yes | Algorithm 1 Noisy Simulated Annealing |
| Open Source Code | No | The paper does not provide an explicit statement or link to the source code for the methodology described. |
| Open Datasets | No | The paper refers to 'benchmark test cases', 'Hajek test case', and the 'Ackley function' which are problem definitions or mathematical functions, not publicly accessible datasets with specific links or citations. For the 'Aircraft Trajectory Optimization' problem, it mentions a 'black box trajectory evaluator' without providing access to any dataset. |
| Dataset Splits | No | The paper discusses performing '1000 Monte Carlo runs' for experiments, which refers to repetitions, not dataset splits. For the mathematical test functions and the black-box aircraft model, there are no datasets described that would require explicit training/test/validation splits. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper does not specify any software names with version numbers that would be required to replicate the experiment. |
| Experiment Setup | Yes | In Section 6 'Numerical Experiments', the paper describes using 'additive Gaussian noise at each evaluation', testing with 'different variance levels', and comparing 'linear increase of the mean number of samples' (nk = k) with 'quadratic one' (nk = k^2). It also specifies '1000 Monte Carlo runs for each noise level and estimation schedule' and uses a 'uniformly (2000 points) discretized version of the Ackley function in one dimension on [-100, 100]'. |