Evaluation of Neural Architecture Search Approaches for Offshore Platform Offset Prediction

• Tomaz M. Suller USP
• Eric O. Gomes USP
• Henrique B. Oliveira USP
• Lucas P. Cotrim USP
• Amir M. Sa’ad USP
• Ismael H. F. Santos Petrobras
• Rodrigo A. Barreira Petrobras
• Eduardo A. Tannuri USP
• Edson S. Gomi USP
• Anna H. R. Costa USP

DOI: https://doi.org/10.5753/eniac.2021.18264

Resumo

This paper proposes a solution based on Multi-Layer Perceptron (MLP) to predict the offset of the center of gravity of an offshore platform. It also performs a comparative study with three optimization algorithms – Random Search, Simulated Annealing, and Bayesian Optimization (BO) – to find the best MLP architecture. Although BO obtained the best architecture in the shortest time, ablation studies developed in this paper with hyperparameters of the optimization process showed that the result is sensitive to them and deserves attention in the Neural Architecture Search process.

Referências

Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A nextgeneration hyperparameter optimization framework. CoRR, abs/1907.10902.

Bergstra, J., Bardenet, R., Bengio, Y., and K´egl, B. (2011). Algorithms for hyperparameter optimization. In Advances in Neural Information Processing Systems, pages 1–9.

Cotrim, L. P., Oliveira, H. B., Filho, A., Santos, I., Barreira, A., Tannuri, E. A., Costa, A. H. R., and Gomi, E. S. (2021). Neural Network Meta-Models for FPSO Motion Prediction from Environmental Data. In Proc. of the Int. Conf. on Offshore Mechanics and Arctic Engineering.

Elsken, T., Metzen, J. H., and Hutter, F. (2019). Neural architecture search: A survey. Journal of Machine Learning Research, 20(55):1–21.

Feurer, M. and Hutter, F. (2019). Hyperparameter Optimization, pages 3–33. Springer International Publishing.

Frazier, P. I. (2018). A tutorial on bayesian optimization. arXiv:1807.02811.

Goffe, W. L., Ferrier, G. D., and Rogers, J. (1994). Global optimization of statistical functions with simulated annealing. Journal of Econometrics, 60(1):65–99.

Gumley, J. M., Henry, M. J., and Potts, A. E. (2016). A novel method for predicting the motion of moored oating bodies. In 35th Int. Conf. on Ocean, Offshore and Arctic Engineering, OMAE2016-54674. DOI = 10.1115/OMAE2016-54674.

Jin, H., Song, Q., and Hu, X. (2018). Efficient neural architecture search with network morphism. CoRR, abs/1806.10282.

Liu, H., Simonyan, K., and Yang, Y. (2018). DARTS: differentiable architecture search. CoRR, abs/1806.09055.

Märtens, M. and Izzo, D. (2019). Neural network architecture search with differentiable cartesian genetic programming for regression. CoRR, abs/1907.01939.

Mazaheri, S., Mesbahi, E., Downie, M., and Incecik, A. (2003). Seakeeping analysis of a turret-moored FPSO by using artificial neural networks. In Proc. of the Int. Conf. on Offshore Mechanics and Arctic Engineering. DOI = 10.1115/OMAE2003-37148.

Nishimoto, K., Fucatu, C. H., and Masetti, I. Q. (2002). Dynasim—A Time Domain Simulator of Anchored FPSO. J. Offshore Mech. Arct. Eng., 124(4):203–211.

Pinos, M., Mrazek, V., and Sekanina, L. (2021). Evolutionary neural architecture search supporting approximate multipliers. CoRR, abs/2101.11883.

Sciuto, C., Yu, K., Jaggi, M., Musat, C., and Salzmann, M. (2019). Evaluating the search phase of neural architecture search. CoRR, abs/1902.08142.

van Laarhoven, P. J. M. and Aarts, E. H. L. (1987). Simulated Annealing: Theory and Applications. Springer Netherlands.

White, C., Neiswanger, W., and Savani, Y. (2019). Bananas: Bayesian optimization with neural architectures for neural architecture search. arXiv:1910.11858.

Zoph, B. and Le, Q. V. (2016). Neural architecture search with reinforcement learning. arXiv:1611.01578.

Zoph, B., Vasudevan, V., Shlens, J., and Le, Q. V. (2017). Learning transferable architectures for scalable image recognition. CoRR, abs/1707.07012.