Publications

On the Undecidability of Artificial Intelligence Alignment: Machines that Halt

Published in ArXiv Preprint. Submitted for review in SciRep AI Alignment Collection, 2024

AI Alignment is undecidable. Nevertheless, there is an enumerable set of provenly aligned AIs that are constructed from a finite set of provenly aligned operations. We propose a halting constraint that guarantees that the AI model always reaches a terminal state in finite execution steps.

Recommended citation: de Melo, G. A., Maximo, M. R. O. D. A., Soma, N. Y., & de Castro, P. A. L. (2024). On the Undecidability of Artificial Intelligence Alignment: Machines that Halt. arXiv [Cs.AI]. Retrieved from http://arxiv.org/abs/2408.08995
Download Paper

High Speed Marker Tracking for Flight Tests

Published in IEEE Latin America Transactions, 2022

We propose a new detecting and tracking model based on CNN, that uses fiducial markers.

Recommended citation: G. A. Melo, M. Máximo and P. A. Castro, "High Speed Marker Tracking for Flight Tests," in IEEE Latin America Transactions, vol. 20, no. 10, pp. 2237-2243, Oct. 2022, doi: 10.1109/TLA.2022.9885171.
Download Paper

A new approach to river flow forecasting: LSTM and GRU multivariate models

Published in IEEE Latin America Transactions, 2019

The monthly flow of a river was estimated using two recurrent neural networks techniques: Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU).

Recommended citation: G. Adriano de Melo, D. N. Sugimoto, P. M. Tasinaffo, A. H. Moreira Santos, A. M. Cunha and L. A. Vieira Dias, "A new approach to river flow forecasting: LSTM and GRU multivariate models," in IEEE Latin America Transactions, vol. 17, no. 12, pp. 1978-1986, December 2019, doi: 10.1109/TLA.2019.9011542.
Download Paper

A MapReduce Based Approach for Circle Detection

Published in SciTePress, 2019

This paper proposes a parallel algorithm for circle detection using Hough Transform and MapReduce paradigm.

Recommended citation: Coelho, Mateus Menezes Azevedo, Dylan Nakandakari Sugimoto, Gabriel Adriano Melo, Vitor Venceslau Curtis, and Juliana de Melo Bezerra. "A MapReduce Based Approach for Circle Detection." In Proceedings of the 14th International Conference on Software Technologies (ICSOFT 2019), edited by Marten van Sinderen and Leszek A. Maciaszek, 454-459. Prague, Czech Republic: SciTePress, 2019. https://doi.org/10.5220/0007827604540459.
Download Paper

Gated Recurrent Unit Hierarchical Architecture for Fundamental Stock Analysis and Forecast

Published in WAIAF (Workshop of Artificial Intelligence Applied to Finance) 2019, 2019

This work proposed a novel technique for fundamental analysis using GRU hierarchical models.

Recommended citation: Gabriel Adriano Melo, Paulo Marcelo Tasinaffo. "Gated Recurrent Unit Hierarchical Architecture for Fundamental Stock Analysis and Forecast." Presented at the Workshop of Artificial Intelligence Applied to Finance 2019 (WAIAF 2019).
Download Paper | Download Slides

Gabriel A. Melo