Publications

Publications by Tiago Manuel Campelos

2014

Data Extraction Tool to Analyse, Transform and Store Real Data from Electricity Markets

Authors
Pereira, IF; Sousa, TM; Praça, I; Freitas, A; Pinto, T; Vale, ZA; Morais, H;

Publication
Distributed Computing and Artificial Intelligence, 11th International Conference, DCAI 2014, Salamanca, Spain, June 4-6, 2014

Abstract

2018

Electricity Price Forecast for Futures Contracts with Artificial Neural Network and Spearman Data Correlation

Authors
Nascimento, J; Pinto, T; Vale, ZA;

Publication
Distributed Computing and Artificial Intelligence, 15th International Conference, DCAI 2018, Toledo, Spain, 20-22 June 2018, Special Sessions I.

Abstract

2014

Short-term wind speed forecasting using Support Vector Machines

Authors
Pinto, T; Ramos, S; Sousa, TM; Vale, ZA;

Publication
2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2014, Orlando, FL, USA, December 9-12, 2014

Abstract

2019

ALBidS: A Decision Support System for Strategic Bidding in Electricity Markets

Authors
Pinto, T; Vale, ZA;

Publication
Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems, AAMAS '19, Montreal, QC, Canada, May 13-17, 2019

Abstract

2021

Sparse Training Theory for Scalable and Efficient Agents

Authors
Mocanu, DC; Mocanu, E; Pinto, T; Curci, S; Nguyen, PH; Gibescu, M; Ernst, D; Vale, ZA;

Publication
AAMAS '21: 20th International Conference on Autonomous Agents and Multiagent Systems, Virtual Event, United Kingdom, May 3-7, 2021.

Abstract
A fundamental task for artificial intelligence is learning. Deep Neural Networks have proven to cope perfectly with all learning paradigms, i.e. supervised, unsupervised, and reinforcement learning. Nevertheless, traditional deep learning approaches make use of cloud computing facilities and do not scale well to autonomous agents with low computational resources. Even in the cloud, they suffer from computational and memory limitations, and they cannot be used to model adequately large physical worlds for agents which assume networks with billions of neurons. These issues are addressed in the last few years by the emerging topic of sparse training, which trains sparse networks from scratch. This paper discusses sparse training state-of-the-art, its challenges and limitations while introducing a couple of new theoretical research directions which has the potential of alleviating sparse training limitations to push deep learning scalability well beyond its current boundaries. Nevertheless, the theoretical advancements impact in complex multi-agents settings is discussed from a real-world perspective, using the smart grid case study.

CloseRead Abstract

2023

Artificial Intelligence as a Booster of Future Power Systems

Authors
Pinto, T;

Publication
ENERGIES

Abstract
Worldwide power and energy systems are changing significantly [...]

CloseRead Abstract