Consensus-Based Learning for MAS: Definition, Implementation and Integration in IVEs.

Carlos Carrascosa; Francisco Enguix; Miguel Rebollo; Jaime Andrés Rincón Arango

doi:10.9781/ijimai.2023.08.004

Authors

Carlos Carrascosa Universitat Politècnica de València
Francisco Enguix Universitat Politècnica de València
Miguel Rebollo Universitat Politècnica de València
Jaime Andrés Rincón Arango Universitat Politècnica de València

DOI:

https://doi.org/10.9781/ijimai.2023.08.004

Keywords:

Complex Networks, Distributed AI, Multi-Agent Systems, Neural Network

Supporting Agencies

This work has been developed thanks to the funding of projects: • Grant PID2021-123673OB-C31 funded by MCIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe” • PROMETEO CIPROM/2021/077 • TED2021-131295B-C32 • Ayudas del Vicerrectorado de Investigación de la UPV (PAIDPD-22)

Abstract

One of the main advancements in distributed learning may be the idea behind Google’s Federated Learning (FL) algorithm. It trains copies of artificial neural networks (ANN) in a distributed way and recombines the weights and biases obtained in a central server. Each unit maintains the privacy of the information since the training datasets are not shared. This idea perfectly fits a Multi-Agent System, where the units learning and sharing the model are agents. FL is a centralized approach, where a server is in charge of receiving, averaging and distributing back the models to the different units making the learning process. In this work, we propose a truly distributed learning process where all the agents have the same role in the system. We suggest using a consensus-based learning algorithm that we call Co-Learning. This process uses a consensus process to share the ANN models each agent learns using its private data and calculates the aggregated model. Co-Learning, as a consensus-based algorithm, calculates the average of the ANN models shared by the agents with their local neighbors. This iterative process converges to the averaged ANN model as a central server does. Apart from the definition of the Co-Learning algorithm, the paper presents its integration in SPADE agents, along with a framework called FIVE allowing to develop Intelligent Virtual Environments for SPADE agents. This framework has been used to test the execution of SPADE agents using Co-Learning algorithm in a simulation of an orange orchard field.

Downloads

Download data is not yet available.

References

C. Carrascosa, J. Rincón, M. Rebollo, “Co-learning: Consensus-based learning for multi-agent systems,” in Advances in Practical Applications of Agents, Multi-Agent Systems, and Complex Systems Simulation. The PAAMS Collection, 2022, pp. 63–75.

H. Sánchez San Blas, A. Carmona Balea, A. Sales, L. Augusto Silva, G. Villarrubia González, “A platform for swimming pool detection and legal verification using a multi-agent system and remote image sensing,” International Journal of Interactive Multimedia and Artificial Intelligence, 2023, pp. 1-13, doi: 10.9781/ijimai.2023.01.002.

H. Brendan McMahan, E. Moore, D. Ramage, S. Hampson, B. Agüera y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Artificial intelligence and statistics, 2017, pp. 1273– 1282, PMLR.

P. Kairouz, H. McMahan, B. Avent, A. Bellet, M. Bennis, A. Bhagoji, K. Bonawitz, Z. Charles, G. Cormode, R. Cummings, et al., “Advances and open problems in federated learning,” Foundations and Trends in ML, vol. 14, no. 1–2, pp. 1–210, 2021.

R. Olfati-Saber, R. M. Murray, “Consensus problems in networks of agents with switching topology and time-delays,” IEEE TAC, vol. 49, no. 9, pp. 1520–1533, 2004.

J. Palanca, A. Terrasa, V. Julian, C. Carrascosa, “SPADE 3: Supporting the new generation of multi-agent systems,” IEEE Access, vol. 8, pp. 182537– 182549, 2020, doi: 10.1109/ACCESS.2020.3027357.

J. Palanca, J. Rincon, V. Julian, C. Carrascosa, A. Terrasa, “Developing iot artifacts in a mas platform,” Electronics, vol. 11, no. 4, p. 655, 2022.

J. Rincon, V. Julian, C. Carrascosa, “Flamas: Federated learning based on a spade mas,” Applied Sciences, vol. 12, no. 7, pp. 1–14, 2022, doi: 10.3390/app12073701.

M. Bratman, Intention, Plans, and Practical Reason. Cambridge: Cambridge, MA: Harvard University Press, 1987.

M. Luck, R. Aylett, “Applying artificial intelligence to virtual reality: Intelligent virtual environments,” Applied artificial intelligence, vol. 14, no. 1, pp. 3–32, 2000.

A. Ikidid, E. F. Abdelaziz, M. Sadgal, “Multi-agent and fuzzy inferencebased framework for traffic light optimization,” International Journal of Interactive Multimedia and Artificial Intelligence, vol. 8, no. 2, pp. 88-97, 2023, doi: 10.9781/ijimai.2021.12.002.

U. Wilensky, “Netlogo (and netlogo user manual),” Center for connected learning and computer-based modeling, Northwestern University. http://ccl.northwestern.edu/netlogo, 1999.

J. Rincon, E. Garcia, V. Julian, C. Carrascosa, “The jacalive framework for mas in ive: A case study in evolving modular robotics,” Neurocomputing, vol. 275, pp. 608–617, 2018.

A. Barella, A. Ricci, O. Boissier, C. Carrascosa, “Mam5: multi-agent model for intelligent virtual environments,” in 10th european workshop on multiagent systems (EUMAS 2012), 2012, pp. 16–30.

R. H. Bordini, J. F. Hübner, M. Wooldridge, Programming multi-agent systems in AgentSpeak using Jason. John Wiley & Sons, 2007.

A. Ricci, M. Viroli, A. Omicini, “Cartago: A framework for prototyping artifact-based environments in mas,” in International Workshop on Environments for Multi-Agent Systems, 2006, pp. 67–86, Springer.

S. Luke, G. C. Balan, L. Panait, C. Cioffi-Revilla, S. Paus, “Mason: A java multi-agent simulation library,” in Proceedings of Agent 2003 Conference on Challenges in Social Simulation, vol. 9, 2003.

F. Enguix Andrés, Desarrollo de un generador de simulaciones en Unity 3D para sistemas multi-agente basados en SPADE. PhD dissertation, Universitat Politècnica de València, 2022.

A. Palomares, M. Rebollo, C. Carrascosa, “Supportive consensus,” PLOS ONE, vol. 15, no. 12, pp. 1–30, 2020.

F. Pedroche, M. Rebollo, C. Carrascosa, A. Palomares, “Convergence of weighted-average consensus for undirected graphs,” International Journal of Complex Systems in Science, vol. 4, no. 1, pp. 13–16, 2014.

J. Palanca, A. Terrasa, V. Julian, C. Carrascosa, “Spade 3: Supporting the new generation of multi-agent systems,” IEEE Access, vol. 8, pp. 182537– 182549, 2020, doi: 10.1109/ACCESS.2020.3027357.

M. Sandler, A. G. Howard, M. Zhu, A. Zhmoginov, L. Chen, “Inverted residuals and linear bottlenecks: Mobile networks for classification, detection and segmentation,” CoRR, vol. abs/1801.04381, 2018.

M. C. Silva, J. C. F. da Silva, R. A. R. Oliveira, “Idissc: Edge-computingbased intelligent diagnosis support system for citrus inspection.,” in ICEIS (1), 2021, pp. 685–692.