Filters








18,839 Hits in 4.3 sec

Evolving intrinsic motivations for altruistic behavior [article]

Jane X. Wang, Edward Hughes, Chrisantha Fernando, Wojciech M. Czarnecki, Edgar A. Duenez-Guzman, Joel Z. Leibo
2019 arXiv   pre-print
We present results in two challenging environments, and interpret these in the context of cultural and ecological evolution.  ...  Multi-agent cooperation is an important feature of the natural world.  ...  We propose that evolution can be applied to remove the hand-crafting of intrinsic motivation, similar to other applications of evolution in deep learning.  ... 
arXiv:1811.05931v2 fatcat:3cyquudnffa3dolz74tefcozv4

CHARISMA: A Context Hierarchy-based cognitive architecture for self-motivated social agents

Matthew Conforth, Yan Meng
2011 The 2011 International Joint Conference on Neural Networks  
In this paper, we are developing the CIVS (Civilization-Inspired Vying Societies) system, which is a novel evolutionary learning multi-agent system loosely inspired by the history of human civilization  ...  By using intrinsic motivations and dynamic knowledge representation, the CHARISMA cognitive architecture can provide autonomous mental development capability for the agent to develop its knowledge and  ...  THE CIVS SYSTEM CIVS is a multi-agent Alife simulation designed to produce intelligent emergent behaviors and broad adaptability in groups of social agents.  ... 
doi:10.1109/ijcnn.2011.6033455 dblp:conf/ijcnn/ConforthM11a fatcat:h3wbg2nanrdczfnbdzvreavq2q

Dynamic Noises of Multi-Agent Environments Can Improve Generalization: Agent-based Models meets Reinforcement Learning [article]

Mohamed Akrout, Amal Feriani, Bob McLeod
2022 arXiv   pre-print
Numerical simulations demonstrate that the intrinsic noise in the ABM-based dynamics of the SIR model not only improve the average reward but also allow the RL agent to generalize on a wider ranges of  ...  of RL agents.  ...  We first compare the evolution of the SIR dynamics using ODEs and ABMs to seize the intrinsic noise in the dynamics of the ABM-based SIR environment.  ... 
arXiv:2204.14076v1 fatcat:e7tud5zpqzeqrlom3bofqvamhi

CURIOUS: Intrinsically Motivated Modular Multi-Goal Reinforcement Learning [article]

Cédric Colas, Pierre Fournier, Olivier Sigaud, Mohamed Chetouani, Pierre-Yves Oudeyer
2019 arXiv   pre-print
In open-ended environments, autonomous learning agents must set their own goals and build their own curriculum through an intrinsically motivated exploration.  ...  Agents focus sequentially on goals of increasing complexity, and focus back on goals that are being forgotten.  ...  In the IMGEP framework however, agents are intrinsically motivated to set their own.  ... 
arXiv:1810.06284v4 fatcat:ivksmu4qr5hqvnnjvyubbvsxjq

Evo-devo-robo workshop program

Stephane Doncieux, Yaochu Jin, Jean-Baptiste Mouret
2012 Proceedings of the fourteenth international conference on Genetic and evolutionary computation conference companion - GECCO Companion '12  
These abstracts cover researches in both fields and give an overview of the potential interactions between developmental and evolutionary robotics.  ...  Developmental robotics (also known as epigenetic robotics) is mainly concerned with modelling the postnatal development of cognitive behaviours in living systems, such as language, emotion, curiosity,  ...  The agent will also use as source of information its peers. I will consider methods from active learning, intrinsic motivation systems and social learning.  ... 
doi:10.1145/2330784.2330836 dblp:conf/gecco/DoncieuxJM12 fatcat:weqba7t6zjc5box7vz6jixckse

Emergent Tool Use From Multi-Agent Autocurricula [article]

Bowen Baker, Ingmar Kanitscheider, Todor Markov, Yi Wu, Glenn Powell, Bob McGrew, Igor Mordatch
2020 arXiv   pre-print
We find clear evidence of six emergent phases in agent strategy in our environment, each of which creates a new pressure for the opposing team to adapt; for instance, agents learn to build multi-object  ...  reinforcement learning methods such as intrinsic motivation.  ...  We also thank Alex Ray for writing parts of our open sourced code.  ... 
arXiv:1909.07528v2 fatcat:u4efcciy7nbl3g5s4kwnillcci

Agent-based interaction analysis of consumer behavior

Lamjed Ben Said, Thierry Bouron, Alexis Drogoul
2002 Proceedings of the first international joint conference on Autonomous agents and multiagent systems part 1 - AAMAS '02  
It shows how genetic algorithms (GA), in addition to multi-agent systems, are used to fit the characteristics of the virtual consumers' population into a global realistic market behavior.  ...  The paper presents also the CUstomer BEhavior Simulator (CUBES), which has been realized for implementing the customer model and leading multi-agents simulations.  ...  Figure 4 . 4 Multi-agent simulation including GA for calibrating consumer agents' characteristics.  ... 
doi:10.1145/544741.544787 dblp:conf/atal/SaidBD02 fatcat:h3cllmslrbegxakrgsjr2hx7fu

CoEvolution of Effective Observers and Observed Multi-agents System [chapter]

Christophe Philemotte, Hugues Bersini
2005 Lecture Notes in Computer Science  
So two Genetic Algorithms (GAs) are intertwined to solve the whole problem: one in the original space and one in the space of observables of the original one.  ...  In order to discover this "clever way", a second search process has to be engaged in the space of the observables.  ...  Secondly, the generalisation of the new approach to other multi-agents systems might be easy: masking the influence of agents neighbourhood to update an agent.  ... 
doi:10.1007/11553090_79 fatcat:hjyq4rzmtrfyvm6kt7qimommfy

Can partisan voting lead to truth?

Naoki Masuda, S Redner
2011 Journal of Statistical Mechanics: Theory and Experiment  
We study an extension of the voter model in which each agent is endowed with an innate preference for one of two states that we term as "truth" or "falsehood".  ...  We determine the conditions under which a population of such agents can ultimately reach a consensus for the truth, a consensus for falsehood, or reach an impasse where an agent tends to adopt the opinion  ...  We thank James Fowler for correspondence that helped motivate this work, and Damon Centola for discussions and literature advice.  ... 
doi:10.1088/1742-5468/2011/02/l02002 fatcat:2dd2rcxckjf3tijcpyukcnlzhe

Intrinsically Motivated Open-Ended Multi-Task Learning Using Transfer Learning to Discover Task Hierarchy

Nicolas Duminy, Sao Mai Nguyen, Junshuai Zhu, Dominique Duhaut, Jerome Kerdreux
2021 Applied Sciences  
Combining both goal-babbling with imitation learning, and active learning with transfer of knowledge based on intrinsic motivation, our algorithm self-organises its learning process.  ...  We show with a simulation and a real industrial robot arm, in cross-task and cross-learner transfer settings, that task composition is key to tackle highly complex tasks.  ...  The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.  ... 
doi:10.3390/app11030975 fatcat:4inw4fj5ajgnzf3nm6x5aghhvu

Modeling and simulating crowdsourcing as a complex biological system: human crowds manifesting collective intelligence on the internet

Thierry Buecheler, Rocky Lonigro, Rudolf M. Füchslin, Rolf Pfeifer
2011 European Conference on Artificial Life  
It introduces the mechanics of a multi-agent system (MAS) by building on insights from empirical science in several disciplines.  ...  The simulator allows running sensitivity analyses of multiple parameters as well as simulation of intractable interactions of complex networks of irrational agents.  ...  Multi-agent system simulations have been shown to have certain advantages over these models when simulating large groups of agents.  ... 
dblp:conf/ecal/BuechelerLFP11 fatcat:4f5b7qkiznezpchuh4uuhfk5zy

Bio-inspired Computational Algorithms in Educational and Serious Games: Some Examples [chapter]

Michela Ponticorvo, Andrea Di Ferdinando, Davide Marocco, Orazio Miglino
2016 Lecture Notes in Computer Science  
In the present paper we will introduce different approaches to embed this kind of models in Serious and Educational Games.  ...  According to a multi-level description of game design process, bio-inspired computational algorithms can be visible to the user, residing at an external, shell level; can be invisible to the user, residing  ...  In particular, three subcomponents of motivation are simulated: intrinsic, reward and fear.  ... 
doi:10.1007/978-3-319-45153-4_80 fatcat:pucvrawx5jeaxar462gl7nkes4

Systems Biology at the giga-scale: large multi-scale models of complex, heterogeneous multicellular systems

Arnau Montagud, Miguel Ponce de León, Alfonso Valencia
2021 Current Opinion in Systems Biology  
This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article.  ...  of record.  ...  ., in preparation) is our current effort to include MPI in PhysiCell, an open-source, flexible, and lattice-free agent-based framework for multi-scale simulation of multi-cellular systems that currently  ... 
doi:10.1016/j.coisb.2021.100385 fatcat:jff4mcrllfhyxnf3stsihfhhwy

An Autonomous Emotional Virtual Character: An Approach with Deep and Goal-Parameterized Reinforcement Learning

Gilzamir Ferreira Gomes, Creto Augusto Vidal, Joaquim Bento Cavalcante Neto, Yuri Lenon Barbosa Nogueira
2020 Journal of Interactive Systems  
The agent is a virtual character who lives in a three-dimensional maze world. We found that emotion drivers can induce the behavior of a trained agent.  ...  We train agents who can randomly assume these goals while trying to maximize a reward function based on intrinsic and extrinsic motivations. A mapping between motivation and emotion was carried out.  ...  Acknowledgment The authors would like to thank CNPq, the State University of Acarau Valley, and the Graduate Program in Computer Science of the Federal University of Ceará (MDCC/UFC) for their financial  ... 
doi:10.5753/jis.2020.751 fatcat:3pmme7d2k5ebjkazsxjvgzn3qe

Human Crowd Simulation: What Can We Learn From Alife?

Rui Filipe Antunes, Magnenat-Thalmann
2016 Zenodo  
One of the key components of the suspension of disbelief in real-time 3D simulations is the apparent authenticity of actions and gestures played by the individuals of the virtual population.  ...  This paper addresses this aspect of simulation, by investigating ways to improve the behavioral realism of virtual humanoid characters in groups and small multitudes.  ...  These are multi-agent systems where individuals appear organized in a hierarchical way (in the form of a food-chain), and traditionally agents have their motivations based on their self-sustenance and  ... 
doi:10.5281/zenodo.820541 fatcat:qbyig74kffaetdbfjdd5qbe4re
« Previous Showing results 1 — 15 out of 18,839 results