Mechanisms of Social Learning in Evolved Artificial Life

Abstract

Adaptation of agents in artificial life scenarios is especially effective when agents may evolve, i.e., inherit traits from their parents, and learn by interacting with the environment. The learning process may be boosted with forms of social learning, i.e., by allowing an agent to learn by combining its experiences with knowledge transferred among agents. In this work, we tackle two specific questions regarding social learning and evolution: (a) from whom learners should learn? (b) how should knowledge be transferred? We address these questions by experimentally investigating two scenarios: a simple one in which the mechanism for evolution and learning is easily interpretable; a more complex and realistic artificial life scenario in which agents compete for survival. Experimental results show that social learning is more profitable when (a) the learners learn from a small set of good teachers and (b) the knowledge to be transferred is determined by teachers experience, rather than learner experience.

The Long-Term Efficacy of “Social Buffering” in Artificial Social Agents: Contextual Affective Perception Matters

“…We discuss our findings further in Section 4, summarise our conclusions in Section 5, and highlight some of the work's limitations and future directions in Section 6. Ghafurian et al, 2019;Youssef et al, 2015); imitation of others (Breazeal et al, 2005;Doering et al, 2019); and social learning through both physical (robot) (Bartoli et al, 2020) and virtual (Le et al, 2020(Le et al, , 2018 partners, as well as in multi-agent systems (Tampuu et al, 2017;Yonenoh et al, 2019;Pérez et al, 2017). Recent work has also considered models of higher-order cognitive functions: adapting behaviours by inferring the intention or affective state of others (Pieters et al, 2017;Görür et al, 2017), or by perspective taking of other social agents (Trafton et al, 2006).…”

“…Addressing social adaptation for embodied (socially-)adaptive agent models remains an ongoing area of research, with a range of approaches being considered. This includes affect-based behavioural adaptation in human-agent (both physical and virtual) interactions ( Hiolle et al, 2014 ; Tanevska et al, 2019 ; Ghafurian et al, 2019 ; Youssef et al, 2015 ); imitation of others ( Breazeal et al, 2005 ; Doering et al, 2019 ); and social learning through both physical (robot) ( Bartoli et al, 2020 ) and virtual ( Le et al, 2020 , 2018 ) partners, as well as in multi-agent systems ( Tampuu et al, 2017 ; Yonenoh et al, 2019 ; Pérez et al, 2017 ). Recent work has also considered models of higher-order cognitive functions: adapting behaviours by inferring the intention or affective state of others ( Pieters et al, 2017 ; Görür et al, 2017 ), or by perspective taking of other social agents ( Trafton et al, 2006 ).…”

On the Mutual Influence of Human and Artificial Life: an Experimental Investigation

“…Other studies crafted artificial worlds, e.g., Tierra (Ray, 1992), PolyWorld (Yaeger et al, 1994), and Avida (Ofria and Wilke, 2004), with several different goals: they mostly investigate questions related to evolutionary biology (Lenski et al, 2003), ecology (Ventrella, 2005), open-ended evolution (Soros and Stanley, 2014), social learning (Bartoli et al, 2020), or are sources of entertainment and gaming (Dewdney, 1984;Grand and Cliff, 1998). Albeit fascinating, none of these addresses the main research question of this paper, i.e., the mutual influence of human life and ALife.…”