The 22nd edition of the Workshop “From Objects to Agents” (WOA) will be held in Bologna as a blended event to serve as a forum for researchers and practitioners working on all aspects of agents and multi-agent systems (MAS). Following the significant interest that all facets of Artificial Intelligence (AI) have been recently obtaining, the topic for WOA 2021 is

Multi-agent systems in the machine learning era Artificial Intelligence is here to stay, along with both its symbolic and sub-symbolic branches. While symbolic AI played a central role in the early days of AI, the last decade has been characterised by the explosion of data-driven, machine-learning-based intelligent systems. Nowadays, the exploitation of machine learning (ML) and big data processing is so pervasive that novel results and applications are being proposed at an unprecedented pace.

While multi-agent systems (MAS) have been entangled with symbolic AI since the very beginning, it is undoubted that they have been extensively exploited in several data-intensive domains including – but not limited to – robotics, telecommunications, simulation, decision support systems, and economics.

MAS and ML have the potential to benefit from each other. In fact, on the one side, it is becoming increasingly complex for agent designers to forecast all possible operating situations, so learning from data is fundamental for next-generation MAS—in order to adapt to the intricacies of an increasingly complex world. On the other side, modern ML-based solutions may easily struggle when different data-driven solution – possibly attained from disparate data sources – must be integrated, combined, reconciled or explained. Under such a perspective, MAS have certainly a role to play to advance the state of the art of data-driven AI towards integration and explainability.

The intertwining between MAS and ML, therefore, aims at bringing about huge benefits to the field of data-driven AI, both at the theoretical and the practical level; however, a well-grounded integration in terms of methodologies and systems properties is far from being reached and acknowledged.

WOA 2021 will bring together researchers to discuss those issues and to suggest future research questions and perspectives laying at the intersection among MAS and ML. Therefore, the workshop aims at working as an interdisciplinary hub for thinkers, students, researchers, and developers who wish to understand and exploit the impact and potential of MAS in the future of ML, and viceversa.

• Agents theory and technology related to human-centred intelligent systems 
• Software engineering techniques for human-centred intelligent systems
• Languages and tools for agent-oriented human-centred intelligent systems
• Agent-oriented architectures for human-centred intelligent systems
• Responsabile and ethical AI in agent-oriented models and architectures
• Responsabile and ethical AI in agent-based simulations and applications
• MAS for interpretability and explainability of intelligent systems
• Trust, security, and privacy in human-centred intelligent systems