Program
| 14:00 | Welcome and Opening [Slides] |
| 14:20 | Invited Contribution (Alessandro Ricci): Levels of Abstraction in Designing and Programming Systems of Cognitive Agents [Slides] |
| 14:50 | MAMS: Multi-Agent MicroServices (Rem Collier) [Slides] |
| 15:10 | Digital Companion for Industry (Florian Michahelles) [Slides] |
| 15:20 | Towards Enabling Internet-Scale Context-as-a-Service (Alexandru Sorici) [Slides] |
| 15:30 | Coffee Break |
| 16:00 | Keynote (Florian Michahelles): As We May Work - Free Minds from Routines and Enable Creativity and Intuition [Slides] |
| 16:30 | Keynote (Mike Amundsen): "Such a rich set of affordances" - Mapping the Hyperagent Ecosystem [Slides] |
| 17:00 | 1 Minute Madness and Breakout Discussions |
| 17:25 | Wrap up |
| Social Evening |
Keynotes
Florian Michahelles: As We May Work - Free Minds from Routines and Enable Creativity and Intuition
Speaker bio: Florian Michahelles heads the Artificial & Human Intelligence research group at Siemens Corporate Technology in Berkeley, California. Florian investigates new approaches of wrapping AI as a handy tool for human users achieving their tasks more effectively. Before joining Siemens, Florian has led the Auto-ID Labs research consortium at ETH Zurich, he has a Ph.D. from ETH Zurich and a Master’s degree from LMU Munich, both in computer science. He has authored 100+ scientific articles in the field of Internet of Things, Wearable Computing, and Industry 4.0.
Abstract: As automation, digitalization and artificial decision-making are advancing, various stakeholders of our society are asking: What’s in it for us? Do we want to live in this world? What’s left to do for us? This talk will address these fears by proposing a more positive vision of machines and humans shaking hands and integrating their strenghts. The talk will build upon the notion of digital companions and present different examples of human-machine collaboration where machines take-over the recurring routine processes and decisions and free humans’ minds for focusing on the more creative parts. The talk concludes on finding the sweet spots of human-machine collaboration, proposes future research directions and opens the field for discussion.
Mike Amundsen: "Such a rich set of affordances" - Mapping the Hyperagent Ecosystem
Speaker bio: An internationally known author and speaker, Mike Amundsen travels the world consulting and talking about network architecture, Web development, and intersection of technology and society. He works with companies large and small to help them capitalize on the opportunities APIs and Microservices present for both consumers and the enterprise.
Amundsen has authored numerous books and papers. He contributed to the O’Reilly Media book, “Continuous API Management” (2018). His “RESTful Web Clients”, was published by O’Reilly in February 2017 and he co-authored “Microservice Architecture” (June 2016). Amundsen’s 2013 collaboration with Leonard Richardson “RESTful Web APIs” and his 2011 book, “Building Hypermedia APIs with HTML5 and Node”, are common references for building adaptable Web applications. His latest book “Design and Build Great APIs” for Pragmatic Publishing is scheduled for release in early 2019.
Abstract: As we move closer to enabling autonomous agents on the Web, we have the opportunity to blend what we’ve learned from related fields such as hypermedia-based learning systems, the WWW, microsevices, machine learning, AI, and even early chatbots like Eliza, Parry, and others. At the same time, Gibson’s Affordance Theory has been applied to the Web as both client technology (the Web Browser) and network software architecture (Fielding’s REST) as well as to human-machine interactions via Donald Norman. Finally, what can we learn from distributed intelligence such as ant colonies on how to accomplish complex tasks over time using limited communication and constrained memory and processing power?
This talk explores the intersection of biology, psychology, and techology and poses a number of questions for discussion. What can we learn from both the machine and human-side of agency and affordance as we map out the future of increased autonomy on the Web? How much of our future success depends on smarter agents, richer runtime ecosystems, and swarm behavior? What can we do today to advance the use and safety of hyperagent-driven systems and what does hyper agency “look like” today?
Invited Contribution
Alessandro Ricci: Levels of Abstraction in Designing and Programming Systems of Cognitive Agents
Speaker bio: Alessandro Ricci is an Associate Professor of the Department of Computer Science and Engineering, University of Bologna (Italy). His research interests concern agents and multi-agent systems as a paradigm for modeling and engineering software systems. His main research contributions are in the context of MAS programming and Agent-Oriented Software Engineering. These include: the “artifact” abstraction in agent and the A&A (Agents and Artifacts) conceptual model; the CArtAgO Platform for environment programming; the JaCaMo platform, integrating CArtAgO with Jason and Moise. More recently, a main part of his research effort has been spent on Mirror Worlds and Augmented Worlds, i.e. kinds of smart environments integrating multi-agent systems with “augmentation technologies” such as Augmented Reality/Mixed Reality to Wearable and pervasive computing.
Abstract: If a main inspiration for the OOP paradigm is our world of objects, the main inspiration for the agent paradigm is our world of humans, from how individual humans autonomously do their tasks, to how they communicate, how they design and use their tools and artifacts to work with, how they coordinate and cooperate in organisations. This level of abstraction makes the paradigm particularly interesting for the design and development of complex systems, featuring different degrees of decentralisation, distribution, openness, flexibility, autonomy. The objective of this talk is first to introduce some key points of this paradigm - referred to as Multi-Agent Oriented Programming (MAOP) - in practice, using JaCaMo as a concrete platform. Second, to provide some points of reflection and discussion about MAOP and the development of Hypermedia MAS.
Accepted Papers
MAMS: Multi-Agent MicroServices (Regular paper)
Authors: Rem Collier, Eoin O’Neill, David Lillis, and Gregory O’Hare
Abstract: This paper explores the intersection between microservices and Multi-Agent Systems (MAS), introducing an approach to building MAS known as Multi-Agent MicroServices (MAMS). This approach is motivated in the context of the main properties of microservices and is illustrated through a worked example of a Vickrey Auction implemented as a microservice. The motivation for this work is to facilitate the creation of MAS that can be deployed using the same infrastructure as Hypermedia Systems; offering a closer and more natural integration with hypermedia resources. Further, we believe that our approach enables the creation of reusable components that can be interwoven within both larger MAS and more traditional microservices ecosystems.
Digital Companion for Industry (Short paper)
Authors: Mareike Kritzler, Jack Hodges, Dan Yu, Kimberly Garcia, Hemant Shukla, and Florian Michahelles
Abstract: The industrial domain offers a high degree of standardization, a variety of very specialized use cases, and an abundance of resources. These characteristics provide perfect conditions for Digital Companion systems. A Digital Companion is a cognitive agent that assists human users by taking on three roles: as guardians, assistants or mentors, and partners. This paper describes the characteristics, conceptual architecture, use cases and open challenges regarding Digital Companions for industry.
Towards Enabling Internet-Scale Context-as-a-Service (Short paper)
Authors: Alexandru Sorici, Andrei Olaru, and Adina Magda Florea
Abstract: Deploying context management systems at a global scale comes with a number of challenges and requirements. We argue that the hypermedia model and the agent-oriented paradigm help achieve the vision of Context-as-a-Service. We categorize challenges according to context processing concerns and use a scenario to exemplify how the proposed architectural principles help overcome the challenges.