Abstract: While AI has recently produced impressive systems that achieve human-like performance at challenging tasks, these systems tell us very little about how human intelligence works. In particular, they do not address the problem of composing knowledge and behaviour incrementally – a phenomenon that is pervasive in individual and collective human intelligence. We argue that achieving more human-like AI requires focusing on diversity in reasoning and behaviour among humans and artificial agents, and that developing systems capable of dealing with such diversity is key to achieving more human-like AI. In these systems intelligence should not only be measured in terms of how a system performs at a certain task, but also in terms of the properties of the process by which each component combines its knowledge and behaviour with that of others, just like humans do.

Citation: M. Rovatsos. Diversity-Awareness – The Key to Human-Like Computing? In S. Muggleton et al, editor, Twentieth Workshop on Machine Intelligence (MI 20), Windsor Park, UK, 23-25 October, 2016. In press.

Download: http://bit.ly/2jd0oMI

Abstract: Sharing economy applications need to coordinate humans, each of whom may have different preferences over the provided service. Traditional approaches model this as a resource allocation problem and solve it by identifying matches between users and resources. These require knowledge of user preferences and, crucially, assume that they act deterministically or, equivalently, that each of them is expected to accept the proposed match. This assumption is unrealistic for applications like ridesharing and house sharing (like airbnb), where user coordination requires handling of the diversity and uncertainty in human behaviour.
We address this shortcoming by proposing a diversity-aware recommender system that leaves the decision-power to users but still assists them in coordinating their activities. We achieve this through taxation, which indirectly modifies users’ preferences over options by imposing a penalty on them. This is applied on options that, if selected, are expected to lead to less favourable outcomes, from the perspective of the collective. The framework we used to identify the options to recommend is composed by three optimisation steps, each of which has a mixed integer linear program at its core. Using a combination of these three programs, we are also able to compute solutions that permit a good trade-off between satisfying the global goals of the collective and the individual users’ interests. We demonstrate the effectiveness of our approach with two experiments in a simulated ridesharing scenario, showing: (a) significantly better coordination results with the approach we propose, than with a set of recommendations in which taxation is not applied and each solution maximises the goal of the collective, (b) that we can propose a recommendation set to users instead of imposing them a single allocation at no loss to the collective, and (c) that our system allows for an adaptive trade-off between conflicting criteria.

Citation: P. Andreadis, S. Ceppi, M. Rovatsos, and S. Ramamoorthy. Diversity-Aware Recommendation for Human Collectives. In Proceedings of the 1st International Workshop on Diversity-Aware Artificial Intelligence (DIVERSITY 2016), The Hague, The Netherlands, 2016

Download: http://bit.ly/2jp8rUr

Abstract: Past research has studied two approaches to utilise pre-defined policy sets in repeated interactions: as experts, to dictate our own actions, and as types, to characterise the behaviour of other agents. In this work, we bring these complementary views together in the form of a novel meta-algorithm, called Expert-HBA (E-HBA), which can be applied to any expert algorithm that considers the average (or total) payoff an expert has yielded in the past. E-HBA gradually mixes the past payoff with a predicted future payoff, which is computed using the type-based characterisation. We present results from a comprehensive set of repeated matrix games, comparing the performance of several well-known expert algorithms with and without the aid of E-HBA. Our results show that E-HBA has the potential to significantly improve the performance of expert algorithms.

Citation: S. Albrecht, J. Crandall, S. Ramamoorthy, E-HBA: Using Action Policies for Expert Advice and Agent Typification, In Proc. AAAI-Workshop on Multiagent Interaction without Prior Coordination (MIPC), 2015.

Download: http://bit.ly/2j5V9MT

Abstract: We report work in progress on the role of models in the formation and maintenance of collectives in Hybrid Diversity-Aware Collective Adaptive Systems (HDA-CASs). HDA-CASs utilize hybrid computations involving machines and humans operating in collectives in a way that manages and leverages the diversity of collectives and machine-based computation. Here we explore the role of models in helping to constitute particular collectives and how models help shape the response of the collective. It appears that models are a potentially critical resource in collecting, sharing and acting on data gathered from the operation of CASs. This points to the potential role for models in the design of HDA-CASs. In particular we are interested in how models provide a sense of identity for a collective and can provide resources that shape the potential for collective action.

Citation: Stuart Anderson, Mark Hartswood and Marina Jirotka (2014). Reflection, collectives and adaptation: the role of models in the design of Collective Adaptive Systems. In the 2014 IEEE Eighth International Conference on Self-Adaptive and Self-Organizing Systems Workshops (SASOW 2014).

Download: http://bit.ly/2jmQMN8

You can find the workshop proceeedings on the ECAI website, or directly download them from here.

To get a glimpse of what the day was like, you can follow #diversityAwareAI on Twitter.

Michael Rovatsos opens DIVERSITY 2016 with a quote from Marvin Minsky.

Marvin Minsky: “What magical trick makes us intelligent? The trick is that there is no trick. The power of intelligence stems from our vast diversity, not from any single, perfect principle.”

The workshop seeks to explore diversity as a phenomenon that both poses a challenge for AI in terms of dealing with and managing diversity in an intelligent system (or ecosystem of intelligent human and/or artificial agents) and presents an opportunity in terms of leveraging diversity (for example through processes like crowdsourcing and collaborative knowledge production) to achieve human-like (and human-friendly) capabilities in more open-ended, incrementally evolving, and interactive AI systems.

The workshop seeks to explore diversity as a phenomenon that both poses a challenge for AI in terms of dealing with and managing diversity in an intelligent system (or ecosystem of intelligent human and/or artificial agents) and presents an opportunity in terms of leveraging diversity (for example through processes like crowdsourcing and collaborative knowledge production) to achieve human-like (and human-friendly) capabilities in more open-ended, incrementally evolving, and interactive AI systems. For more details, please see the original announcement, here.

The deadline for paper submission to HAIDM 2015 has been extended to the February the 23rd.

Topics covered by this year’s HAIDM include amongst others:

• Trust between humans and agents
• Smart society applications including energy systems, ride-sharing, healthcare augmentation, and disaster response
• Coalition formation and optimisation models involving models of agents and humans
• Human-Robot Interaction
• Crowdsourcing
• Citizen science
• Enhanced models of human behaviour and theory of human behaviour
• Applications of human behaviour models,
• Behavioural game theory
• Techniques for learning human behaviour
• Quantitative and qualitative studies of human-agent interaction

Important Dates:
11th February: 23rd of February:  Submission deadline
10th March: Notifications
19th March: Deadline for Camera-Ready copies

For more details, including the submission procedure, visit the HAIDM 2015 site here.

SmartSociety is proud to have co-organised this event!

Important Dates

Submission deadline: February 10th, 2014
Notification of acceptance: March 3, 2014
Workshop takes place: May 6th, 2014

Abstract. The physical space is populated with sense, which is gathered, manipulated and reinvented during its interpretation. Here is proposed a provisional list on how sense  imbricates places and some possible ways of making use of it for design purposes can be argued. Memory, emotions, reflexivity, sociality, and bewildering therefore become  qualities that can be explored in the physical space from HCI.

Author Keywords: Sense making, physical space, spatial reasoning, memory, emotions, reflexivity, sociality, bewildering.

ACM Classification Keywords: Design, Human Factors, Theory.

http://studiolab.ide.tudelft.nl/studiolab/sxdchi13/

Citation: S. Torsi (2013). Building Sense in the Infosphere. In Giaccardi, E.,Ciolfi, L., Hornecker, E., Speed, C., Bardzell, S. (2013) CHI 2013 Workshop “Exploration in Social Interaction Design”. Paris (FR), 28th. April 2013.

Download: http://bit.ly/1MBj7vW

Abstract. The physical and virtual dimensions of life are becoming more and more deeply interwoven. Society is merging with technology, giving rise to a global socio-technical ecosystem. In a society comprising people and machines as actors we often see people-to-people interactions mediated by machine and machine-to-machine interaction mediated by people. The speed and scale of this change and the differences in culture, language and interests make the problem of establishing effective means of communication and coordination increasingly challenging. Our vision, embodied in the SmartSociety project1, is that a new generation of systems, hybrid (i.e., including humans and artificial peers, as well as social groups), distributed, open and large-scale, is needed to tackle these issues. In such systems, multitudes of heterogeneous peers will produce and handle massive amounts of data; peers will join/leave the system following unpredictable patterns with no central coordination and will interoperate at different spatial and temporal scales. Aware of the ethical issues, and by identifying the right incentive schemes and privacy levels, these systems will assist individuals and collectives in their everyday activities, coping with the diversity of the world and working in the presence of incomplete and incorrect information.

http://eprints.biblio.unitn.it/4214/

Citation: F. Giunchiglia, V. Maltese, S. Anderson, D. Miorandi (2013). Towards Hybrid and Diversity-Aware Collective Adaptive Systems. First FOCAS Workshop on Fundamentals of Collective Systems @ECAL 2013.

Download: http://bit.ly/1STlQ0l

The 2nd Social-IST Workshop took place on May 29 in Edinburgh. Click here to find out more!