Workshop on: Collective Adaptation in Very Large Scale Ubicomp:
Towards a Superorganism of Wearables
-----------------
Workshop at Ubicomp 2016: September 13, 2016
-----------------
SUMMARY AND THEME
The 3rd workshop (after the 2014 WS in Seattle and the 2015 WS in Osaka)
asks questions on the potential and opportunities of turning massively
deployed wearable systems to a globe-spanning superorganism of
socially interactive personal digital assistants. While individual
wearables are of heterogeneous provenance and typically act
autonomously, it stands to reason that they can (and will) self-organize
into large scale cooperative collectives, with humans being mostly
out-of-the-loop. A common objective or central controller may thereby
not be assumed, but rather volatile network topologies, co-dependence
and internal competition, non-linear and non-continuous dynamics, and
sub-ideal, failure-prone operation. We refer to these emerging massive
collectives of wearables as a "superorganism", since they exhibit
properties of a living organism (like e.g. 'collective intelligence') on
their own. One essential aspect of such globe-spanning collective
ensembles is that they often exhibit properties typically observed in
complex systems, like (i) spontaneous, dynamic network configuration,
with (ii) individual nodes acting in parallel, (iii) constantly acting
and reacting to what the other agents are doing, and (iv) where the
control tends to be highly dispersed and decentralized. If there is to
be any coherent behavior in the system, it (v) has to arise from
competition and cooperation among the individual nodes, so that the
overall behavior of the system is the result of a huge number of
decisions made every moment by many individual entities.
In order to properly exploit such superorganisms, this workshop concerns
itself with the development of a deeper scientific understanding of the
foundational principles by which they operate. To this end, the workshop
attempts to address the following foundational research concerns:
- Understanding the trade-offs between the power of top-down (by design)
adaptation means and bottom-up (by emergence) ones, also by studying how
the two approaches co-exist in modern wearable ICT systems, and possibly
contributing to smoothing the tension between the two approaches.
- Understanding the "power of the masses" principle as far as
participatory wearable ICT processes are involved. In particular, this
implies understanding how and to what extent even very simple collective
phenomena and algorithms - when involving billions of wearables - can
express forms of intelligence much superior than that of more
traditional AI techniques.
- Understanding the issue of diversity and of diversity increase in
complex systems and in service/data systems and how diversity of
structure and behavior is currently accommodated in wearable ICT
systems. As of now, most studies focus on a limited number of different
classes, which is far from approximating the diversity of existing
systems.
- Laying down new foundations for the modelling of large-scale Human-ICT
organisms and their adaptive behaviors, also including lessons from
applied psychology, sociology, and social anthropology, other than from
systemic biology, ecology and complexity science.
- Identifying models and tools by which individual organs of the systems
can influence and direct "by design" the emergent adaptive behavior of
the whole system, or at least of substantial parts of it.
Further, the workshop attempts to address the following systems research
concerns:
- Opportunistic information collection: Systems need to be able to
function in complex, dynamic environments where they have to deal with
unpredictable changes in available infrastructures and learn to
cooperate with other systems and human beings in complex self-organized
ensembles.
- Collaborative Reasoning and Emergent Effects: Reasoning methods and
system models are needed that combine machine learning methods with
complexity theory to account for global emergent effects resulting from
feedback loops between collaborative, interconnected devices and their
users.
- Social Awareness: Whereas today's context-aware systems are able to
make sense of the activity of single users and their immediate
environment, future systems should be able to analyze, understand and
predict complex social phenomena on a broad range of spatial and
temporal scales. Examples of the derived information could be: shifts in
collective opinions and social attitudes, changes in consumer behavior,
the emergence of tensions in communities, demographics, migration,
mobility patterns, or health trends.
PRELIMINARY AGENDA
8:30 Registration / Help desk opens
12:00 Lunch
13:30 Workshop Opening
14:00 Keynote - "Collective Addaptive Wearables - Technology
Needs and Societal Breakthroughs"
Franco Zambonelli
Universita di Modena e Reggio Emilia, Italy
15:00 Paper Session 1
Bernhard Anzengruber et al., "Understanding Individuals
in Masses: Case Study Vienna City Marathon"
15:30 Coffee break
16:00 Paper Session 2
Ali Farahani et al., "Self-* Properties in Collective
Adaptive Systems"
Mirko Viroli et al., "On Execution Platforms for
Large-Scale Aggregate Computing"
Bernhard Anzengruber et al., "AL17 - Triggering
Interactions in an Ecosystem of Wearables"
Andrea Omicini et al., "Challenges of Decentralized
Coordination in Large-scale Ubicomp Systems"
18:00 Drafting of Workshop White Paper
19:00 Workshop closed
ORGANIZERS
- Alois Ferscha (University of Linz, Austria)
- Paul Lukowicz (DFKI, Germany)
- Franco Zambonelli (Universita di Modena e Reggio Emilia, Italy)
All Details and News: http://www.pervasive.jku.at/ubicomp16/
|
