unik4950/9950 - multiagent systems lecture 1 introduction … · highlights lecture 1 –...
TRANSCRIPT
![Page 1: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/1.jpg)
UNIK4950/9950 - Multiagent systems Lecture 1
Introduction Jonas Moen
![Page 2: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/2.jpg)
23.08.2017 3
My background
• Scientist FFI, leader swarm group, Research in MultiX (Multi-agent systems, multi-function systems, multi-objective optimization)
• Associate professor UiO/ITS, teacher and supervisor • Formal academic background
– PhD in evolutionary computation, UiO/Robin 2013 – Master in economics, NHH Bergen 2002 – Master in physics, UiB 1999
![Page 3: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/3.jpg)
23.08.2017 4
What are Multi-Agent Systems (MAS)?
Multiagent systems are systems composed of multiple interacting computing elements, known as agents, [Wooldridge, 2009] 1. Agents are capable of autonomous actions satisfying
design objectives 2. Agents engage in social activities like cooperation,
coordination, negotiation, competition, and the like.
![Page 4: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/4.jpg)
23.08.2017 5
What are multiagent systems?
1. A natural metaphor for artificial social systems.
2. An appropriate software paradigm for modelling & building
massive open distributed systems.
![Page 5: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/5.jpg)
23.08.2017 6
Reserach goal of multiagent systems
To connect micro scale behaviour with macro scale (often emergent) properties/effects, and vice versa.
![Page 6: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/6.jpg)
23.08.2017 7
Outline of the course
Part I - Agents Part II - Game theory Part III - Swarm intelligence Weekly lecture, 2 hours theory and 1 hour exercises To pass course, 2 ‘obliger’ + exam
![Page 7: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/7.jpg)
23.08.2017 8
Books in the course
«Multiagent systems», Michael Wooldrigde, 2009
«Swarm Intelligence» Dorigo et al., 1999
![Page 8: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/8.jpg)
23.08.2017 9
Highlights lecture 1 – Introduction*
• What is a multiagent system? – Key concepts – In relation to trends in computing – When to use and apply, and how to avoid pitfalls
• A brief history of multiagent systems • Some examples of multiagent system research and
application
*Wooldridge, 2009: chapter 1, 9, 10, preface and appendix A.
![Page 9: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/9.jpg)
23.08.2017 10
Trends in computing
5 important and continuing trends [Wooldridge, 2009] 1. Ubiquity 2. Interconnection 3. Intelligence 4. Delegation 5. Human-orientation
![Page 10: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/10.jpg)
Trends in computing
1. Ubiquity The reduced cost of computation (i.e. Moore’s Law) has made it possible to introduce processing power in almost all devices and places.
23.08.2017 11
![Page 11: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/11.jpg)
Trends in computing
2. Interconnection Computers are networked into large distributed systems (i.e. the Internet)
23.08.2017 12
![Page 12: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/12.jpg)
Trends in computing
3. Intelligence The increased complexity of tasks that we are able to automate and computerize in terms of research into algorithm design (i.e. AI)
23.08.2017 13
![Page 13: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/13.jpg)
Trends in computing
4. Delegation Implies that we give control to computer systems (e.g. safety-critical tasks in aerospace and energy)
23.08.2017 14
![Page 14: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/14.jpg)
Trends in computing
5. Human-orientation Human-machine interface history; from assembler languages, through procedural abstraction, via abstract data types, to object-oriented programming and, now most recently, agent-based systems.
23.08.2017 15
![Page 15: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/15.jpg)
Major challenges in software development
1. With respect to ubiquity, interconnection and AI, how do we develop techniques that exploit this massive processing power without explicitly telling it what to do? (e.g. ~1010 processing units) Can we make an optimal system that is not hard-coded? Do we need an alternative software paradigm in order to tap into this immense computing power?
23.08.2017 16
![Page 16: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/16.jpg)
Major challenges in software development
2. With respect to delegation and human-machine interaction, how do we build computer systems that can act effectively on our behalf? They must be
– Independent and autonomous – Represent our best interests
In a dynamical world of other agents and humans (on the net, in cyber-physical systems, robots, autonomous vehicles, and in the social sciences and biological domain).
23.08.2017 17
![Page 17: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/17.jpg)
Multiagent systems
A relatively new field in computer science trying to answer and these 5 trends in software development 1. An agent is a computer system capable of independent
action on behalf of its user or owner. 2. A multiagent system consist of many such agents
interacting with each other (through some network or sensor system)
Agents need skills and abilities to cooperate, coordinate and negotiate with each other on behalf of their users.
23.08.2017 18
![Page 18: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/18.jpg)
Key challenges with multiagent systems
1. The agent design problem (the micro level) How do we build agents that are capable of independent, autonomous action in order to successfully carry out tasks that we delegate to them?
2. The social design problem (the macro level) How do we build agents that are capable of interacting with other agents in order to successfully carry out the tasks that we delegate to them, especially when the agents do not share common goals or intentions?
23.08.2017 19
![Page 19: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/19.jpg)
Key challenges with multiagent systems
A truly successful multiagent system makes an explicit connection between the micro level and the macro level modelling of the complex system.
23.08.2017 20
![Page 20: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/20.jpg)
Social abilities of multiagent systems
1. How can cooperation emerge in societies of self-interested agents?
2. How can self-interested agents recognize when their desires and beliefs (goals and actions) conflict and avoid resorting to conflict?
3. How can autonomous agents coordinate their activities so as to cooperatively achieve goals?
23.08.2017 21
![Page 21: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/21.jpg)
Social abilities of multiagent systems
4. What sort of common language can agents use to communicate their desires and beliefs?
5. How can we support this kind of cooperative system when agents have different software and hardware platforms?
And most importantly, multiagent systems are based on artificial computational entities, that often are conceptually inspired by social human systems and biologic systems.
23.08.2017 22
![Page 22: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/22.jpg)
Some views of the field
Agents as paradigm for software engineering Some researchers believe that, in the future, computation can be viewed basically as a process of interaction. Agents seems like a strong candidate for this new paradigm.
23.08.2017 23
![Page 23: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/23.jpg)
Some views of the field
Agents as paradigm for software engineering 1. Self-interested computation
Multiagent systems focus on agents maximizing desires and beliefs, leading to intentions and actions. Previously, structured nodes and data sets.
2. The grid (e.g. the Internet) Multiagent systems focus on cooperative problem solving in dynamic environments. Previously, focus was mainly on middelware development.
23.08.2017 24
![Page 24: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/24.jpg)
Some views of the field
Agents as paradigm for software engineering 3. Ubiquitous computing
Multiagent systems focus on massive cooperative behaviour. Previously, we had mainframes, personal computers and HTML.
4. The Semantic Web A language for the meaning of information on web pages, called the semantic markup. Agents would be able to analyse all the data on the net. In contrast to the limiting web language of HTML
23.08.2017 25
![Page 25: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/25.jpg)
Some views of the field
Agents as paradigm for software engineering 5. Automatic computing
Systems that heal themselves and adapt autonomously to changing circumstances.
23.08.2017 26
![Page 26: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/26.jpg)
Some views of the field
Agents as tools for understanding social behaviour Simulating system dynamics from first-principle or collective phenomena, give insight into the micro/macro scale modelling. 1. Societies: EOS models Palaeolithic culture in France,
[Doran and Palmer, 1995] 2. Biology: Ants and bees, chemotaxis and Levy flight 3. Economy: Trading stock robots 4. Cooperation: Games of cooperation and competition 23.08.2017 27
![Page 27: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/27.jpg)
The relation between multiagent systems and
1. Distributed/concurrent systems – Agents are autonomous decision-makers at run-time – Self-interested agents of different desires and beliefs
23.08.2017 28
![Page 28: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/28.jpg)
The relation between multiagent systems and
2. Artificial intelligence – Classical AI is mostly concerned with components of type learning,
planning, reasoning, etc – Multiagent systems are sometimes wrongly said to focus on the
integration of all these AI components – Multiagent systems do not need very complex agent AI in order to
model system behaviour – Classical AI has mostly ignored the social aspect of agency
23.08.2017 29
![Page 29: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/29.jpg)
The relation between multiagent systems and
3. Game theory – Multiagent systems are the application side of game theory, often
with focus on computational aspects and approximate solutions. – Multiagent systems do not require (self-interested) rational agents.
23.08.2017 30
![Page 30: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/30.jpg)
The relation between multiagent systems and
4. The social sciences – Multiagent systems take inspiration from real human and biological
societies. – Societies can be modelled and simulated by multiagents systems.
23.08.2017 31
![Page 31: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/31.jpg)
When is an agent-based solution appropriate?
[Bond and Gasser, 1988; Jennings and Wooldridge, 1998] 1. The environment is open, or at least highly dynamic,
uncertain or complex. In these settings autonomous agents might be the only solution.
2. Agents are natural metaphors for – Societies, organizations, business, etc. – Intelligent interface as in ‘expert assistant’.
23.08.2017 32
![Page 32: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/32.jpg)
When is an agent-based solution appropriate?
3. Distribution of data, control or expertise When centralized solutions are difficult (like the synchronization of many autonomous databases).
4. Legacy software Wrap legacy software in an agent layer.
23.08.2017 33
![Page 33: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/33.jpg)
Some pitfalls of agent-based development
[Wooldridge, 2009] 1. Agents as ‘silver bullets’
– You oversell agents – You get dogmatic about agents – You see agents everywhere – You can’t justify why you adopt agents
23.08.2017 34
![Page 34: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/34.jpg)
Some pitfalls of agent-based development
2. Agents as ‘software infrastructure’
– You focus on implementing optimal infrastructure – You decide you want your own agent architecture – You build a generic solution to a one-off problem – Your agents use too much AI
23.08.2017 35
![Page 35: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/35.jpg)
Some pitfalls of agent-based development
3. Multiagent systems as ‘dynamic distributed systems’
– Your design does not exploit concurrency – You forget you are designing multithreaded software – You have too few agents – Your agents interact too freely for analysis
23.08.2017 36
![Page 36: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/36.jpg)
A brief history of multiagent systems research
A history of agents The notion of ‘agent’ is evident in early AI literature, e.g. the Turing test and the term AI coined by John McCarthy at the famous Dartmouth workshop in 1956.
23.08.2017 37
![Page 37: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/37.jpg)
A brief history of multiagent systems research
1940-1980: Symbolic AI Until the mid 1980s the AI community was mainly occupied with formal logic and deductive reasoning systems known as symbolic AI. STRIPS [Fikes and Nilsson, 1971] is prominent example of this era. AI researchers were working in subfields like learning, planning, reasoning and so forth, but not really interested in a ‘holistic’ approach to agency.
23.08.2017 38
![Page 38: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/38.jpg)
A brief history of multiagent systems research
1980s: ‘new AI’ In the mid 80s the ‘new AI’, also called ‘behavioural AI’ and ‘reactive AI’ were introduced in opposition to symbolic AI. 1. Inspiration from biology, spawning new research fields like
evolutionary computation, Alife, swam intelligence,etc. 2. Realistic scenarios were adopted in research and
development. 3. Hybrid agents combining deliberate reasoning and reactive
behaviour in same architecture. 23.08.2017 39
![Page 39: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/39.jpg)
A brief history of multiagent systems research
A history of multiagent systems Multiagent systems, as a research field of its own, did not gain widespread popularity and attention in the AI community until early 1980s.
23.08.2017 40
![Page 40: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/40.jpg)
A brief history of multiagent systems research
1960-1970: The blackboard model [Newell, 1962] 1970s: The actor model of computation, [Hewitt, 1977] • Actors are reactive but not proactive • Closely connected to the development of object-oriented
languages • The Contract Net [Smith, 1977] introduced the economic
metaphore in multiagent systems 23.08.2017 41
![Page 41: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/41.jpg)
A brief history of multiagent systems research
1980-1990: The ‘distributed AI’ paradigm • Spawned several workshops, conferences and publications • The concept of self-interested agents introduced game
theory to multiagent systems, [Rosenschein and Genesereth, 1985].
• MACE (MultiAgent Computing Environment) pioneered the acquaintance model, [Gasser et al., 1987]
• ARCHON (Architecture for cooperating heterogenous online systems) as first real industrial application of agent systems. Financed by EU in late 1980s, 14 partners in 9 countries.
23.08.2017 42
![Page 42: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/42.jpg)
A brief history of multiagent systems research
1990->: The Internet • In 1990s interest in multiagent systems grew very rapidly. • E-commerce represents a natural and lucrative application
domain with the introduction of online auctions, shopping agents, personal web robots and now IoT.
• Standardization became a major issue and by 2000 FIPA emerged as the central standard of multiagent systems.
• Multiagent systems are now mainstream computer science.
23.08.2017 43
![Page 43: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/43.jpg)
A brief history of multiagent systems research
2010->: Autonomous mobile robots In the beginning of the 2010s autonomous robots started to move out of the academic university laboratories and into real world field trials. This was due to reduced cost and increased availability of appropriate hardware (and SatNav).
23.08.2017 44
![Page 44: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/44.jpg)
A brief history of multiagent systems research
2010->: Autonomous mobile robots • Autonomus systems take inspiration from the like of
multiagent system research, evolutionary computation methods, swarm intelligence, etc.
• Produced massive interest in autonomous cars and in novel applications like swarm systems of drones and other vehicles, applied to transport, energy, agriculture, etc.
• Note that mobile agents are also software programs that can transmit themselves over the net and recommence execution at remote site.
23.08.2017 45
![Page 45: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/45.jpg)
Multiagent system research and applications
Some examples: 1. A software perspective on IoT 2. Modelling social behaviour: Hotelling’s beach 3. What is cooperation? Nowak’s spatial game 4. Modern swam robotics: Locust and Kilobots
23.08.2017 46
![Page 46: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/46.jpg)
Applications
Multiagent system applications can be divided into two groups 1. Distributed systems
Agents are processing nodes in a distributed multiagent system, with emphasis on ‘multi’.
2. Personal software assistants Agents as proactive assistents to users, with emphasis on ‘individual’.
23.08.2017 47
![Page 47: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/47.jpg)
A software perspective on IoT
23.08.2017 48
Image: IoT Tech Expo
![Page 48: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/48.jpg)
A software perspective on IoT
23.08.2017 49
Image: IBM Big Data & Analytics Hub
![Page 49: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/49.jpg)
A software perspective on IoT
23.08.2017 50
Image: IBM Big Data & Analytics Hub
![Page 50: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/50.jpg)
A software perspective on IoT
23.08.2017 51
Image: IBM Big Data & Analytics Hub
Order washing powder
![Page 51: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/51.jpg)
A software perspective on IoT
23.08.2017 52
Images: IBM Big Data & Analytics Hub Amazon Tide Push Button
![Page 52: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/52.jpg)
A software perspective on IoT
23.08.2017 53
Image: IBM Big Data & Analytics Hub
Optimize electric consumption
![Page 53: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/53.jpg)
A software perspective on IoT
23.08.2017 54
Images: Aidon Power Grid Management SmartGrid.gov, DoE
![Page 54: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/54.jpg)
A software perspective on IoT
23.08.2017 55
Images: Aidon Power Grid Managament SmartGrid.gov, DoE
![Page 55: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/55.jpg)
Modelling social behaviour: Clustering of services
23.08.2017 56
Image: TedEd
![Page 56: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/56.jpg)
Modelling social behaviour: Hotelling’s beach*
23.08.2017 57
Image: TedEd *Hotelling, «Stability In Competition», TEJ, 1929.
![Page 57: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/57.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 58
Image: TedEd
![Page 58: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/58.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 59
Image: TedEd
![Page 59: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/59.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 60
Image: TedEd
![Page 60: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/60.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 61
Image: TedEd
![Page 61: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/61.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 62
Image: TedEd
The Socially Optimal Outcome:
![Page 62: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/62.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 63
Image: TedEd
The Socially Optimal Outcome: Is this stable?
![Page 63: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/63.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 64
Image: TedEd
![Page 64: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/64.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 65
Image: TedEd
![Page 65: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/65.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 66
Image: TedEd
Pareto Optimal Outcome: Nash- equilibrium
![Page 66: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/66.jpg)
Modelling social behaviour: Hotelling’s beach
23.08.2017 67
Image: TedEd
![Page 67: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/67.jpg)
Modelling social behaviour: Clustering of services
23.08.2017 68
Socially optimal solution: Pareto optimal solution: Sales:1/2 Customer: 1/4 Sales:1/2 Customer: 1/2
![Page 68: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/68.jpg)
Modelling social behaviour: Clustering of services
23.08.2017 69
Now, based on Hotelling’s analysis, how would a set of mobile autonomous 4G base stations allocate in order to cover an ad-hoc crowd of people addicted to SoMe? 1. Self-interested robots would cluster in Nash-equilibrium 2. Altruistic robots would spread out to the social optimum
The really big question: is it possible to predict a socially optimal solution using rational self-interested agents?
![Page 69: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/69.jpg)
What is cooperation? Nowak’s spatial game*
23.08.2017 70
The cooperation puzzle: • How can cooperation be maintained when there is a clear
advantage in defecting? • How can cooperation be established when there is a clear
advantage in defecting?
Some researchers, among them Nowak, believe that life itself presuppose cooperation (which is always under the threat of opportunism).
*Nowak et al., ”Evolutionary games and spatial chaos”. Nature 1992
![Page 70: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/70.jpg)
What is cooperation? Nowak’s spatial game
23.08.2017 71
Agents 1 and 2 cooperate on task T Π 𝑇𝑇 = 𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈 − 𝐶𝐶𝐶𝐶𝐶𝐶𝑈𝑈 Agent 1 and 2 cooperate Π = 𝑈𝑈/2 − 𝐶𝐶/2 for each Agent 1 or 2 defect Π = 𝑈𝑈/2 for defector Π = 𝑈𝑈/2 − 𝑘𝑘 ∙ 𝐶𝐶 for cooperater Agent 1 and 2 defect Π = 0 for each
![Page 71: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/71.jpg)
What is cooperation? Nowak’s spatial game
23.08.2017 72
D D C C C C D D C
![Page 72: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/72.jpg)
NetLogo of NWU Framework for MAS Blue = 𝐶𝐶 (Cooperator) Red = 𝐷𝐷 (Defector)
What is cooperation? Nowak’s spatial game
23.08.2017 73
![Page 73: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/73.jpg)
What is cooperation? Nowak’s spatial game
23.08.2017 74
Professor Martin Nowak Harvard University
![Page 74: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/74.jpg)
Modern swam robotics
23.08.2017 75
Some examples: • FFI Sparrows • NPS Locust • Harvard Kilobots
Image: UiO/FFI
![Page 75: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/75.jpg)
Modern swam robotics: FFI Sparrows
23.08.2017 76
Why UAVs?
Image: Thoresen et al., 2014
Frequency 2.4 GHz
![Page 76: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/76.jpg)
Multiple UAVs improve geolocation
23.08.2017 77
Uncertainty 22 m
Average over 7 measurements
Image: Thoresen et al., 2014
![Page 79: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/79.jpg)
From the book - The vision thing
1. Space-probe fixing an unexpected system failure on its own.
2. A nearby autonomous Air-Traffic-Control system backs up a failed local ATC.
3. Web based agents negotiate a good vacation package on behalf of user.
4. EU puts out thousands of contracts for public tender. How much cost could be saved by using autonomous agents instead?
23.08.2017 80
![Page 80: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/80.jpg)
Applications
Agents for: 1. Workflow and business process management – customer
request handled by corporate divison/department agents 2. Human-computer interface - proactive expert assistents 3. Electronic commerce – best deal on the net 4. Social simulations – EOS project and policy modelling 5. Industrial applications – ATC and ARCHON 6. Virtual environments – believable agents in games,
cinemas, etc 23.08.2017 81
![Page 81: UNIK4950/9950 - Multiagent systems Lecture 1 Introduction … · Highlights lecture 1 – Introduction* • What is a multiagent system? – Key concepts – In relation to trends](https://reader036.vdocument.in/reader036/viewer/2022070211/60fcf0b4db1b20796c231fbd/html5/thumbnails/81.jpg)
Summary of lecture 1
Multiagent systems are about ‘computers’ working together in: • Pure software systems • Cyber-physical systems • Robotics systems and the research ranges from applied technology to fundamental science.
23.08.2017 82