L. Steels. The origins of intelligence. In Proceedings on the Carlo Erba foundation, Workshop on Artificial Intelligence, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the simulation are the (hopefully) emerging patterns, and organization. Steels idea is that language can be looked at as a pattern of organization that is emerging in such multi agent systems (Steels, 1998) Furthermore he thinks the nature of intelligence in general can be described this way (Steels, 1996)) The basic method he applies to develop e.g. emergent common vocabularies and phoneme systems is the so called language game. The behavior of agents consists of a series of games that they play with randomly chosen other agents. One possible game for example is the discrimination game where the ....

....learning the language is not essentially different from the emergence of language. As he puts it : no separate mechanism for language acquisition is necessary because the mechanisms that explain the origin of language also explain how it is acquired be new agents entering the community. (Steels, 1996) Thus the model of the emergence of language and language evolution is also a model of language acquisition. Keeping this in mind let us take a look at how de Boer applies Steels approach for phoneme acquisition. First of all we have to note that de Boer uses only standalone vowels. His ....

Steels, L. (1996). The origins of intelligence. In Proceedings of the Carlo Erba Foundation Meeting on Artificial Life, Milano. Fundazione Carlo Erba.

....We also discuss work currently underway to allow the robot s behavior to be improved by the emergent representations. Our work builds on research from several disciplines. These include: behavior based robotics (Brooks and Stein 1994) the dynamical nature of representation and intelligence (Steels 1995, 1996), and the philosophical insights of Maturana and Varela (1980) and Clark (1997) on the self organizing nature of living systems and their coupling with their environments. Further support for our approach comes from Holland s (1986, 1998) ideas on emergence in the context of classifier systems, ....

....(Luger 1994) These models also provide suitable tests for the assertion that representations only have meaning in the context of embedding experiences. Our control architecture implicitly defines intelligence with the four characteristics of evolving complex adaptive systems proposed by Steels (1996). The first of these criteria is self maintenance (we prefer the term autopoeisis from Maturana and Varela (1980) who also describe a mutual maintenance relationship among system components) The remaining criteria for describing intelligence are adaptivity, information preservation, and, in ....

[Article contains additional citation context not shown here]

Steels, L. (1996). The origins of intelligence. In Proceedings of the Carlo Erba Foundation Meeting on Artificial Life . Berlin: Springer-Verlag.

....a robot must autonomously adapt to its changing surroundings, for instance in unmanned space landings. Our work builds on research from several disciplines. These include: work on behavior based robotics (Brooks and Stein 1994) work on the dynamical nature of representation and intelligence (Steels 1995 and 1996); work on the organization of living systems and their coupling with an environment (Maturana and Varela 1980, Clark 1997) and work on fluid representation in the unique architectures of the Copycat family from Mitchell and Hofstadter (Mitchell 1993) The active synthesis of representations from ....

Steels, L. (1996). The origins of intelligence. In Proceedings of the Carlo Erba Foundation Meeting on Artificial Life . Berlin: Springer-Verlag.

....of situated and embodied agents has been actively espoused by members of the behaviour based robotics community, in recognition of these problems, for many years (see An Architecture for Cooperation among Autonomous Agents. PhD Thesis David Jung 20 [Brooks, 1991; Matari#, 1992a; Pfeifer, 1995; Steels, 1996] for a selection) This hypothesis is called the physical grounding hypothesis [Brooks, 1990] By being embodied, a robot has a particular physiology that directly determines the structure of iconic representations, and thus indirectly influences indexical and symbolic ones. By being situated in ....

....is learnt Little animal behaviour is innate. Almost all behaviour is learnt with a small amount of pre wired behaviour or behavioural biases to bootstrap learning [McFarland and Bsser, 1993] A number of robotics researchers have recognised this problem, either implicitly or explicitly (e.g. [Steels, 1996d; Stein, 1997; Gaussier et al. 1998] For example, in Steels use of his process networks he resists labelling any set of processes as a behaviour. Any process may play a role in a number of difference behaviour systems ( Steels and Brooks, 1995] Chapter 5) Although some proposals for a ....

[Article contains additional citation context not shown here]

Steels, Luc, "The origins of intelligence", Proceedings of the Carlo Erba Foundation, Meeting on Artificial Life. Fondazione Carlo Erba. Milano, March 12, 1996.

.... is termed the frame problem (see Ford and Hayes, 1991; Pylyshym, 1987) The importance of situated and embodied agents has been actively espoused by members of the behavior based robotics community, in recognition of these problems, for many years (see Brooks, 1991; Mataric, 1992a; Pfeifer, 1995; Steels, 1996 for a selection) This hypothesis is called the physical grounding hypothesis (Brooks, 1990) Consequently, we adopted the behaviorbased approach for our implementation. 2.3 Cooperative biological systems In this subsection we describe five selected biological cooperative systems and classify ....

Steels, Luc 1996. The origins of intelligence, Proceedings of the Carlo Erba Foundation, Meeting on Artificial Life. Fondazione Carlo Erba. Milano.

....the frame problem (see [Ford and Hayes, 1991; Pylyshym, 1987] The importance of situated and embodied agents has been actively espoused by members of the behavior based robotics community, in recognition of these problems, for many years (see [Brooks, 19910 Hh#h. v # 1992a; Pfeifer, 1995; Steels, 1996] for a selection) This hypothesis is called the physical grounding hypothesis [Brooks, 1990] By being embodied, a robot has a particular physiology that directly determines the structure of iconic representations, and thus indirectly influences indexical and symbolic ones. By being situated ....

....in multi robot systems, an analogous mechanism is necessary. Since the first symbolic representational systems used by robots will be very simple in comparison to their human counterparts, the mechanisms for creating the necessary shared grounding will be correspondingly simple. See [Steels, 1997; Steels, 1996a] for related research on shared meaning creation. As part of our research into cooperative multi robot systems, we implemented various solutions to a cleaning task using two autonomous mobile robots. The robots are heterogeneous and one of our solutions required symbolic level communication. The ....

Steels, Luc, "The origins of intelligence", Proceedings of the Carlo Erba Foundation, Meeting on Artificial Life. Fondazione Carlo Erba. Milano, March 12, 1996.

No context found.

L. Steels. The origins of intelligence. In Proceedings on the Carlo Erba foundation, Workshop on Artificial Intelligence, 1996.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC