A System-Theoretic Analysis
of Focused Cognition,
and its Implications for
the Emergence of Self and Attention
Ben Goertzel
Novamente LLC
November
4, 2006
Abstract: A unifying framework for the description and
analysis of focused cognitive processes is presented, based on the
system-theoretic notions of forward and backward synthesis. Forward synthesis iteratively creates
combinations, seeded from an initial focus-set of mental items; backward
synthesis takes a set of mental items and tries to create them iteratively via
forward-synthesis. The utility of
a dynamic involving alternating forward and backward synthesis is
discussed. The phenomenal self and
the shifting focus of attention, two critical aspects of cognitive systems, are
hypothesized to emerge as strange attractors of this alternating dynamic. In a companion paper, this framework is
used to provide a systematic typology for the various cognitive agents in the
Novamente AI system, including probabilistic inference, evolutionary learning,
attention allocation, credit assignment, concept creation and others.
1. Introduction
Human cognition is complex, involving a combination of multiple complex
mechanisms with overlapping purposes.
I conjecture that this complexity is not entirely the consequence of the
ÒmessinessÓ of evolved systems like the brain. Rather, I suggest, any system that must perform advanced
cognition under severe computational resource constraints will inevitably
display a significant level of complexity and multi-facetedness.
This complexity, however, proves problematic for those attempting to
grapple with the mind from the perspective of science and engineering. In the context of experimental
psychology, it means that controlled experiments are very rarely going to be
able to get at the most interest aspects of intelligence and mind. And in the context of AGI design
(artificial general intelligence; see Goertzel and Pennachin, 2006), it means
that it is very difficult to construct detailed low-level cognitive mechanisms
in such a way as to give rise to desired high-level cognitive behaviors. In this paper I will focus on the AGI
aspect rather than the human-psychology aspect, but many of the same issues
exist in both cases.
I believe the complexity of mind can be grappled with effectively – both in human
psychology and in AGI – but only if theorists and practitioners take more
of a system-theoretic perspective and seek to understand both natural and
artificial intelligences as complex, self-organizing systems with dynamics
dominated by large-scale emergent structures. In past writings (Goertzel, 1993, 1994, 1997, 2006) I have
sought to take steps in this direction; and in this paper I will attempt to
push this programme further, by discussing the complex systemic cognitive
dynamics that I hypothesize to give rise to the critical emergent structures of
ÒselfÓ and Òattention.Ó
These particular emergent structures are obviously critical for AGI, for
human psychology and for mind science in general.
In (Goertzel, 2006) I follow up further on these concepts in an AGI
context, showing how the systems theoretic notions introduced here may be used
to give a systematic typology of the cognitive mechanisms involved in the
Novamente AGI architecture, and an explanation of why it seems plausible to
hypothesize that a fully implemented Novamente system, if properly educated in
the context of an embodiment and shared environment, could give rise to self
and attention as emergent structures.
The first theoretical step taken here is to introduce the general
notions of forward synthesis and backward
synthesis. as an elaboration of the
theory of component-systems and self-generating systems proposed in Chaotic
Logic; (Goertzel, 1994). The hypothesis is made that these
general schematic processes encompass all forms of focused cognition carried out in intelligent systems operating under
strongly limited computational resources.
Furthermore, I will lay stress here on the importance of the oscillatory
dynamic in which forward and backward synthesis processes repeatedly follow
each other. This fundamental
cognitive dynamic, I will argue, is important among other reasons because of
the attractors it leads to. The
second key theoretical step taken here is the hypothesis that fundamental
emergent structures such as the self and the Òmoving bubble of focused
attentionÓ may be fruitfully conceptualized as strange attractors of this
oscillatory cognitive dynamic.
2. AGI and the Complexity of Cognition
The vast bulk of approaches to AI and even AGI, I feel, deal with the
problem of the complexity of cognition by essentially ignoring it. For instance, many existing
approaches to AGI focus on some particular problem-solving mechanism and seek
to utilize this mechanism for a broad variety of purposes, or to explain why
the problems solved well by this mechanism are the most critical ones. The problem with this sort of approach
is that, given the combination of complex functions that is required of an AGI
system, it seems very difficult to find any single mechanism that is capable of carrying out
all the functions required of an AGI.
As an example of this I will cite an AGI approach of which I am very
fond: WangÕs (1995, 2006) NARS system.
NARS places uncertain logic at the center of intelligence, but my
feeling is that uncertain logic in itself is not adequate for procedure
learning nor for perceptual pattern recognition, to name just two of many
aspects. Cyc (Lenat and Guha,
1990) has a similar issue, placing crisp predicate logic at the center and then
seeking to augment it with a language-processing front end and context-specific
Bayesian nets, without confronting the issue that crisp theorem-proving may not
be adequate for carrying out most of the functionalities critical to general
intelligence.
Other AGI approaches take a hybrid strategy, in which an overall
architecture is posited and then various diverse algorithms and structures are
proposed to fill in the various functions defined by the architecture. The problem that arises here is
that, unless the different algorithms and structures are specifically designed
to work effectively together, the odds of them interoperating productively in a
sufficient variety of real-world situations are small. Much of intelligence consists of
emergent behaviors arising from the cooperative action of numerous complex
problem-solving mechanisms; but the appropriate emergent behaviors will not
simply emerge from the insertion of vaguely appropriate cognitive mechanisms
inside each of a set of boxes defined by a high-level cognitive theory. Rather, the cognitive mechanisms inside
each of the boxes must be specifically designed and tuned for operation in the
whole-cognitive-system context; i.e., with appropriate emergent behaviors and
structures in mind.
To make this point, I will cite another AGI architecture of which I am
very fond: Stan FranklinÕs (2006) LIDA architecture. LIDA is a very well
thought out architecture which is well grounded in cognitive science research,
but it is not clear to me whether the combination of learning mechanisms used
within LIDA is going to be appropriately chosen and tuned to give rise to the
emergent structures and dynamics characteristic of general intelligence, such
as the self and the moving bubble of attention. LIDA is a very general approach, which could be used as a
container for a haphazard assemblage of learning techniques, or for a carefully
assembled combination of learning techniques designed to lead to appropriate
emergence. So, this is not a
criticism of LIDA as such, but rather an argument that without concrete choices
regarding the specifics of the learning algorithms, it is not possible to tell
whether or not the LIDA system is going to be plausibly capable of a reasonable
level of general intelligence.
The Novamente design seeks to avoid these various potential problems
via the incorporation of a variety of cognitive mechanisms specifically
designed for effective interoperation and for the induction of appropriate
emergent behaviors and structures.
I believe this approach is conceptually sound, however it does have the
drawback of leading to a rather complex design in which the accurate
description and development of each component requires careful consideration of
all other components. For this
reason it is worthwhile to seek simplification and consolidation of cognitive
mechanisms, insofar as is possible.
In this paper I introduce a conceptual framework that has been developed
in order to provide a simplifying unifying perspective on the various cognitive
mechanisms existing in the Novamente design, and an abstract and coherent
argument regarding the dynamics by which these mechanisms may give rise to
appropriate emergent structures.
The framework presented here is a further development of the
system-theoretic perspective on cognition introduced in Chaotic Logic (Goertzel, 1994) and reiterated in The Hidden
Pattern (Goertzel, 2006), and in
spite of its origins in specific analysis of the Novamente system, is intended
to possess a more general applicability.
In the last couple paragraphs I have explained the historical origins
of the ideas to be presented here: the notions of forward and backward synthesis
were originated as part of an effort to simplify the collection of cognitive
mechanisms utilized in the Novamente system. These notions were then recognized as possessing potentially
more general importance. In the
remainder of the paper I will proceed in the opposite direction: presenting
forward and backward synthesis as general system-theoretic (and mathematical)
notions, and exploring their general implications for the philosophy of
cognition. In another paper
(Goertzel, 2006) these are applied to provide a systematic typology of the
collection of Novamente cognitive processes.
Furthermore, the (hypothesized, not yet observed in experiments)
emergence of self and attention from the overall dynamics of the Novamente
system, which in prior publications has largely been discussed either in very
general conceptual terms or else in terms of the specific interactions between
specific system components, may now be viewed as a particular case of the
general emergence of self and attention as strange attractors of
forward-backward synthesis dynamics.
This is often the sort of conclusion one wants to get out of systems
theory. It rarely directly tells
one specific new things about specific systems -- but it frequently allows one
to better organize and understand specific things about specific systems, thus
in some cases pointing the way to new discoveries.
3. Forward and Backward Synthesis as General Cognitive
Dynamics
The notion of forward and
backward synthesis presented here is an elaboration of a system-theoretic
approach to cognition developed by George Kampis and the author in the early
1990Õs. This section presents
forward and backward synthesis in this general context.
3.1. Component-Systems and Self-Generating Systems
Let us begin with the concept
of a Òcomponent-systemÓ, as described in George KampisÕs (1991) book Self-Modifying
Systems in Biology and Cognitive Science,
and as modified into the concept of a Òself-generating systemÓ or SGS in Chaotic
Logic. Roughly speaking, a Kampis-style component-system
consists of a set of components that combine with each other to form other,
compound components. The metaphor
Kampis uses is that of Lego blocks, combining to form bigger Lego structures. Compound structures may in turn be
combined together to form yet bigger compound structures. A self-generating system is
basically the same concept as a component-system, but understood to be
computable, whereas Kampis claims that component-systems are uncomputable.
Next, in SGS theory there is
also a notion of reduction (not present in the Lego metaphor): sometimes when
components are combined in a certain way, a ÒreactionÓ happens, which may lead
to the elimination of some of the components. One relevant metaphor here is chemistry. Another is abstract algebra: for
instance, if we combine a component f with its ÒinverseÓ component f-1,
both components are eliminated.
Thus, we may think about two stages in the interaction of sets of
components: combination, and reduction.
Reduction may be thought of as algebraic simplification, governed by a
set of rules that apply to a newly created compound component, based on the
components that are assembled within it.
Formally, suppose {C1, C2,...}
is the set of components present in a discrete-time component-system at time
t. Then, the components present at
time t+1 are a subset of the set of components of the form
Reduce( Join (Ci(1),
... ,Ci(r)))
where Join is a joining
operation, and Reduce is a reduction operator. The joining operation is assumed to map tuples of components
into components, and the reduction operator is assumed to map the space of
components into itself. Of course,
the specific nature of a component system is totally dependent on the
particular definitions of the reduction and joining operators; below I will
specify these operators in the context of the Novamente AGI system, but for the
purpose of the general theoretical discussion in this section they may be left
general.
It is also important that
(simple or compound) components may have various quantitative properties. Given appropriate theoretical
understanding, these properties may sometimes be inferred by knowing the
ingredients that went into making up a compound component, and the reductions
that occurred. Or, sometimes,
experiments must be done on the component to calculate its quantitative
properties.
3.2. Forward and Backward Synthesis
Now we move on to the main
point. The basic idea put forth in
this paper is that all or nearly all focused cognitive processes are
expressible using two general process-schemata called forward and backward
synthesis, to be presented below.
The notion of Òfocused cognitive processÓ will be exemplified more
thoroughly below, but in essence what is meant is a cognitive process that
begins with a small number of items (drawn from memory or perception) as its
focus, and has as its goal discovering something about these items, or
discovering something about something else in the context of these items or in
a way strongly biased by these items.
This is different from, for example, a cognitive process whose goal is
more broadly-based and explicitly involves all or a large percentage of the
knowledge in an intelligent systemÕs memory store.
Figure 1. The General Process of Forward
Synthesis
The forward and backward
synthesis processes as I conceive them, in the general framework of SGS theory,
are as follows:
Forward synthesis:
- Begin with some
initial components (the initial Òcurrent poolÓ), an additional set of
components identified as ÒcombinatorsÓ (combination operators), and a goal
function
- Combine the components
in the current pool, utilizing the combinators, to form product components
in various ways, carrying out reductions as appropriate, and calculating
relevant quantities associated with components as needed
- Select the product
components that seem most promising according to the goal function, and
add these to the current pool (or else simply define these as the current
pool)
- Return to Step 2
Figure 2. The General Process of Backward
Synthesis
Backward synthesis:
- Begin with some
components (the initial Òcurrent poolÓ), and a goal function
- Seek components so
that, if one combines them to form product components using the
- combinators and then
performs appropriate reductions, one obtains (as many as possible of) the
components in the current pool
- Use the newly found
constructions of the components in the current pool, to update the
quantitative properties of the components in the current pool, and also (via
the current pool) the quantitative properties of the components in the
initial pool
- Out of the components
found in Step 2, select the ones that seem most promising according to the
goal function, and add these to the current pool (or else simply define
these as the current pool)
- Return to Step 2
Less technically and more
conceptually, one may rephrase these process descriptions as follows:
Forward synthesis: Iteratively build compounds from the initial
component pool using the combinators, greedily seeking compounds that seem
likely to achieve the goal
Backward synthesis: Iteratively search (the systemÕs
long-term memory) for component-sets that combine using the combinators to form
the initial component pool (or subsets thereof), greedily seeking
component-sets that seem likely to achieve the goal
More formally, forward synthesis may
be specified as follows. Let X
denote the set of combinators, and let Y0 denote the initial pool of
components (the initial focus of the cognitive process). Given Yi, let Zi
denote the set
Reduce( Join (Ci(1),
... ,Ci(r)))
where the Ci are
drawn from Yi or from X.
We may then say
Yi+1 = Filter(Zi)
where Filter is a function
that selects a subset of its arguments.
Backward synthesis, on the
other hand, begins with a set W of components, and a set X of combinators, and
tries to find a series Yi so that according to the process of
forward synthesis, Yn=W.
In practice, of course, the
implementation of a forward synthesis process need not involve the explicit
construction of the full set Zi. Rather, the filtering operation takes place implicitly
during hte construction of Yi+1. The result, however, is that one gets some subset of
the compounds producible via joining and reduction from the set of components
present in Yi plus the combinators X.
Conceptually one may view
forward-synthesis as a very generic sort of Ògrowth process,Ó and
backward-chaining as a very generic sort of Òfiguring out how to grow something.Ó The intuitive idea underlying the
present proposal is that these forward-going and backward-going Ògrowth
processesÓ are among the the essential foundations of cognitive control, and
that a conceptually sound design for cognitive control should explicitly make
use of this fact. To abstract away
from the details, what these processes are about is:
1. taking the general dynamic of compound-formation and
reduction as outlined in Kampis and Chaotic Logic
2. introducing goal-directed pruning (ÒfilteringÓ) into this
dynamic so as to account for the limitations of computational resources that
are a necessary part of pragmatic intelligence
3.3. The Dynamic of Iterative Forward-Backward Synthesis
While forward and backward
synthesis are both very useful on their own, they achieve their greatest power
when harnessed together. It is my
hypothesis that the dynamic pattern of alternating forward and backward
synthesis has a fundamental role in cognition. Put simply, forward synthesis creates new mental forms by
combining existing ones. Then,
backward synthesis seeks simple explanations for the forms in the mind,
including the newly created ones; and, this explanation itself then comprises
additional new forms in the mind, to be used as fodder for the next round of
forward synthesis. Or, to put it
yet more simply:
É Combine É Explain É Combine É Explain É
Combine É
It is not hard to express
this alternating dynamic more formally, as well.
Let X denote any set of
components.
Let F(X) denote a set of components
which is the result of forward synthesis on X.
Let B(X) denote a set of
components which is the result of backward synthesis of X. We assume also a heuristic biasing the
synthesis process toward simple constructs.
Let S(t) denote a set of components
at time t, representing part of a systemÕs knowledge base.
Let I(t) denote components
resulting from the external environment at time t.
Then, we may consider a
dynamical iteration of the form
S(t+1) = B( F(S(t) + I(t)) )
This expresses the notion of
alternating forward and backward synthesis formally, as a dynamical iteration
on the space of sets of components.
We may then speak about attractors of this iteration: fixed points,
limit cycles and strange attractors.
One of the key hypotheses I wish to put forward here is that some key
emergent cognitive structures are strange attractors of this equation. The iterative dynamic of combination
and explanation leads to the emergence of certain complex structures that are,
in essence, maintained when one
recombines their parts and then seeks to explain the recombinations. These structures are built in the first
place through iterative recombination and explanation, and then survive in the
mind because they are conserved by this process. They then ongoingly guide the construction and destruction
of various other temporary mental structures that are not so conserved.
4. Self and Focused Attention as Approximate Attractors
of the Dynamic of Iterated Forward/Backward Synthesis
In The Hidden Pattern I have argued that two key aspects of intelligence
are emergent structures that may be called the ÒselfÓ and the Òattentional
focus.Ó[1] These, it is suggested, are aspects of
intelligence that may not effectively be wired into the infrastructure of an
intelligent system, though of course the infrastructure may be configured in
such a way as to encourage their emergence. Rather, these aspects, by their nature, are only likely to
be effective if they emerge from the cooperative activity of various cognitive
processes acting within a broad based of knowledge.
In the previous section I
have described the pattern of ongoing habitual oscillation between forward and
backward synthesis as a kind of Òdynamical iteration.Ó Here I will argue that both self
and attentional focus may be viewed as strange attractors of this
iteration. The mode of argument is
relatively informal. References
will be given into the cognitive science literature, but the essential
processes under consideration are ones that are poorly understood from an
empirical perspective, due to the extreme difficulty involved in studying them
experimentally. For understanding
self and attentional focus, we are stuck in large part with introspection,
which is famously unreliable in some contexts, yet still (I feel) dramatically
better than having no information at all.
Anyhow, the perspective on self and attentional focus given here is a
synthesis of empirical and introspective notions, drawn largely from the published
thinking and research of others but with a few original twists.
4.1. Self
The ÒselfÓ in the present
context refers to the Òphenomenal selfÓ (Metzinger, 2004) or Òself-modelÓ
(Epstein, 1978). That is, the self
is the model that a system builds internally, reflecting the patterns observed
in the (external and internal) world that directly pertain to the system
itself. As is well known in
everyday human life, self-models need not be completely accurate to be useful;
and in the presence of certain psychological factors, a more accurate
self-model may not necessarily be advantageous. But a self-model that is too badly inaccurate will lead to a
badly-functioning system that is unable to effectively act toward the
achievement of its own goals.
The value of a self-model for
any intelligent system carrying out embodied agentive cognition is
obvious. And beyond this, another
primary use of the self is as a foundation for metaphors and analogies in
various domains. Patterns
recognized pertaining the self are analogically extended to other
entities. In some cases this leads
to conceptual pathologies, such as the anthropomorphization of trees, rocks and
other such objects that one sees in some precivilized cultures. But in other cases this kind of analogy
leads to robust sorts of reasoning – for instance, in reading Lakoff and
NunezÕs (2002) intriguing explorations of the cognitive foundations of
mathematics, it is pretty easy to see that most of the metaphors on which they
hypothesize mathematics to be based, are grounded in the mindÕs
conceptualization of itself as a spatiotemporally embedded entity, which in
turn is predicated on the mindÕs having a conceptualization of itself (a self)
in the first place.
A self-model can in many
cases form a self-fulfilling prophecy (to make an obvious
double-entendreÕ!). Actions are generated based on oneÕs
model of what sorts of actions one can and/or should take; and the results of
these actions are then incorporated into oneÕs self-model. If a self-model proves a generally bad
guide to action selection, this may never be discovered, unless said self-model
includes the knowledge that semi-random experimentation is often useful.
In what sense, then, may it
be said that self is an attractor of iterated forward-backward synthesis? Backward synthesis infers the self from
observations of system behavior.
The system asks: What kind of system might I be, in order to give rise
to these behaviors that I observe myself carrying out? Based on asking itself this
question, it constructs a model of itself, i.e. it constructs a self. Then, this self guides the systemÕs
behavior: it builds new logical relationships its self-model and various other
entities, in order to guide its future actions oriented toward achieving its
goals. Based on the behaviors new
induced via this constructive, forward-synthesis activity, the system may then
engage in backward synthesis again and ask: What must I be now, in order to
have carried out these new actions?
And so on.
My hypothesis is that after
repeated iterations of this sort, in infancy, finally during early childhood a
kind of self-reinforcing attractor occurs, and we have a self-model that is
resilient and doesnÕt change dramatically when new instances of action- or
explanation-generation occur.
This is not strictly a mathematical attractor, though, because over a
long period of time the self may well shift significantly. But, for a mature self, many hundreds
of thousands or millions of forward-backward synthesis cycles may occur before
the self-model is dramatically modified.
For relatively long periods of time, small changes within the context of
the existing self may suffice to allow the system to control itself
intelligently.
Finally, it is interesting to
speculate regarding how self may differ in future AI systems as opposed to in humans. The relative stability we see in human
selves may not exist in AI systems that can self-improve and change more
fundamentally and rapidly than humans can. There may be a situation in which, as soon as a system has
understood itself decently, it radically modifies itself and hence violates its
existing self-model. Thus:
intelligence without a long-term stable self. In this case the Òattractor-ishÓ nature of the self holds
only over much shorter time scales than for human minds or human-like minds. But the alternating process of forward
and backward synthesis for self-construction is still critical, even though no
reasonably stable self-constituting attractor ever emerges. The psychology of
such intelligent systems will almost surely be beyond human beings' capacity
for comprehension and empathy.
4.2. Attentional Focus
Next, the notion of an
Òattentional focusÓ is similar to BaarsÕ (1988) notion of a Global Workspace: a
collection of mental entities that are, at a given moment, receiving far more
than the usual share of an intelligent systemÕs computational resources. Due to the amount of attention paid to
items in the attentional focus, at any given moment these items are in large
part driving the cognitive processes going on elsewhere in the mind as well
– because the cognitive processes acting on the items in the attentional
focus are often involved in other mental items, not in attentional focus, as
well (and sometimes this results in pulling these other items into attentional
focus). An intelligent system must
constantly shift its attentional focus from one set of entities to another
based on changes in its environment and based on its own shifting
discoveries.
In the human mind, there is a
self-reinforcing dynamic pertaining to the collection of entities in the
attentional focus at any given point in time, resulting from the observation
that If A is in the attentional focus, and A and B have often been
associated in the past, then odds are increased that B will soon be in the
attentional focus. This basic observation has been refined tremendously
via a large body of cognitive psychology work; and neurologically it follows
not only from HebbÕs (1949) classic work on neural reinforcement learning, but
also from numerous more modern refinements (Sutton and Barto, 1998). But it implies that two items A
and B, if both in the attentional focus, can reinforce each othersÕ presence in
the attentional focus, hence forming a kind of conspiracy to keep each other in
the limelight. But of course, this
kind of dynamic must be counteracted by a pragmatic tendency to remove items
from the attentional focus if giving them attention is not providing sufficient
utility in terms of the achievement of system goals.
The forward and backward
synthesis perspective provides a more systematic perspective on this
self-reinforcing dynamic. Forward
synthesis occurs in the attentional focus when two or more items in the focus
are combined to form new items, new relationships, new ideas. This happens continually, as one of the
main purposes of the attentional focus is combinational. On the other hand, backward synthesis
then occurs when a combination that has been speculatively formed is then
linked in with the remainder of the mind (the ÒunconsciousÓ, the vast body of
knowledge that is not in the attentional focus at the given moment in
time). Backward synthesis
basically checks to see what support the new combination has within the
existing knowledge store of the system.
Thus, forward/backward synthesis basically comes down to Ògenerate and
testÓ, where the testing takes the form of attempting to integrate the
generated structures with the ideas in the unconscious long-term memory. One of the most obvious examples of
this kind of dynamic is creative thinking (Boden, 1994; Goertzel, 1997), where
the attentional focus continually combinationally creates new ideas, which are
then tested via checking which ones can be validated in terms of (built up
from) existing knowledge.
The backward synthesis stage
may result in items being pushed out of the attentional focus, to be replaced
by others. Likewise may the
forward synthesis stage: the combinations may overshadow and then replace the
things combined. However, in human
minds and functional AI minds, the attentional focus will not be a complete
chaos with constant turnover: sometimes the same set of ideas – or a
shifting set of ideas within the same overall family of ideas -- will remain in
focus for a while. When this
occurs it is because this set or family of ideas forms an approximate attractor
for the dynamics of the attentional focus, in particular for the
forward/backward synthesis dynamic of speculative combination and integrative
explanation. Often, for instance,
a small Òcore setÓ of ideas will remain in the attentional focus for a while,
but will not exhaust the attentional focus: the rest of the attentional focus
will then, at any point in time, be occupied with other ideas related to the
ones in the core set. Often
this may mean that, for a while, the whole of the attentional focus will move
around quasi-randomly through a Òstrange attractorÓ consisting of the set of
ideas related to those in the core set.
5. Conclusion
The ideas presented above
(the notions of forward and backward synthesis, and the hypothesis of self and
attentional focus as attractors of the iterative forward-backward synthesis
dynamic) are quite generic and are hypothetically proposed to be applicable to
any cognitive system, natural or artificial. In another paper (Goertzel, 2006), I get more specific and discuss the
manifestation of the above ideas in the context of the Novamente AGI
architecture. I have found that
the forward/backward synthesis approach is a valuable tool for conceptualizing
NovamenteÕs cognitive dynamics.
And, I conjecture that a similar utility may be found more generally.
Next, so as not to end on too
blaseÕ of a note, I will also make a stronger hypothesis. My hypothesis is that, in order for a
physical or software system to achieve intelligence that is roughly human-level
in both capability and generality, using computational resources on the same
order of magnitude as the human brain, this system must
- manifest the dynamic
of iterated forward and backward synthesis
- do so in such a way as
to lead to self and attentional focus as emergent structures that serve as
approximate attractors of this dynamic, over time periods that are long
relative to the basic Òcognitive cycle timeÓ of the systemÕs
forward/backward synthesis dynamics
To prove the truth of a
hypothesis of this nature would seem to require mathematics fairly far beyond
anything that currently exists.
Nonetheless, however, I feel it is important to formulate and discuss
such hypotheses, so as to point the way for future investigations both
theoretical and pragmatic.
References
á
Baars, Bernard J. (1988). A Cognitive Theory of Consciousness. New York: Cambridge
University Press
á
Boden, Margeret (1994). The
Creative Mind. Routledge
á
Epstein, Seymour (1980). The Self-Concept: A Review and the Proposal of
an Integrated Theory of Personality, p. 27-39 in Personality: Basic Issues
and Current Research, Englewood Cliffs: Prentice-Hall
á
Goertzel (2006). Virtual
Easter Egg Hunting: A Thought-Experiment in Embodied Social Learning, Cognitive Process Integration, and
the Dynamic Emergence of the Self , Proceedings of 2006 AGI Workshop, Bethesda MD,
IOS Press
á
Goertzel, Ben (1993). The
Evolving Mind. Gordon and Breach
á
Goertzel, Ben (1993). The
Structure of Intelligence. Springer-Verlag
á
Goertzel, Ben (1997). From
Complexity to Creativity. Plenum Press
á
Goertzel, Ben (1997). Chaotic
Logic. Plenum Press
á
Goertzel, Ben (2002). Creating
Internet Intelligence. Plenum Press
á
Goertzel, Ben and Cassio Pennachin, Editors (2006). Artificial General Intelligence. Springer-Verlag.
á
Goertzel, Ben and Stephan Vladimir Bugaj (2006). The Path to Posthumanity. Academica Press
á
Goertzel, Ben (2006). The
Hidden Pattern: A Patternist Philosophy of Mind, Brown-Walker Press, to
appear
á
Hebb, Donald (1949). The
Organization of Behavior. Wiley
á
Kampis, George (1991). Component-Systems
in Biology and Cognitive Science.
Plenum Press
á
Lakoff, George and Rafael Nunez (2002). Where Mathematics Comes From, Basic Books
á
Lenat, D. and R. V. Guha. (1990). Building Large Knowledge-Based
Systems: Representation and Inference in the Cyc Project. Addison-Wesley
á
Metzinger, Thomas (2004). Being
No One. MIT Press
á
Ramamurthy, U., Baars, B., D'Mello, S. K., & Franklin, S. (2006).
LIDA: A Working Model of Cognition. Proceedings of the 7th International
Conference on Cognitive Modeling. Eds: Danilo Fum, Fabio Del Missier and Andrea
Stocco; pp 244-249. Edizioni Goliardiche, Trieste, Italy.
á
Sutton, Richard and Andrew Barto (1998). Reinforcement Learning.
MIT Press.
á
Wang, Pei (2006). Rigid
Flexibility: The Logic of Intelligence.
Springer-Verlag.