Abstract
As well as having the ability to formulate models of the world capable of experimental falsification, it is evident that human cognitive capability embraces some degree of representational plasticity, having the scope (at least in infancy) to modify the primitives in terms of which the world is delineated. We hence employ the term ‘cognitive bootstrapping’ to refer to the autonomous updating of an embodied agent’s perceptual framework in response to the perceived requirements of the environment in such a way as to retain the ability to refine the environment model in a consistent fashion across perceptual changes.
We will thus argue that the concept of cognitive bootstrapping is epistemically ill-founded unless there exists an a priori percept/motor interrelation capable of maintaining an empirical distinction between the various possibilities of perceptual categorization and the inherent uncertainties of environment modeling.
As an instantiation of this idea, we shall specify a very general, logically-inductive model of perception-action learning capable of compact re-parameterization of the percept space. In consequence of the a priori percept/action coupling, the novel perceptual state transitions so generated always exist in bijective correlation with a set of novel action states, giving rise to the required empirical validation criterion for perceptual inferences. Environmental description is correspondingly accomplished in terms of progressively higher-level affordance conjectures which are likewise validated by exploratory action.
Application of this mechanism within simulated perception-action environments indicates that, as well as significantly reducing the size and specificity of the a priori perceptual parameter-space, the method can significantly reduce the number of iterations required for accurate convergence of the world-model. It does so by virtue of the active learning characteristics implicit in the notion of cognitive bootstrapping.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
{P′ exp} is a subset of {P exp} since, in general, not all of the exploratory actions will result in percepts that could be regarded as intentional; for instance, if an action end-point is unstable.
- 2.
- 3.
Indeed, for a non-interacting cognitive agent, there are effectively no meaningful autonomously-derived criteria that can be applied to perceptual inferences in consequence of the fact that no non-arbitrary cost function capable of distinguishing between rival perceptual hypotheses exists a priori (cf Quine [37] on the ontological underdetermination problem).
- 4.
- 5.
Though, importantly, without the same guarantee of uniform sampling of the legitimate actions (see Conclusions).
- 6.
Indeed, in certain phenomenological models [45], this differential in actual and potential action capabilities constitutes the agent’s environment.
- 7.
Note that this supporting surface is different for each object, since differing objects slot into differing holes, with the holes acting as support-entities otherwise.
- 8.
In a more realistically complex cognitive environment it would be possible to eliminate the a priori awareness of the morphological correspondence between holes and shapes by resolving it into a suitable composite of the two perceptual categories of positional occupancy and spatial adjacency. In fact, these are more like the true Kantian a priori perceptual categories. (A priori in the sense of their being empirically undiscoverable, being rather the conditions of empirical discovery).
- 9.
Note that the predicates of the form not(…) do not generate output variables that can be meaningfully employed to address the percept space, and are hence indicated by dotted lines in figure 1 to denote their removal from the remapping process.
- 10.
The only permitted exception to this rule being predicates with mode declarations that specify only input variables (ie, those having the form modeb(1, p([ + variable type]))), such as hole{ _}shape{ _}match(A, B). In these cases, the predicate can be regarded as a variable-less terminating node of the directed acyclic clause structure.
- 11.
It is always possible to map percepts to smaller subsets such that the ability to discriminate ‘difficult areas’ is lost, and falsifying action possibilities are no longer perceived.
- 12.
However, this paper does not necessarily represent the opinion of the European Community, and the European Community is not responsible for any use which may be made of its contents.
References
Michael L. Anderson. Embodied cognition: A field guide. Artif. Intell., 149(1):91–130, 2003.
R. Bowden, L. Ellis, J. Kittler, M. Shevchenko, and D. Windridge. Unsupervised symbol grounding and cognitive bootstrapping in cognitive vision. In Proc. 13th Int. Conference on Image Analysis and Processing, pages 27–36, September 2005.
David A. Cohn, Zoubin Ghahramani, and Michael I. Jordan. Active learning with statistical models. J. Artif. Intell. Res. (JAIR), 4:129–145, 1996.
Cottingham, Stoothoff, Murdoch, and Kenney. The Philosophical Writings of Descartes. Cambridge University Press, Cambridge, UK, 1991.
Sanjoy Dasgupta, Adam Tauman Kalai, and Claire Monteleoni. Analysis of perceptron-based active learning. In Peter Auer and Ron Meir, editors, COLT, volume 3559 of Lecture Notes in Computer Science, pages 249–263. Springer, 2005.
J. Dewey. The reflex arc concept in psychology. The Psychological Review, (3):356–370, 1896.
Thomas G. Dietterich. Hierarchical reinforcement learning with the maxq value function decomposition. J. Artif. Intell. Res. (JAIR), 13:227–303, 2000.
J.A. Fodor. The Modularity of Mind. MIT Press, Cambridge, MA, 1983.
S. Gallagher. How the Body Shapes the Mind. Oxford University Press, Oxford, 2005.
Ph. Gaussier. Towards a cognitive system algebra: A perception/action perspective. In European Workshop on Learning Robots, pages 88–100, Prague, Sept 2001.
A. Geir. Explanation and Understanding: The Hermeneutic Arc. PhD thesis, University of Oslo Department of Philosophy, 2001.
R. W. Jr Gibbs. Embodiment and Cognitive Science. Cambridge University Press, 2005. ISBN-10: 0521811740 — ISBN-13: 9780521811743.
J. J. Gibson. The ecological approach to visual perception. Houghton-Mifflin, Boston, 1979.
G. Granlund. Organization of architectures for cognitive vision systems. In Proceedings of Workshop on Cognitive Vision, Schloss Dagstuhl, Germany, 2003.
G. H. Granlund. An associative perception-action structure using a localized space variant information representation. In Proceedings of Algebraic Frames for the Perception-Action Cycle (AFPAC), Kiel, Germany, September 2000.
G. H. Granlund and A. Moe. Unrestricted recognition of 3d objects for robotics using multilevel triplet invariants. AI Magazine, 25(2):51–67, 2004.
S. Harnad. The symbol grounding problem. Physica D, (42):335–346, 1990.
M. Heidegger. Being and Time. Blackwell, 1996.
Scott B. Huffman and John E. Laird. Flexibly instructable agents. Journal of Artificial Intelligence Research, 3:271–324, 1995.
E. Husserl. Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy – First Book: General Introduction to a Pure Phenomenology. The Hague, 1982.
“Leslie Pack Kaelbling, Michael L. Littman, and Andrew P. Moore”. Reinforcement learning: A survey. J. Artif. Intell. Res. (JAIR), 4:237–285, 1996.
Immanuel Kant. Critique of Pure Reason. Cambridge University Press, 1999. ISBN: 0521657296.
S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi. Optimization by simulated annealing. Science, Number 4598, 13 May 1983, 220, 4598:671–680, 1983.
G. Lakoff and M Johnson. Philosophy in the Flesh : The Embodied Mind and Its Challenge to Western Thought. Harper Collins Publishers, 1999.
G. Lakoff and R. Núñez. Where Mathematics Comes From: How the Embodied Mind Brings Mathematics into Being. Basic Books, 2000.
D. Magee, C. J. Needham, P. Santos, A. G. Cohn, and D. C. Hogg. Autonomous learning for a cognitive agent using continuous models and inductive logic programming from audio-visual input. In Proc. of the AAAI Workshop on Anchoring Symbols to Sensor Data, 2004.
J. McCarthy and P.J. Hayes. Some philosophical problems from the standpoint of artificial intelligence. Machine Intelligence, (4):463–502, 1969.
J. McGrenere and W. Ho. Affordances: Clarifying and evolving a concept. In Proceedings of Graphics Interface 2000, pages 179–186, Montreal, Canada, 2000.
M. Merleau-Ponty. Phenomenology of Perception. New York: Humanities Press, 1962.
R. G. Millikan. Language, Thought, and Other Biological Categories: New Foundations for Realism. The MIT Press; Reprint edition, December 1987. ISBN: 0262631156.
R. G. Millikan. White Queen Psychology and Other Essays for Alice. The MIT Press; Reprint edition, March 1995. ISBN: 0262631628.
Andrew Moore. Variable resolution dynamic programming: Efficiently learning action maps in multivariate real-valued state-spaces. In L. Birnbaum and G. Collins, editors, Machine Learning: Proceedings of the Eighth International Conference, 340 Pine Street, 6th Fl., San Francisco, CA 94104, June 1991. Morgan Kaufmann.
S. Muggleton. Inductive Logic Programming. Academic Press, 1992.
Stephen. Muggleton. Inverse entailment and progol. New Gen. Comput., 13:245–286, 1995.
A. Newell and H. Simon. The theory of human problem solving; reprinted in collins & smith (eds.). In Readings in Cognitive Science, section 1.3., 1976.
K. Popper. The Logic of Scientific Discovery. (translation of Logik der Forschung). Hutchinson, London, 1959.
W. V. O. Quine. Ontological Relativity. Columbia, 1977.
P. Ricoeur. Hermeneutics and the Human Sciences: Essays on Language, Action and Interpretation. Cambridge: Cambridge University Press, 1981.
H. S. Seung, Manfred Opper, and Haim Sompolinsky. Query by committee. In Computational Learning Theory, pages 287–294, 1992.
G. Sommer, E. Bayro-Corrochano, and Th. Blow. Geometric algebra as a framework for the perception-action cycle. In Workshop on Theoretical Foundation of Computer Vision, Ed. F. Solina, Wien, 1997. Springer Verlag.
L. Steels. The origins of syntax in visually grounded robotic agents. In M. Pollack, editor, Proceedings of the 10th IJCAI, Nagoya, pages 1632–1641. AAAI Press, Menlo-Park Ca., 1997.
L. Steels and F. Kaplan. Bootstrapping grounded word semantics. In Ted Briscoe, editor, Linguistic Evolution through Language Acquisition: Formal and Computational Models, chapter 3. Cambridge University Press, 2002.
R. Sun. Desiderata for cognitive architectures. Philosophical Psychology, 17(3), 2004.
R. Sun, T. Peterson, and E. Merrill. A hybrid architecture for situated learning of reactive sequential decision making. Applied Intelligence, 11:109–127, 1999.
Bas. van Fraassen. The Scientific Image. Oxford: Oxford University Press, 1980. ISBN 0-19-824427-4.
D Windridge and J Kittler. Cognitive bootstrapping: A survey. Technical report, CVSSP, University of Surrey, UK, 2005.
D Windridge and J Kittler. Epistemic constraints on autonomous symbolic representation in natural and artificial agents. Applications of Computational Intelligence in Biology: Current Trends and Open Problems, Studies in Computational Intelligence (SCI), 122, 2008.
L. Wittgenstein. Philosophical investigations : the German text with a revised English translation by Ludwig Wittgenstein. Oxford : Blackwell, 2001. ISBN 0631231277.
J G Wolff. Cognitive development as optimisation. In L Bolc, editor, Computational Models of Learning, pages 161–205, Heidelberg, 1987. Springer-Verlag.
Acknowledgments
The work presented here was supported by the the European Union, grant DIPLECS (FP 7 ICT project no. 215078)Footnote 12. We also gratefully acknowledge funding from the UK Engineering and Physical Sciences Research Council (EPSRC) (grant EP/F069421/1)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer Science+Business Media, LLC
About this paper
Cite this paper
Windridge, D., Kittler, J. (2010). Perception-Action Learning as an Epistemologically-Consistent Model for Self-Updating Cognitive Representation. In: Hussain, A., Aleksander, I., Smith, L., Barros, A., Chrisley, R., Cutsuridis, V. (eds) Brain Inspired Cognitive Systems 2008. Advances in Experimental Medicine and Biology, vol 657. Springer, New York, NY. https://doi.org/10.1007/978-0-387-79100-5_6
Download citation
DOI: https://doi.org/10.1007/978-0-387-79100-5_6
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-79099-2
Online ISBN: 978-0-387-79100-5
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)