MORCOM@IS4SI 2019 Berkeley - ABSTRACTS
Morphological, Natural, Analog and Other Unconventional Forms of Computing for Cognition and Intelligence
The questions regarding the modern information processing technology in performing tasks traditionally considered as exclusively human, or even considered as defining human being such as thinking, intelligence, consciousness, or goal-oriented agency are essentially the same questions as those asked by natural philosophers through the ages. We have now more powerful intellectual and technological tools in searching for answers, but the existing pervasive and convenient tool-kit brings also a danger of following the old habits of thinking.
This is why it is necessary to re-consider and re-examine even most fundamental concepts, such as computing or cognition and intelligence. There are examples of novel studies, for instance of morphological computing and embodied cognition, that succeed in escaping the inertia of thinking habits and question conventional theoretical and practical models.
See the event from the last year: https://sites.google.com/view/morphologicalcomputing
In this event we bring together perspectives on morphological-, physical-, natural-, analog- and embodied cognitive computation and other forms of unconventional conceptualization of computing, cognition and intelligence. We encourage open and constructive debate on the perceived differences in the various perspectives on constructivist and computationalist accounts of the dynamics of information in its natural and artifactual realizations. Abstracts of contributions should be send to organizers by April 15, 2019 (addresses below).
Gordana Dodig-Crnkovic, Professor, Chalmers University of Technology & Mälardalen University, Sweden, firstname.lastname@example.org
Marcin J. Schroeder, Ph.D., Professor
& Dean of Academic Affairs, Akita
International University (国際教養大学), Akita,
Japan. Editor-in-Chief, Philosophies
SESSION A – MORCOM (Tuesday, 2019 06 04)
Gordana Dodig-Crnkovic, Chalmers University of Technology,
Title: “Morphological, Natural, Analog and Other Unconventional Forms of Computing for Cognition and Intelligence”
Abstract. What is the relationship between Cognition and Intelligence? How does cognitive computing relate to AI? What is the difference between Natural, Analog and Morphological computing? At the moment there is a huge variety of use of those terms that causes confusion. In this presentation I would present the taxonomy of computing originally developed in collaboration with Mark Burgin, and extended by some recent work on cognition as information processing.
Dodig Crnkovic, G. (2018) Cognition as Embodied Morphological Computation. Philosophy and Theory of Artificial Intelligence 2017: 19-23. http://dx.doi.org/10.1007/978-3-319-96448-5
Dodig-Crnkovic, G., Nature as a Network of Morphological Infocomputational Processes for Cognitive Agents, The European Physical Journal Special Topics, DOI: 10.1140/epjst/e2016-60362-9 Eur. Phys. J. 2017, 226, 181–195.
Mark Burgin, University of California, Los Angeles, email@example.com
Title: “Processing Information by Symmetric Inductive Machines”
Abstract. To reflect important properties of computers, Marcin Schroeder introduced a new model of computation - symmetric Turing machines or S-machines (Schroeder, 2013; 2013a). In a conventional Turing machine, the head (processor) performs operations with data in the memory (tape) using a fixed system of instructions – its program. In a symmetric Turing machine, information processing goes not only from the head to the memory but also backward. On the one hand, the head (processor) performs operations with data in the memory using a fixed system of instructions – its program. On the other hand, the memory performs operations with instructions from the head (processor).
Physical computers also perform operations with their programs using special tools such as interpreters, compilers and translators. There are also program optimizers, which improve characteristic of programs.
Automata that perform transformations with their programs, such as reflexive Turing machines, were explored in (Burgin, 1992). It was proved that these machines have the same computing power as Turing machines but could be much more efficient.
Using technique similar to the one employed in (Burgin, 1992), it is possible to prove that functioning of a symmetric Turing machine can be simulated by a conventional Turing machine with two tapes and two heads. It means that symmetric Turing machines have the same computing power as Turing machines. At the same time it is also possible to prove that symmetric Turing machines can be much more efficient than Turing machines.
To achieve higher computing power, here we introduce and study inductive symmetric machines, which further develop the structure and possibilities of inductive Turing machines allowing to model natural computations in various situations.
Burgin, M. Reflexive Calculi and Logic of Expert Systems, in Creative processes modeling by means of knowledge bases, Sofia, 1992, pp. 139-160 (in Russian)
Schroeder,M.J. Dualism of Selective and Structural Manifestations of Information in Modelling of Information Dynamics, In: G. Dodig-Crnkovic, R. Giovagnoli (Eds.) Computing Nature, SAPERE 7, Springer, Berlin, Germany, pp. 125-137, (2013)
Schroeder,M.J. From Proactive to Interactive Theory of Computation, M. Bishop and Y. J. Erden (Eds.): The 6th AISB Symposium on Computing and Philosophy: The Scandal of Computation – What is Computation? pp.47 - 51, 2013a
SESSION E – MORCOM (Tuesday)
Ricardo Q. Figueroa*, Genaro J. Martinez Andrew Adamatzky, Luz N. Oliva-Moreno
1 Artiﬁcial Life Robotics Laboratory, Escuela Superior de C´omputo, Instituto Polit´ecnico Nacional, M´exico.
2 Unconventional Computing Lab, University of the West of England, Bristol, United Kingdom.
3 Unidad Profesional Interdisciplinaria de Ingenier´ıa Hidalgo, Instituto Polit´ecnico Nacional, M´exico
Title: “Robots Simulating Turing Machines”
Abstract. We will discuss how modular robots can be operated for simulate computers. This research explores a classic problem in computer science: machines simulating machines. The main goal of this research consist in proof how a set of Cubelets robots may be organized to implement a Turing machine. Additionally, the Turing machine constructed with Cubelets robots shows how a universal Turing machine works in this device. This robotic Turing machine moves its head to read a ﬁxed type containing a binary string and this one changes in each iteration. The machine is complemented with some LEGO pieces to stabilize the concatenation of these robots. A characteristic of this machine is that we can change its conﬁguration for to get another kind of robot, probably useful for the same goal or to another propose.
Andrew Adamatzky (Ed.) (2017) Advances in Unconventional Computing (Volume Prototypes, Models and Algorithms), Springer International Publishing.
Ricardo Q. Figueroa, Daniel A. Zamorano, Genaro J. Mart´ınez, Andrew Adamatzky (2019) A Turing Machine Constructed with Cubelets Robots, Journal of Robotics, Networking and Artiﬁcial Life 5(4) 1–4.
John von Neumann (1966) Theory of Self-reproducing Automata (edited and completed by A. W. Burks), University of Illinois Press, Urbana and London.
Rao Mikkilineni, Golden Gate University, firstname.lastname@example.org
Title: “Structural Machines as Unconventional Knowledge Processors”
Abstract. Knowledge systems often have very sophisticated structures. For instance, representation of knowledge in the form of a text involves structures of this text. Their structure is represented by hypertexts, which are networks (sometimes very complex ones) consisting of linguistic objects, such as words, phrases and sentences, with diverse links connecting these objects. Coming to multimedia, we encounter even more multifarious structures. At the same time, computational machines and automata are mostly oriented at sequential processing of information. For instance, Turing machines process words letter by letter. Thus, to work with knowledge using Turing machines, it is necessary in advance to present knowledge by linear structures. To improve efficiency and allow processing not only symbols but also links between them, more advanced automata, such as Kolmogorov algorithms storage modification machines and relational machines. However, all these relations define only structures of the first order while knowledge structure can have much higher orders. To eliminate this restriction and further advance efficiency, structural machines were introduced. Here, we present knowledge processing by structural machines. Knowledge contains information as matter contains energy and structural machines work with knowledge structures of arbitrary order transforming not only elements of these structures or the content of these elements, as conventional models of computation do, but also relations of different orders in the processed structures. This allows achieving higher flexibility and efficiency in comparison with regular models of computation including both conventional and unconventional computing systems. Structural machines can also simulate such advanced computational models as Kolmogorov algorithms, limit Turing machines, storage modification machines, relational machines and other models of computation. Being structurally universal abstract automata, structural machines work directly with knowledge structures, molecular and atomic structures, with structures studied and utilized in the topological quantum field theory (TQFT) and with structures of quantum information such as qubits.
SESSION I – MORCOM (Tuesday)
Genaro J. Martinez * 1,2, Andrew Adamatzky2, Ricardo Q. Figueroa1, and Dmitr A. Zaitsev3,
1 Computer Science Laboratory, Escuela Superior de C´omputo, Instituto Polit´ecnico Nacional, Mexico.
2 Unconventional Computing Lab, University of the West of England, Bristol, United Kingdom.
3 International Humanitarian University, Odessa, Ukraine
Title: ”Propagation of patterns in non-linear media as a paradigm of unconventional computers”
Abstract. Cellular automata are classic models to design unconventional computing in several ways. Historically, a lot of diﬀerent proposals work handling signals or composition of them interpreted as particles (gliders, mobile-self localizations). Patterns, originating from diﬀerent sources of perturbations, propagating in a precipitating chemical, physical or biological medium do usually compete for the space. They sub-divide the medium onto the regions unique for an initial conﬁguration of disturbances. This sub-division can be expressed in terms of computation. We adopt an analogy between precipitating chemical, physical or biological media and semi-totalistic binary two-dimensional cellular automata. We demonstrate how to implement basic logic and arithmetical operations (its computability) by patterns propagating in geometrically constrained cellular automata medium. Non-serial logic gates are designed and implemented to look a possible circuit. Finally, we show practical implementations of these theoretical designs across of Cubelets robots. In this case, a concatenations of Cubelets robots represent a channel of communication and package of electrons propagates as light and they represent binary signals, the junction of these wires open a new wire conformed with other Cubelets robots which display the result.
Adamatzky, A. (2009) Hot ice computer, Physics Letters A 374(2) 264– 271.
Fischer, T., Kewenig, M., Bozhko, D.A., Serga, A.A., Syvorotka, I.I., Ciubotaru, F., Adelmann, C., Hillebrands, B. & Chumak, A.V. (2017) Experimental prototype of a spin-wave majority gate, American Institute of Physics 17(2) 86–91.
Gregory, L.S., Orlov, A.O., Amlani, I., Bernstein, G.H., Lent, C.S., Merz, J.L., & Porod, W. (1999) Quantum-Dot Cellular Automata: Line and Majority Logic Gate, Japanese Journal of Applied Physics 38 7227–7229.
Mart´ınez, G.J., Adamatzky, A., & Costello, B.L. (2008) On logical gates in precipitating medium: cellular automaton model, Physics Letters A 1(48) 1–5.
Mart´ınez, G.J., Adamatzky, A., Morita, K., & Margenstern, M. (2010) Computation with competing patterns in Life-like automaton, In: Game of Life Automata, A. Adamatzky (ed.), Springer, chapter 27, pages 547–572.
Mitchell, M. (2001) Life and evolution in computers, History and Philosophy of the Life Sciences 23 361–383.
Mart´ınez, G.J., Morita, K., Adamatzky, A., & Margenstern, M. (2010) Majority adder implementation by competing patterns in Life-like rule B2/S2345, Lecture Notes in Computer Science 6079 93–104.
Mart´ınez, G.J., Seck, J.C.T.M., & Zenil, H. (2013) Computation and Universality: Class IV versus Class III Cellular Automata, Journal of Cellular Automata 7(5-6) 393–430.
Toﬀoli, T. (1998) Non-Conventional Computers, Encyclopedia of Electrical and Electronics Engineering (John Webster Ed.), 14:455–471, Wiley & Sons.
Lorenzo Magnani, University of Pavia, Italy,
Title: “Disseminated Computation, Cognitive Domestication of New Ignorant Substrates, and Overcomputationalization”
Abstract. What I called ”eco-cognitive computationalism” considers computation in context, following some of the main tenets advanced by the recent cognitive science views on embodied, situated, and distributed cognition. It is in the framework of this eco-cognitive perspective that we can usefully analyze the recent attention in computer science devoted to the importance of cognitive domestication of new substrates, such as in the case of morphological computation: this new perspective shows how the computational domestication of ignorant substrates can originate new unconventional cognitive embodiments, which expand the processes of computationalization already occurring in our societies. I will also introduce and discuss the concept of overcomputationalism, as intertwined with the traditional concepts of pancognitivism, paniformationalism, and pancomputationalism, seeing them in a more naturalized intellectual disposition, appropriate to the aim of bypass ontological or metaphysical overstatements. What I call overcomputationalization refers to the presence of too many entities and artifacts that carry computational tasks and powers. Overcomputationalization 1) often promotes a plenty of possible unresolvable disorganizational consequences, and 2) tends to favor philosophical reflections that depict an oversimplified vision of the world. Moreover, it tends to generate too many cognitive constraints and limitations, which lead to a weakening of human creative (abductive) cognitive activities, as I have illustrated in the last chapter of my recent book The Abductive Structure of Scientific Creativity (2017), and, because of the excess of redundant cognitive/informational features attributed to entities (features often exogenous to the original functions of them) it tends to prevent human intellectual freedom to benefit from that cognitive simplification that is characteristic of the absence of informational overloads.
SESSION M – MORCOM (Tuesday)
Marcin Schroeder, Akita International University,
Title: “Intelligent Computing: Oxymoron?
Abstract. The question about conditions qualifying an object, whether artificial or natural, an entity or its functioning as intelligent is more about qualifier (intelligence) than qualified. Of course, if we had an established definition of intelligence, then we could sort objects, actions, processes accordingly. But we don’t. Turing attempted to escape the problem in the context of artificial systems (“machine”) using his “imitation game”, but instead of closing the discussion of “intelligent machinery”, he opened Pandora’s box of ever-lasting disputes in which intelligence is frequently mixed with the ability to think, capacity of being conscious, etc. It does not mean that the problem is restricted to intelligent artefacts, since the qualification of human beings as intelligent has been never clarified and typically discussions are about many different “intelligences”, even if in everyday practice people refer without hesitation to the so called “intelligence quotient” (IQ). So whether computers or computing can be intelligent or not depends on the way we understand intelligence. Of course, this freedom of the choice of definition has some limits coming from already established tradition of the use of the term “intelligence”, in particular in the common sense discourse. From this perspective, we cannot ignore the most frequent objection to the intelligence of artefacts based on the doubt that they can have the capacity of symbolic association between the sign and its denotation. Thus, they do not have the capacity to “understand” symbols. The objection is not surprising and its source is not new. We can trace it to Brentano’s claim that intention (associating denotation with signs) is a specific mental capacity of the mind which is absent on the body side of the mind-body division. Since starting from the mind-body division would place us in the maze of centuries old discussions leading nowhere, it is better to rephrase this objection to be more suitable for the reflection on computers and computing. Everyone agrees that computing based on the model of the Turing machine is a transformation of compound symbols (built of digits, Tukey’s bits, letters, or any other finite number of elementary units from which symbols are built) through the manipulation of their components. The components (digits) have their meaning for the machine determined by instructions. It was originally expressed by Turing that machine (we would say “head”) can see or scan the present square for the unitary sign - digit and can act according to the present instruction (state of its “mind”). The action of the machine or head can be understood as an expression of the meaning of the unitary sign for the machine or its head. Obviously, machine’s head does not have any representation of the tape with its configuration of digits, nor even any sensor to monitor more than one square (or in some variations of Turing machines a fixed, finite number of squares). Thus, machine cannot understand the entire compound symbol consisting of possible long configuration of digits. Someone could object that the tape is a part of the machine, so machine has a representation of the configuration on the tape in the form of the tape. This however kills the concept of the symbolic representation by restricting symbolic representation to the strict identity. Human beings definitely are capable of symbolic representation beyond the meaning understood as the identity of the sign and denotation. Does it mean that the Turing machine type computers are doomed to be non-intelligent? No. Someone can claim that the meaning is emergent in both human symbolic thinking and in computing. The difference between actual typical implementations of computers using Tukey’s bits (0 and 1) and typical human brains with possibly large, but finite set of elementary units may be misleading. Nothing prevents us from using implementation of the Turing machine with a very large set of digits. The view that the meaning is emergent eliminates the distinction between semantics and syntactic may seem exotic, but not without precedence. After all, it is at the foundation of constructivism. There is nothing fallacious in the assumption that we live in reality which our mind constructs from a finite number of elementary units (“digits”). Maybe, we actually understand only the finite number of simple components and we react to the instructions telling our brain how to construct complex and diverse reality. But, if this is the case, how do we know that the reality is complex and diverse, if we can understand only simple components? How can we direct our actions to eliminate complexity and diversity? In case of computers (or Turing machines) we know that Turing machine cannot reduce the algorithmic complexity of the configuration on the tape, or it cannot even assess computability of the input configuration. Computing is a one-way process of construction, but not deconstruction. It is a human programmer who decomposes in the process of programming the complex task into an algorithm (intelligent part of the task) and leaves the non-intelligent task of performing the construction of the desired outcome. This is not far from the objection to the intelligence of computers coming from the common sense discussions. When we compare the intelligence of different people, we consider as more intelligent the individuals who have the ability to reduce complexity, usually by making complex tasks simple through the deconstruction and leaving these simple tasks to less intelligent collaborators. Thus, the answer given by the author in this part of the paper is: Yes, oxymoron. But the question remains about what type of capacities have to be added to Turing machine to make it intelligent at the human level or beyond. Partial answer to this question was given by the author in his earlier publications. Computers have to be equipped with the ability to integrate information. The remaining part of the paper is about this and other conditions for making computers intelligent and about the perspectives of implementation of such conditions.
Rao Mikkilineni1, Mark Burgin2 & Eugene Eberbach3
1Golden Gate University, email@example.com
3 Scopium AI, Toronto, Canada & Seekonk, firstname.lastname@example.org
Title: “Cloud Computing as a Step to a Higher-Order AI”
Abstract. Cloud computing approach addresses how to make the right resources available to the right computation to improve scaling, resiliency and efficiency of the computation. In this paper we argue that cloud computing indeed, is a new paradigm for computation upgrading it to a higher order of artificial intelligence, and we put forward cloud automata as a new model for computation. A high-level artificial intelligence requires infusing features of the human brain into AI systems. One of the central features is that the brain learns all the time and learning is incremental. Consequently, for AI, we need to use computational models, which reflect incremental learning without stopping (sentience). These features are inherent in reflexive Turing machines, inductive Turing machines and limit Turing machines.
It is possible to distinguish several paradigms of computation, including Mainframe, PC, Network, Internet, Distributed, Grid and Cloud Computing. New computing paradigms may involve various technologies besides VLSI, such as quantum computing, biologically inspired computing, nanocomputing, optical computing, neurocomputing. Theoretical models of computing are naturally divided into three classes: sub-recursive, recursive and super-recursive algorithms and automata. To construct cloud automata, we use the mathematical theory of Oracles, which include Oracles of Turing machines as its special case. This allows developing a hierarchical approach to artificial intelligence based on Oracles with different ranks. The developed approach includes Oracle AI as a special case providing new tools for exploration of artificial intelligence in general and Oracle artificial intelligence. Discussing named-set approach and Oracle-based evolution of computations, we describe an implementation of a high-performance edge cloud using hierarchical name-oriented networking and Oracle-based orchestration. We demonstrate how cloud automata provide means for improving resiliency, scalability and efficiency of computations. A control overlay allows microservice network provisioning, monitoring and reconfiguration to address fluctuations in their behavior.
SESSION D – MORCOM (Wednesday, 2019 06 05) Skype presentations
Zenil, Algorithmic Dynamics Lab, Unit of
Computational Medicine, Center for Molecular
Medicine, Karolinska Institutet, Stockholm,
Abstract. In this talk I will explain how current approaches of machine, and deep learning based on traditional statistics and information theory fail to capture fundamental properties of our world and are ill-equipped to deal with high-level functions such as inference, abstraction, and understanding, they are fragile and can easily be deceived. In contrast, we will explore recent attempts to combine symbolic and differentiable computation in a form of unconventional hybrid computation that is more powerful and may eventually display and grasp these higher level elements of human intelligence. In particular, I will introduce the field of Algorithmic Information Dynamics and that of Algorithmic Machine Intelligence based on the theories of computability and algorithmic probability, and how these approaches promise to shed light on the weaknesses of current AI and how to attempt to circumvent some of their limitations.
Vincent C. Muller, Technical University of Eindhoven (TU/e) The Netherlands, University of Leeds and Alan Turing Institute, London, UK. www.sophia.de
Title: ”Morphological Computation and the Discussion About Whether Computation involves Meaningful Symbols”
Abstract. The discussion about morphological computation and about whether computation involves meaningful symbols, rather than merely syntactic operations, have been going on for some time now. My rather conservative position has been to say that computation is essentially syntactic algorithmic processing (as the Church-Turing thesis suggested) done by humans with machines. But there are other very fruitful and plausible notions of computing. The question is what kind of question is this? Do we expect a discovery to find out the truth, or can we slice the world in several plausible ways? Is this the same question as realism and anti-realism about computing?