Chapter 9. Spike-Based Symbolic Computations on Bit Strings and Numbers

The brain uses recurrent spiking neural networks for higher cognitive functions such as symbolic computations, in particular, mathematical computations. We review the current state of research on spike-based symbolic computations of this type. In addition, we present new results which show that surprisingly small spiking neural networks can perform symbolic computations on bit sequences and numbers and even learn such computations using a biologically plausible learning rule. The resulting networks operate in a rather low firing rate regime, where they could not simply emulate artificial neural networks by encoding continuous values through firing rates. Thus, we propose here a new paradigm for symbolic computation in neural networks that provides concrete hypotheses about the organization of symbolic computations in the brain. The employed spike-based network models are the basis for drastically more energy-efficient computer hardware – neuromorphic hardware. Hence, our results can be seen as creating a bridge from symbolic artificial intelligence to energy-efficient implementation in spike-based neuromorphic hardware.

Infusing Autopoietic and Cognitive Behaviors into Digital Automata to Improve their Sentience, Resilience and Intelligence

The holy grail of Artificial Intelligence (AI) has been to mimic human intelligence using computing machines. Autopoiesis which refers to a system with well-defined identity and is capable of re-producing and maintaining itself and cognition which is the ability to process information, apply knowledge, and change the circumstance are associated with resilience and intelligence. While classical computer science (CCS) with symbolic and sub-symbolic computing has given us tools to decipher the mysteries of physical, chemical and biological systems in nature and allowed us to model, analyze various observations and use information to optimize our interactions with each other and with our environment, it falls short in reproducing even the basic behaviors of living organisms. We present the foundational shortcomings of CCS and discuss the science of infor-mation processing structures (SIPS) that allows us to fill the gaps. SIPS allows us to model su-per-symbolic computations and infuse autopoietic and cognitive behaviors into digital machines. They use common knowledge representation from the information gained using both symbolic and sub-symbolic computations in the form of system-wide knowledge networks consisting of knowledge nodes and information sharing channels with other knowledge nodes. The knowledge nodes wired together fire together to exhibit autopoietic and cognitive behaviors.

Specifity of Including of Structural Nonlinearity in Model of Dynamics of Cable-Driven Robot

The paper deals with a problem of modeling of the dynamics of a parallel cable-driven robot with the inclusion of structural nonlinearity of cables in a mathematical model. Mathematical model is implemented in a computer model with the possibility of using of symbolic calculations. Parallel cable robots as a type of robotics have been developing in the last two or three decades. The research in the theoretical field was being carried out and the mathematical model of the cable system was being refined with the spread of the practical use of cable robots. This is a non-trivial task to draw up a dynamic model of a cable-driven robot. Cable-driven robots are highly nonlinear systems, because of the main reason for the nonlinearity is the properties of the cable system. As an element of a mechanical system, the cable or the wire rope is a unilateral constraint, since the cable works only for stretching, but not for compression. Thus, the cables are structurally nonlinear elements of the system. On the other hand, cables have the property of sagging under their own weight. Thus, the cables are geometrically nonlinear elements of the system. Under the condition of a payload mass that is utterly greater than the mass of each cable, the cables can be considered strained without sagging and geometric nonlinearity can be neglected. Since symbolic computations can be used in a computer model which implements a mathematical model of the dynamics of a robot, in such a way it must provide the possibility of symbolic computations with the condition of structural nonlinearity. The main aim of this work is to develop a method that ensures the inclusion of the structural nonlinearity of the cable system in the mathematical model. It is supposed to consider the possibility of implementation of the computer model with symbolic computations. The problem of including a mathematical model of cables as unilateral constraints in the model of highly loaded cable robots is considered. The justification for including the activation functions in a system of differential equations of dynamics of cable-driven robot is formulated. A model of wire ropes as unilateral constraints is represented via including the activation functions in a system of differential equations. With using of the proposed method, numerical solution of a problem of forward dynamics has been obtained for high-loaded parallel cable-driven robot.

Minimizing Curvature in Euclidean and Lorentz Geometry

In this paper, an interesting symmetry in Euclidean geometry, which is broken in Lorentz geometry, is studied. As it turns out, attempting to minimize the integral of the square of the scalar curvature leads to completely different results in these two cases. The main concern in this paper is about metrics in R3, which are close to being invariant under rotation. If we add a time-axis and let the metric start to rotate with time, it turns out that, in the case of (locally) Euclidean geometry, the (four-dimensional) scalar curvature will increase with the speed of rotation as expected. However, in the case of Lorentz geometry, the curvature will instead initially decrease. In other words, rotating metrics can, in this case, be said to be less curved than non-rotating ones. This phenomenon seems to be very general, but because of the enormous amount of computations required, it will only be proved for a class of metrics which are close to the flat one, and the main (symbolic) computations have been carried out on a computer. Although the results here are purely mathematical, there is also a connection to physics. In general, a deeper understanding of Lorentz geometry is of fundamental importance for many applied problems.

Spike-based symbolic computations on bit strings and numbers

The brain uses recurrent spiking neural networks for higher cognitive functions such as symbolic computations, in particular, mathematical computations. We review the current state of research on spike-based symbolic computations of this type. In addition, we present new results which show that surprisingly small spiking neural networks can perform symbolic computations on bit sequences and numbers and even learn such computations using a biologically plausible learning rule. The resulting networks operate in a rather low firing rate regime, where they could not simply emulate artificial neural networks by encoding continuous values through firing rates. Thus, we propose here a new paradigm for symbolic computation in neural networks that provides concrete hypotheses about the organization of symbolic computations in the brain. The employed spike-based network models are the basis for drastically more energy-efficient computer hardware -- neuromorphic hardware. Hence, our results can be seen as creating a bridge from symbolic artificial intelligence to energy-efficient implementation in spike-based neuromorphic hardware.

Optical solitons and other soluions for Radhakrishnan–Kundu–Lakshmanan equation in birefringent fibers by an efficient computational technique

In this article, we are interested to discuss the exact optical soiltons and other solutions in birefringent fibers modeled by Radhakrishnan-Kundu-Lakshmanan equation in two component form for vector solitons. We extract the solutions in the form of hyperbolic, trigonometric and exponential functions including solitary wave solutions like multiple-optical soliton, mixed complex soliton solutions. The strategy that is used to explain the dynamics of soliton is known as generalized exponential rational function method. Moreover, singular periodic wave solutions are recovered and the constraint conditions for the existence of soliton solutions are also reported. Besides, the physical action of the solution attained are recorded in terms of 3D, 2D and contour plots for distinct parameters. The achieved outcomes show that the applied computational strategy is direct, efficient, concise and can be implemented in more complex phenomena with the assistant of symbolic computations. The primary benefit of this technique is to develop a significant relationships between NLPDEs and others simple NLODEs and we have succeeded in a single move to get and organize various types of new solutions. The obtained outcomes show that the applied method is concise, direct, elementary and can be imposed in more complex phenomena with the assistant of symbolic computations

A Different Look at Polynomials within Man-Machine Interaction

Undoubtedly, one of the most powerful applications that allow symbolic computations is Wolfram Mathematica. However, it turns out that sometimes Mathematica does not give the desired result despite its continuous improvement. Moreover, these gaps are not filled by many authors of books and tutorials. For example, our attempts to obtain a compact symbolic description of the roots of polynomials or coefficients of a polynomial with known roots using Mathematica have often failed and they still fail. Years of our work with theory, computations, and different kinds of applications in the area of polynomials indicate that an application ‘offering’ the user alternative methods of solving a given problem would be extremely useful. Such an application would be valuable not only for people who look for solutions to very specific problems but also for people who need different descriptions of solutions to known problems than those given by classical methods. Therefore, we propose the development of an application that would be not only a program doing calculations but also containing an interactive database about polynomials. In this paper, we present examples of methods and information which could be included in the described project.

Lump and Lump–Kink Soliton Solutions of an Extended Boiti–Leon–Manna–Pempinelli Equation

In this paper, the extended Boiti–Leon–Manna–Pempinelli equation (eBLMP) is first proposed, and by Ma’s [1] method, a class of lump and lump–kink soliton solutions is explicitly generated by symbolic computations. The propagation orbit, velocity and extremum of the lump solutions on (x,y) plane are studied in detail. Interaction solutions composed of lump and kink soliton are derived by means of choosing appropriate real values on obtained parameter solutions. Furthermore, 3-dimensional plots, 2-dimensional curves, density plots and contour plots with particular choices of the involved parameters are depicted to demonstrate the dynamic characteristics of the presented lump and lump–kink solutions for the potential function v = 2ln( f(x))x.