Biologically Inspired Computing: The Neural Network
John Bryden
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
Artificial intelligence has been the inspiration and goal of computing since the discipline was first conceived by Alan Turing. Our understanding of the brain has increased in parallel with the development of computers capable of modelling its functions. While the human brain is vastly complex, too much so for the computation abilities of modern super computers, interesting results have been found while modelling the nervous system of smaller creatures such as the salamander [3].
Key takeaways
- If enough signals reach the soma to excite it (reach its action potential), then it will fire the whole neuron.
- There are 100 billion neurons [2] (p190) in the human brain; each neuron is commonly connected to thousands of other neurons.
- Taking a set of n MPC neurons in an attractor net, the net will have 2 n states (each neuron can either be at one or two).
- The hidden neurons are all connected to each other, themselves and every input and output neuron, with every connection having an individual weight.
- Neurons (biological and artificial) that commonly fire each other reinforce each other.
Related papers
Cellular Information Processing & Artificial Neurons
2019
In the present review we analyse the key features of cells, seen fundamentally as information processing machines. Even though the comparison between cells and machines or cells and computers provides a powerful tool for understanding how information is stored and processed by the cellular machinery, a series of misleading and simplistic assumptions are critically reviewed. Moreover, the seminal studies on understanding and artificially recreating the neuronal structure are analysed, both for exploiting its powerful computational abilities in information processing devices and for providing medical solutions to neurodegenerative diseases, neurological trauma and neuroprosthetics.
Artificial and Biological Neurons: Interdisciplinary Issues and Future Perspectives. White Paper
2021
Recent developments in the technological domain have increased the interactions between artificial and natural spheres, leading to a growing interest in the ethical, legal and philosophical implications of AI research. The present paper aims at creating an interdisciplinary discussion on issues raised by the use and the implementation of artificial intelligence algorithms, robotics, and applied solutions in the neuroscience and biotechnology field. Building on the findings of the webinar “Workshop neuroni artificial e biologici: etica e diritto”, this work explores the issues discussed in the workshop, it attempts to show both the existing challenges and opportunities and it seeks to propose ways forward to overcome some of the investigated problems.
Scholars Research Library Comparative study of biological and artificial neural networks
In this research project, the features of biological and artificial neural networks were studied by reviewing the existing works of authorities in print and electronics on biological and artificial neural networks. The features were then assessed and evaluated and comparative analysis of the two networks was carried out. The metrics such as structures, layers, size and functional capabilities of neurons, learning capabilities, style of computation, processing elements, processing speed, connections, strength, information storage, information transmission, communication media selection, signal transduction and fault tolerance were used as basis for comparison. A major finding in the research showed that artificial neural networks served as the platform for neuro-computing technology and as such a major driver of the development of neuron-like computing system. It was also discovered that Information processing of the future computer systems will greatly be influenced by the adoption of artificial neural network model.
Neural networks in neuroscience: a brief overview
Ever since the publication of Santiago Ramón y Cajal's drawings of neurons - in his words, those "mysterious butterflies of the soul" - it has been clear that the nervous system is composed of a large number of such cells connected to one another to form a network. Long axons, ending in terminals which form synapses to the dendrites which branch out from neighbouring neurons, transmit bursts of electric current and enable neurons somehow to cooperate and yield the astonishing emergent phenomenon known as thought.
Study of a new neuron
Arxiv preprint adap-org/9406001, 1994
We study a modular neuron alternative to the McCulloch-Pitts neuron that arises naturally in analog devices in which the neuron inputs are represented as coherent oscillatory wave signals. Although the modular neuron can compute XOR at the one neuron level, it is still characterized by the same Vapnik-Chervonenkis dimension as the standard neuron. We give the formulas needed for constructing networks using the new neuron and training them using back-propagation. A numerical study of the modular neuron on two data sets is presented, which demonstrates that the new neuron performs at least as well as the standard neuron.
Biologically Plausible Artificial Neural Networks
Artificial Neural Networks - Architectures and Applications, 2013
Revisiting the Neuron Doctrine
The neuron doctrine is the dominant theory of the structure and function of the nervous system. Standard histories of neuroscience celebrate Cajal, who first formulated this theory, and overlook Golgi, who resisted adopting it. A few historians have turned up evidence that challenges this story, suggesting that Golgi accepted the main claim that defines the neuron doctrine. The revised history says that Golgi and Cajal agreed on the facts, but differed in their theoretical preconceptions. I add further details to this revised history. First, I show that the standard history was largely manufactured by Cajal. Second, I argue that Golgi and Cajal agreed only on the fact of there not being anastomosis between cells, but not on anatomical independence. Third, I investigate the reasons why Golgi refused to accept anatomical independence, while Cajal did. I show how the particular brain areas Golgi and Cajal each took as their model system played a part in the construction of their theories.
Dynamic Neurons
Dynamic Neurons, Santiago Ramón y Cajal, 2019
Santiago Ramón y Cajal. Discoveries in the neurosciences made possible by technical advances in the late 19 th century had great influence in the field of psychology. The idea that one can manipulate the very structure of the brain by what one experiences has its roots in the research that led to the discovery of the synapse. Scientists of the late 19 th century diverged from hundreds of years of assumptions about the structure and function of the nervous system. Traditional views were bitterly guarded even as evidence against them mounted. In the end, strong observational research and exciting speculation about the nature of the nervous system laid the groundwork for work now being done in the fields of neuroplasticity and neurogenesis. The field of psychology was to be changed dramatically by the discovery of the dynamic nature of neurons.
Related papers
The Formulation and Evaluation of a Neuron Model Based on Biological Neurons
2008
This thesis formulates and evaluates a mathematical model from an engineer's point of view based on the currently-known information-processing processes and structures of biological neurons. The specification and evaluation of the RealNeuron model form a baseline for current use in engineering solutions and future developments.
How Neurons Mean
2000
Questions concerning the nature of representation and what representations are about have been a staple of Western philosophy since Aristotle. Recently, these same questions have begun to concern neuroscientists, who have developed new techniques and theories for understanding how the locus of neurobiological representation, the brain, operates. My dissertation draws on philosophy and neuroscience to develop a novel theory of representational content. I would also like to single out both Charles H. Anderson, my mentor in computational neuroscience, and William Bechtel who, as my dissertation advisor, has patiently and expertly guided my progress. As well, Pete Mandik and Chase Wrenn deserve special mention for constructive discussions that focused my efforts early on. Finally, I'd like to thank those who are part of the evolution of this thesis in a different sense, and without whom I would not have started, let alone finished:
On Modelling Biological Neurons With Artificial Neural Networks
The first few pages of any good introductory book on neurocomputing contain a cursory description of neurophysiology and how it has been abstracted to form the basis of artificial neural networks as we know them today. In particular, artificial neurons simplify considerably the behavior of their biological counterparts. It is our view that in order to gain a better understanding of how biological systems learn and remember it is necessary to have accurate models on which to base computerized experimentation. In this paper we describe an artificial neuron that is more realistic than most other models used currently. The model is based on conventional artificial neural networks (and is easily computerized) and is currently being used in our investigations into learning and memory.
The physiological construction of the neurone concept
Steps in the physiological construction of the neurone concept are described. Early ideas on the function of the nerve cell led to later polemics on the neurone doctrine and the speculative attitude of histophysiology. Researches of Sherrington and Adrian emerged from a specific British context, and confronted American oscillography and Berger rhythm. At the end of various polemics, the neurone was constructed by the intracellular technique and the use of concepts borrowed from other sub-disciplines. Analysis of these paths demonstrates underlying disciplinary interactions as essential factors.
The formulation and evaluation of a neuron model based on biological neurons / Louwrence Daniël Erasmus
2007
This thesis formulates and evaluates a mathematical model from an engineer's point of view based on the currently-known information-processing processes and structures of biological neurons. The specification and evaluation of the RealNeuron model form a baseline for current use in engineering solutions and future developments. v vi I am thankful to all my teachers, mentors, colleagues and friends who have a part in forming me into the engineer I am today. I am very grateful to my parents for their love and support through all these years of study. Only after having children of my own, can I fully appreciate the sacrifices made by them throughout the years. My sincerest thanks go to my wife Ronel, and daughters Malisa and Bernice for their love, support, encouragement, sacrifices and understanding throughout my research. They are my spiritual support team for big projects. Everyday I experience the omnipresence of Elohim. He looks after me, guides me, and gives me the inner strength to continue my work; Sola Gratia. He lovingly reveals his own workings and works to me. He allows me to ask the difficult questions, to challenge current beliefs and to enrich my own personal relationship with Him; Soli Deo Gloria.
Neural systems engineering
2008
Biological brains and engineered electronic computers fall into different categories. Both are examples of complex information processing systems, but beyond this point their differences outweigh their similarities. Brains are flexible, imprecise, error-prone and slow; computers are inflexible, precise, deterministic and fast. The sets of functions at which each excels are largely non-intersecting. They simply seem to be different types of system.
NEUROSCIENCE: The Neuron Doctrine, Redux
Science, 2005
Information processing, past and present.The Neuron Doctrine transformed the 19th-century view of the nervous system which saw the brain as a network of interconnected nerve fibers (upper left). A century later, the modern view (lower right) holds the neuron as a discrete cell that processes information in more ways than original envisaged: Intercellular communication by gap junctions, slow electrical potentials, action potentials initiated in dendrites, neuromodulatory effects, extrasynaptic release of neurotransmitters, and information flow between neurons and glia all contribute to information processing. ]. (Right) T. thermophilus ribosome at 5.5 Å resolution [from (7, 9)]. Both are oriented such that the small subunit [ribosomal RNA (light blue) and protein (dark blue)] is in the front.
Comparative study of biological and artificial neural networks
In this research project, the features of biological and artificial neural networks were studied by reviewing the existing works of authorities in print and electronics on biological and artificial neural networks. The features were then assessed and evaluated and comparative analysis of the two networks was carried out. The metrics such as structures, layers, size and functional capabilities of neurons, learning capabilities, style of computation, processing elements, processing speed, connections, strength, information storage, information transmission, communication media selection, signal transduction and fault tolerance were used as basis for comparison. A major finding in the research showed that artificial neural networks served as the platform for neuro-computing technology and as such a major driver of the development of neuron-like computing system. It was also discovered that Information processing of the future computer systems will greatly be influenced by the adoption of artificial neural network model.
The Neuroid: A Novel and Simplified Neuron-Model
In this paper we introduce a novel computational neuron-model, the Neuroid, which is based on three basic operations that are carried out by nerve cells to process the incoming information, such as comparison, and frequency pulse modulation-demodulation. The model was implemented using LabVIEW 10.0, in order to assign to each of these operations, an execution block (Virtual Instrument). The results of its implementation showed a very similar behavior to that exhibited by real neurons. Furthermore, due to its simplicity and low computational requirements, it is expected that the Neuroid can be used to create several software models of biological neural systems, either for research or teaching purposes.
Arithmetical operations performed by nerve cells
Brain Research, 1974
A simple neurone model is constructed and analysed mathematically to see what types of operation it can perform on its synaptic inputs. The neurone consists essentially of a soma together with semi-infinite dendrite. Provided their conductance changes are sufficiently small, excitatory synapses are linearly additive and inhibitory synapses are linearly subtractive, irrespective of location. Inhibitory synapses, producing large conductance changes and located preferably on the soma, are ideally suited to carry out division. The extension of this model to the more complex situation of branching dendritic trees is briefly examined and its relevance to real nerve ceils is discussed.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.