IV. Ecosmomics: An Independent, UniVersal, Source Code-Script of Generative Complex Network Systems
2. Biteracy: Natural Algorithmic Computation
Burgin, Mark and Eugene Eberbach. Evolutionary Computation and the Processes of Life. ubiquity.acm.org/symposia2012.cfm?volume=2012. Ubiquity symposia are organized by the Association for Computing Machinery (ACM) which consider “Information Everywhere.” Some other events are The Technology Singularity and The Science in Computer Science. This is a review of presentations from a symposia with the title name such as The Essence of Evolutionary Computation by Xin Yao, Darwinian Software Engineering by Moshe Sipper and Information, Biological and Evolutionary Computing by Walter Riofrio. To sum up, a new 21st century synthesis seems to be arising whence a creative self-optimizing program can be realized at work prior to selective effects. As a follow up, we added this section in 2015 to report a growing interest in and turn to this vital dimension. Compare then these 2012 inklings with later 2017 postings herein by Hector Zenil, Wolfgang Banzhaf and more.
Evolution is one of the indispensable processes of life. After biologists found basic laws of evolution, computer scientists began simulating evolutionary processes and using operations discovered in nature for solving problems with computers. As a result, they brought forth evolutionary computation, inventing different kinds operations and procedures, such as genetic algorithms or genetic programming, which imitated natural biological processes. Thus, the main goal of our Symposium is exploration of the essence and characteristic properties of evolutionary computation in the context of life and computation. (Peter Denning Editor)
Cabessa, Jeremie and Hava Siegelmann.
The Computation Power of Interactive Recurrent Neural Networks.
Network: Computation in Neural Systems.
University of Massachusetts, Amherst, computational neuroscientists take these cerebral complexities to exemplify how nature evolves, develops and learns. We are then invited to realize that the same dynamical trial and error, feedback to move forward, iterative process is in effect everywhere. See also Turing on Super-Turing and Adaptivity by Hava Siegelmann in Progress in Biophysics and Molecular Biology (113/117, 2013), and search Richard Watson 2014 herein.
In classical computation, rational- and real-weighted recurrent neural networks were shown to be respectively equivalent to and strictly more powerful than the standard Turing machine model. Here, we study the computational power of recurrent neural networks in a more biologically oriented computational framework, capturing the aspects of sequential interactivity and persistence of memory. In this context, we prove that so-called interactive rational- and real-weighted neural networks show the same computational powers as interactive Turing machines and interactive Turing machines with advice, respectively. A mathematical characterization of each of these computational powers is also provided. It follows from these results that interactive real-weighted neural networks can perform uncountably many more translations of information than interactive Turing machines, making them capable of super-Turing capabilities. (Abstract)
Cardelli, Luca, et al. Efficient Switches in Biology and Computer Science. PLoS Computational Biology. 13/1, 2017. Oxford University and King’s College London researchers consider cross-integrations of computational algorithms with systems biology and genomics. By this synthesis, life becomes distinguished and explained by complex information flows and network phenomena.
We have seen that fundamental ideas can cross disciplines, and the findings of one discipline can be readily used to solve problems in another. As we have seen cross fertilizations in far, distinct fields, like how banking systems can be driven by rules learned from ecology, we believe that computer science and biology still have many insights to share. Both disciplines investigate paradigms such as robustness, efficiency, and reliability. Influence can go both ways: biological findings can be used to improve the design of algorithms and theories, and computational concepts can help us use the increasing amount of experimental data to improve our understanding of complex biological systems. (12)
Cardinot, Marcos, et al. Evoplex: A Platform for Agent-Based Modeling on Networks. SoftwareX. 9/199, 2019. We cite this entry from the National University of Ireland, Galway and University of Maribor, Slovenia (Matjaz Perc) as an example of how computer code programs can likewise be found to take on these ubiquitous complexity formats.
Agent-based modeling and network science have been used extensively to advance our understanding of emergent collective behavior in systems that are composed of a large number of simple interacting individuals or agents. Evoplex meets this need by providing a fast, robust and extensible platform for developing agent-based models and multi-agent systems on networks. Each agent is represented as a node and interacts with its neighbors, as defined by the network structure. Evoplex is ideal for modeling complex systems, for example in evolutionary game theory and computational social science. (Abstract excerpt)
Caucheteux, Charlotte and Jean-Remi King. Brains and Algorithms Partially Converge in Natural Language Processing. Communications Biology. 5/134, 2022. Facebook AI Research, Paris and University of Paris Saclay researchers find an intrinsic, parallel affinity between these disparate cognitive phases. In regard, it appears that life and mind ought to parsimoniously use the same patterns and processes in every instance.
Deep learning algorithms trained to predict words from large amount of text have recently been shown to generate activations similar to those of the human brain. However, what drives this similarity is still unknown. Here, we compare deep language models to identify computational principles that generate brain-like representations of sentences. Our analyses reveal that the similarity between the algorithms and the brain primarily depends on their ability to predict words from context. Secondly, this common mode gives rise to and maintains perceptual, lexical, and compositional representations within each cortical region. Overall, this study shows that modern language algorithms converge towards brain-like solutions, which then suggests a way to unravel the foundations of natural language processing. (Abstract)
Chaitin, Gregory. Proving Darwin: Making Biology Mathematical. New York: Pantheon, 2012. The international polymathematician broaches a popular synopsis of his meta-math theories whence life’s regnant complexity is seen to arise from and be distinguished by an algorithmic program. As a gloss, DNA is nature’s evolving software, which leads to an “information-theoretic analysis” of Darwinian selection via hill-climbing on fitness landscapes. The space of possible organisms then consists of these programs, as subject to contingent mutations. Circa 2017, a list of publications can be found at https://independent.academia.edu/GregoryChaitin. His wider import is to be an inspiration and guide to parallel endeavors such as Hector Zenil’s Algorithmic Nature Group.
Chan, Bert. Lenia and Expanded Universe. arXiv:2005.03742. The Hong Kong computational scholar posts his latest, elaborate illustration of how lively and complex the algorithms of John Conway’s Game of Life (see Siobhan Roberts herein) can become. Also Google the author’s name to reach his YouTube videos of Mathematical Life Forms.
We report experimental extensions of Lenia, a continuous cellular automata family capable of producing lifelike self-organizing autonomous patterns. The rule of Lenia was generalized into higher dimensions, multiple kernels, and multiple channels with a final architecture akin to a recurrent convolutional neural network. Genetic algorithm automations led to phenomena like polyhedral symmetries, individuality, self-replication, emission, growth by ingestion, and "virtual eukaryotes" with an internal division of labor and type differentiation. (Abstract)
Chastain, Erick, et al. Algorithms, Games, and Evolution. Proceedings of the National Academy of Sciences. 111/10620, 2014. With Adi Livnat, Christos Papadimitriou, and Umesh Vazirani, theoretical biologists pose a working comparison between natural selection seen as a search, iterate, and optimize process, and a machine learning procedure called “multiplicative weight updates algorithm” (MWUA). Both modes involve explorations and samplings of diverse populations, subject to trials, errors, then retests, so to reach a “good enough” state or solution. An alternative, computational view of selection is entered, but not yet seen to infer a prior, independent program. A Commentary in the same issue, Diverse Forms of Selection in Evolution and Computer Science by Nicholas Barton, et al, supports the finding. The authors and colleagues explored these perceptions at a series of Simons Institute spring 2014 Evolutionary Biology and Computation seminars (search).
Even the most seasoned students of evolution, starting with Darwin himself, have occasionally expressed amazement that the mechanism of natural selection has produced the whole of Life as we see it around us. There is a computational way to articulate the same amazement: “What algorithm could possibly achieve all this in a mere three and a half billion years?” In this paper we propose an answer: We demonstrate that in the regime of weak selection, the standard equations of population genetics describing natural selection in the presence of sex become identical to those of a repeated game between genes played according to multiplicative weight updates (MWUA), an algorithm known in computer science to be surprisingly powerful and versatile. MWUA maximizes a tradeoff between cumulative performance and entropy, which suggests a new view on the maintenance of diversity in evolution. (Abstract)
Chazelle, Bernard. Natural Algorithms and Influence Systems. Communications of the ACM. 55/12, 2012. The Princeton University professor of computer science is a pioneer theorist of a distributed computational source for consequent evolutionary complex creativity. By way of intricate concepts, the project is to discern and decipher an innate mathematical quintessence. There is a need for clarification and translation of terms, but the project is aided by a universality of dynamic collective behaviors due to active agents guided by a common program. If living processes are powered by the “software” of nature, then natural selection is the ultimate code optimizer. This lead think piece is introduced by Ali Jadbabaie, a University of Pennsylvania systems engineer, as Natural Algorithms in a Networked World, see quote below.
Algorithms offer a rich, expressive language for modelers of biological and social systems. They lay the grounds for numerical simulations and, crucially, provide a powerful framework for their analysis. The new area of natural algorithms may reprise in the life sciences the role differential equations have long played in the physical sciences. For this to happen, however, an “algorithmic calculus” is needed. We discuss what this program entails in the context of influence systems, a broad family of multiagent models arising in social dynamics. (Abstract)
Chazelle, Bernard. The Convergence of Bird Flocking. Journal of the ACM. 61/4, 2014. The Princeton University computer scientist presents a deeply mathematical exercise upon “a class of natural algorithms known as nondiffusive influences systems.” As this contribution references many studies from statistical physics and complex systems science upon the phenomena they become similar to computational and algorithmic interpretations. The related approaches may each provide a window and description upon the same generative phenomena. A further frontier would be to translate and realize that what we all are trying to express is a natural genetic code. See also his How Many Bits Can a Flock of Birds Compute? in the online journal Theory of Computing (10/Art. 16, 2014).
We found the time it takes for a group of birds to stabilize in a standard flocking model. Each bird averages its velocity with its neighbors lying within a fixed radius. We resolve the worst-case complexity of this natural algorithm by providing asymptotically tight bounds on the time to equilibrium. We reduce the problem to two distinct questions in computational geometry and circuit complexity. (Abstract)
Chazelle, Bernard. Why Natural Algorithms are the Language of the Living World. . An hour-long pithy presentation at Technion – Israel Institute of Technology on April 23, 2013 by the Princeton mathematician and advocate of such a dual software and soft matter reality.
The glory of 20th century physics was in many ways the triumph of mathematics. Lacking the requisite symmetries, the life sciences of today are unlikely to witness a repeat of this miraculous match. Unlike electromagnetism, for example, cancer will not be explained by a few differential equations. The high descriptive complexity of biology seems to call for a new language --- not a language of equations but of algorithms. The challenge is to find it and then decipher it within the world of biology. Just as equations are studied via other equations, so natural algorithms must be approached through the lens of other algorithms, which in turn points to the need for an "algorithmic calculus." I'll sketch what such a program might entail in the context of "influence systems," which form a broad family of multiagent dynamics encountered in the living world.
Chibbaro, Sergio, et al. Reductionism, Emergence and Levels of Reality. Berlin: Springer, 2014. Chibbaro, University of Pierre and Marie Curie, Paris, with Lamberto Rondoni, Torino Polytechnic, and Angelo Vulpiani, Sapienza University of Rome, physicists provide an intense review of scientific method from Galileo through nature’s subdivisions into statistical and quantum mechanics, and onto nascent reassemblies. Along the way an earlier determinism becomes taken over by chaotic dynamics. A 2010s prognosis turns to J. A. Wheeler’s participatory cosmos, and ultimately to Algorithmic Complexity as a “key to understanding nature.”