Artificial intelligence finds amazing patterns in the extinction of the Earth’s biological mass

Charles Darwin’s landmark, On the origin of the species, Concludes with a beautiful summary of his theory of evolution, “This view of life has a grandeur, with its many powers, originally breathed in some form or one; and that, while the planet has been cycling. From such a simple beginning, the infinite forms have been the most beautiful and the most wonderful, and are happening, evolving. “In fact, scientists now know that most of the species that exist have become extinct. This extinction of species has been almost balanced by the emergence of new people throughout the history of the planet, with some major temporary imbalances scientists calling mass extinction events. Scientists have long believed that mass extinction leads to a productive period of species evolution, or “radiation,” which is called “creative extinction.” A new study led by scientists affiliated with the Earth-Life Science Institute (ELSI) at the Tokyo Institute of Technology used machine learning to investigate the co-occurrence of fossil species and found that radiation and extinction are rarely linked, and large The possibility of extinction is rare. Causes radiation of comparable scale.

Creative destruction is central to the classic concepts of evolution. It seems clear that there are some periods in which many species suddenly disappear, and many new species suddenly appear. However, radiation of a scale comparable to mass extinction, which this study calls mass radiation, has received far less analysis than extinction events. In this study, the effects of both extinction and radiation during the period in which the residue is available are compared, the so-called phenorozoic ions. Phenerozoic (Greek meaning “apparent life”), a total of 4 Earth. Billion represents the most recent 550-million-year period in history, and is significant for paleontologists: before this period most organisms were microorganisms that did not easily form fossils, so it is difficult to observe previous evolutionary records. The new study suggests that species did not give a good description of how they evolved or became extinct during the Phenyrozic, and suggests that when life entered a new evolutionary and ecological realm, there were many significant periods of evolutionary radiation, such as the eruption of animal diversity during the Cambrian and the forest. Carboniferous expansion of biomes. It is not known whether this is true for the last 3 3 billion years dominated by microbes, as the lack of recorded information on such ancient diversity does not allow for similar analysis.

Paleontologists have identified some of the most serious, mass extinctions in the Phenerozoic fossil record. These mainly include the Big Five mass extinctions, such as the End-Permian mass extinction, with more than 70% of the species estimated to be extinct. Biologists have suggested that we may now enter the “sixth mass extinction”, which they believe is mainly due to human activity, including changes in hunting and land use, due to the expansion of agriculture. The most commonly reported example of the previous “Big Five” mass extinction is the Cretaceous-Tertiary one (commonly known as “Katie”) which appears to have occurred when a meteorite hit Earth 3 million years ago. , Clearing non-avian dinosaurs. By observing the fossil record, scientists believed that mass extinction events produced particularly productive radiation. In the Katy dinosaur-extermination event, for example, it is traditionally believed that a debris was created, allowing mammals to re-mix and “radiate” the organism, eventually laying the foundation. For the emergence of man. In other words, if the Katie event of “Creative Destruction” hadn’t happened, we probably wouldn’t be here to discuss this question.

The new study begins with a casual discussion in ELSI’s “Agora”, in which ELSI scientists and visitors mostly have lunch and have new conversations. The two authors of the paper, evolutionary biologist Jennifer Howell Kuthil (now a research fellow at the University of Essex in the UK) and physicist / machine learning expert Nicholas Gutenberg (now a postdoctoral researcher at Cross Labs) , Kicking around the question of whether machine learning can be used to visualize and understand fossil records. During the ELSI visit, before the Covid-19 epidemic began to ban international travel, they worked tirelessly to enhance their analysis to investigate the relationship between extinction and radiation events. These discussions allowed them to combine their new data with the breadth of existing ideas on mass extinction and radiation. They quickly discovered that evolutionary methods known as machine learning differed significantly from traditional interpretations.

The team used an innovative application of machine learning to investigate the temporary co-occurrence of species in the phenytoin fossil record, examining more than one million entries in a huge curated, public database, including nearly two million species.

“Some of the most challenging aspects of understanding the history of life are the enormous time period and the number of species involved. New machine learning programs can help us visualize this information in human-readable form,” said lead author Dr. H. Hoyle Kuthill. This means that if we can say, we can hold the evolution of half a billion years in the palms of our hands and gain new insights from what we see. “

Using their objective methods, they found that the “big five” mass extinction events previously identified by paleontologists were captured by machine learning methods, with the top 5% of significant barriers involving extinction developed radiation or .Lutton, There were seven additional mass extinctions, two combined mass extinction-radiation events, and fifteen mass radiations. Surprisingly, in contrast to previous stories emphasizing the importance of post-delayed radiation, this work found that most comparative mass radiation and extinction were only rarely linked in time, which negates the notion of a causal relationship between them.

Co-author Dr. Nicholas Guttenberg said, “The ecosystem is dynamic, you don’t have to cut the existing part to let something new appear.”

The team further discovered that radiation could in fact bring about major changes in existing ecosystems, which the authors call a “destructive creation”. They found that, during the phanerozygous ion, on average, the species that formed the ecosystem at any one time passed away after about 19 million years. But when mass extinction or radiation occurs, this rate of turnover is much higher.

This gives a new perspective on how modern “sixth extinction” occurs. The Quaternary period, which began 2.5 million years ago, has seen frequent climatic upheavals, including dramatic shifts in glaciers, while high latitudes on Earth are covered with ice. This means that the current “sixth extinction” is erasing biodiversity that has already been disrupted, and the authors suggest that it will take at least 8 million years to return to the long-term average of 19 million years. Dr. H. Hoyle Kuthil comments that “every extinction created on our clock erases a species that has probably existed for millions of years now, complicating the normal process of ‘origin of new species’ instead of being lost.”

###

Reference

J. F. Hoyle Kuthil^{1,2,3 *}, Ann Gutenberg^{Held at 2,4,5} And GE Bud⁶, Specification and effects of extinction measured by the evolutionary decay clock, Nature, DOI: 10.1038 / s41586-020-3003-4

1. Analytics and Data Sciences and School of Life Sciences, University of Essex, Vivenho Park, Colchester, CO4 3SQ, UK.
2. Earth-Life Science Institute, Tokyo Institute of Technology, Tokyo, 152-8550, Japan
3. Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge, CB2 3EQ, UK
4. Cross Labs, Cross Compass Ltd., 2-9-11-9F Shinkawa, Chuo-ku, Tokyo 104-0033, Japan
5. Goodaii, no Pattins, 213/23 B, 169 00, Prague, Czech Republic
6. Department of Earth Sciences, Paleobiology Program, Uppsala University, Villavegen 16, SE752 36, Uppsala, Sweden

More info

Tokyo Institute of Technology (Tokyo Tech) As a leading university for science and technology in Japan, it is at the forefront of research and higher education. Tokyo Tech researchers excel in areas ranging from physics to biology, computer science and physics. Founded in 1881, Tokyo Tech hosts more than 10,000 undergraduate and graduate students each year, evolving into scientific leaders and some of the most sought-after engineers in the industry. Embracing the Japanese philosophy of “monotsukuri,” meaning “technological ingenuity and innovation,” the Tokyo tech community seeks to contribute to society through high-impact research.

Institute of Economics and Life Sciences (ELSI) Japan is an ambitious world premier international research center, aiming to achieve progress in interdisciplinary scientific fields by inspiring the world’s great standards to come to Japan and collaborate on very challenging scientific issues. The main objective of ELSI is to consider the origin and co-evolution of the earth and life.

World Premier International Research Center Initiative (WPI) Launched in 2007 by the Ministry of Education, Culture, Sports, Science and Technology to help build globally visible research centers in Japan. These institutes promote high research standards and an outstanding research environment that attracts frontline researchers from around the world. These centers are highly autonomous, allowing them to revolutionize traditional methods of research and administration in Japan.

Cross Labs Machine and Intelligence is a research institute focused on cognitive science and artificial life, with the aim of exposing the mathematical principles of intelligence in natural and artificial systems. Cross Labs was established in 2019 as a bridge between academia and industry, Cross Compass Ltd., a leading AI company providing advanced solutions to manufacturing, robotics, gaming, healthcare, design, and global industry leaders. Marketing. Founded in 2015 in Tokyo, Cross Compass develops advanced techniques for deep learning, machine learning and artificial life, to enhance safety, quality and productivity for the benefit of society. For more information, please visit crosslabs.org and cross-comps.com.

GoodAI Prague-based research and development organization with the goal of helping humanity and understanding the universe – as fast as possible – to develop common artificial intelligence. It was founded by CEO / CTO Marek Rosa in January 2014 and now has more than 30 research scientists, engineers and consultants working in its departments.