Past, meet future. Artificial Intelligence, in the form of a program running on an artificial neural network, is being used to pinpoint the best locations to dig for fossils.
How does it work? The model’s trainers isolated the clues that archaeologists use when covering the ground on foot, and taught the AI program to evaluate the Great Divide Basin in Wyoming based on those criteria.
AI-Fossil Finders: The Research Team
Robert Anemone, Charles Emerson, and Glenn Conroy are the researchers behind this new method of finding fossils. Robert Anemone is Professor of Anthropology and Charles Emerson is Associate Professor of Geography – both at Western Michigan University. Glenn Conroy is Professor of Anatomy and Anthropology at Washington University School of Medicine, in St. Louis, MO.
Decoded Science had the opportunity to ask Professor Conroy about this research.
Decoded Science: Did you train the software through supervised learning, unsupervised learning, or a combination of the two?
Professor Conroy: Training the neural network classifier involved supervised learning, although my colleague, Dr. Charles Emerson, did do some unsupervised classifications early on when he was picking training sites for some of the other land cover classes such as soils of different colors, sagebrush, etc.
Decoded Science: How long did the training process take?
Professor Conroy: Training time varied according to the complexity of the model, but the one we eventually used took about 90 minutes from start to finish (training plus classification).
Supervised Learning, Unsupervised Learning, or Both?
Machine learning is a process by which an AI program is ‘taught’ to perform tasks. Programmers can use supervised learning, unsupervised learning, or a combination of both.
- Supervised learning is a type of machine learning in which the program is given examples to learn from, including the exact process used to reach the desired result.
- In unsupervised learning, the program is only given the data, and the desired result. Programs that use unsupervised learning are able to ‘choose’ the most efficient and effective method of processing the data, to achieve the desired result.
AI Technology for Other Locations
The program used to locate fossils was specific to one area in Wyoming, but this research is applicable to other areas as well. According to the researchers,
“While we have developed and tested this model on fossil mammal localities in deposits of Paleocene and Eocene age in the Great Divide Basin of southwestern Wyoming, a similar analytical approach can be easily applied to fossil-bearing sedimentary deposits of any age in any part of the world.”
Anemone, R., Emerson, C., Conroy, G. Finding Fossils in New Ways: An Artificial Neural Network Approach to Predicting the Location of Productive Fossil Localities. Evolutionary Anthropology 20:169–180 (2011).
Washington University in St. Louis. Human, artificial intelligence join forces to pinpoint fossil locations. (2011). Accessed November 26, 2011.
Decoding Science. One article at a time.