Use and Misuse of Machine Learning in Anthropology

Abstract

Machine learning (ML), being now widely accessible to the research community at large, has fostered a proliferation of new and striking applications of these emergent mathematical techniques across a wide range of disciplines. In this paper, we will focus on a particular case study: the field of paleoanthropology, which seeks to understand the evolution of the human species based on biological (e.g. bones, genetics) and cultural (e.g. stone tools) evidence. As we will show, the easy availability of ML algorithms and lack of expertise on their proper use among the anthropological research community has led to foundational misapplications that have appeared throughout the literature. The resulting unreliable results not only undermine efforts to legitimately incorporate ML into anthropological research, but produce potentially faulty understandings about our human evolutionary and behavioral past.The aim of this paper is to provide a brief introduction to some of the ways in which ML has been applied within paleoanthropology; we also include a survey of some basic ML algorithms for those who are not fully conversant with the field, which remains under active development. We discuss a series of missteps, errors, and violations of correct protocols of ML methods that appear disconcertingly often within the accumulating body of anthropological literature. These mistakes include use of outdated algorithms and practices; inappropriate testing/training splits, sample composition, and textual explanations; as well as an absence of transparency due to the lack of data/code sharing, and the subsequent limitations imposed on independent replication. We assert that expanding samples, sharing data and code, re-evaluating approaches to peer review, and, most importantly, developing interdisciplinary teams that include experts in ML are all necessary for progress in future research incorporating ML within anthropology and beyond.

Recruiting a skeleton crew—Methods for simulating and augmenting paleoanthropological data using Monte Carlo based algorithms

“…2, Table 5), which included the separation of training data and testing data prior to the calculation of PCA, algorithms maintain high classification rates (Table 5), with even lower loss values (RMSE = 0.017). This proves that even when trying to reduce the possible contamination of the testing set related to PCA preprocessing as described by Calder et al (2022), MT approaches presented here are able to produce highly powerful classification algorithms that may be more realistically applicable to the fossil record.…”

“…Experiment 1 performed train:test splits after PCA had been performed on the entire dataset, and was then subjected to MT, following the approaches that other studies have used in the past (Courtenay et al, 2021). Experiment 2, however, performed the train:test split on the Procrustes Superimposed coordinates, so as to avoid the possible contamination implied by Calder et al (2022). The train set was then used to calculate PC scores, which were then augmented and the synthetic values used to train a SVM.…”

“…For this purpose, each of the samples was divided into 70:30% train:test sets, and then subjected to the same processes as described above. As a means of testing the possible contamination that may be present product of performing the train:test split after performing PCA (sensu Calder et al, 2022), two experiments were devised. Experiment 1 performed train:test splits after PCA had been performed on the entire dataset, and was then subjected to MT, following the approaches that other studies have used in the past (Courtenay et al, 2021).…”

“…the possible contamination implied byCalder et al (2022). The train set was then used to calculate PC scores, which were then augmented and the synthetic values used to train a SVM.…”