Abstract
Deep learning models are based on a combination of neural network architectures, optimization parameters and activation functions. All of them provide exponential combinations whose computational fitness is difficult to pinpoint. The intricate resemblance of the microscopic features that are found in bone surface modifications make their differentiation challenging, and determining a baseline combination of optimizers and activation functions for modeling seems necessary for computational economy. Here, we experiment with combinations of the most resolutive activation functions (relu, swish, and mish) and the most efficient optimizers (stochastic gradient descent (SGD) and Adam) for bone surface modification analysis. We show that despite a wide variability of outcomes, a baseline of relu–SGD is advised for raw bone surface modification data. For imbalanced samples, augmented datasets generated through generative adversarial networks are implemented, resulting in balanced accuracy and an inherent bias regarding mark replication. In summary, although baseline procedures are advised, these do not prevent to overcome Wolpert’s “no free lunch” theorem and extend it beyond model architectures.
Highlights
The use of computer vision (CV) through deep learning (DL) methods has substantially modified the resolution of taphonomic studies
On small samples (n ≤ 100), CV yielded an accuracy of classification > 90% when human experts were systematically producing < 60% correct identifications of tested bone surface modifications (BSM)
For BSM, the augmented samples can be biasing if expanding the least common types of marks only because they are the ones that avoid confusion with the other categories
Summary
The use of computer vision (CV) through deep learning (DL) methods has substantially modified the resolution of taphonomic studies. DCNN models go as far as to differentiate cut marks imparted on bones when carcasses were fleshed or defleshed [2]. These methods are even capable of detecting BSM morphing through dynamic impact of biostratinomic abrasion processes [3]. Some DL models have even successfully classified tooth marks from different felid types [5,6]. All this shows the promising path ahead in the use of these techniques for taphonomic research
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.