This doctoral thesis is an interdisciplinary investigation of the subjectivity inherent in the analysts’ classifications of palaeoenvironmental remains, namely pollen grains and faunal osseous remains. The significant research contributions span from improvements in the post-hoc interpretation of convolutional neural networks, state-of-the-art classification models in pollen classification, and the first application of convolutional neural networks in the classification of bones to species from images.

2022