Week 8 - Nov. 9

Location Our classroom

Theme Practical Necromancy… erm, artificial neural networks (yes, the things that everyone now calls ‘ai’)

Have Read for Today

Perry S. The Enchantment of the Archaeological Record. European Journal of Archaeology. 2019;22(3):354-371. doi:10.1017/eaa.2019.24 link
Huggett, J. (2022). ‘Is Less More? Slow Data and Datafication in Archaeology’. In K. Garstki (Ed.), Critical Archaeology in the Digital Age. UCLA Cotsen Institute of Archaeology Press. pp. 97–110. https://escholarship.org/uc/item/0vh9t9jq#page=112 (scroll to chapter 6 if that link doesn’t quite take you there)
Magnani, Matthew, and Jon Clindaniel. “Artificial Intelligence and the Interpretation of the Past.” Advances in Archaeological Practice 14, no. 2 (2026): 218–33. https://doi-org.proxy.library.carleton.ca/10.1017/aap.2025.10110.
Optional: A complete free book on archaeology, history, and AI by me. link, to have handy should you want it.

The Plan I’ll talk about how we got to now with regard to ‘ai’, versus the much longer use of computers in archaeology that has gone under a variety of names; in particular I want to talk about machine learning in archaeology. We’ll take a look at this: https://predictingthepast.com/aeneas too. And I’ll talk about how we used these techniques in our project looking at the online trade in human remains, and our project looking at climbing culture on Mount Everest.

Skill Building

We’re going to build a very stupid language model from scratch that knows about archaeology.
We’ll build an image classifier
We’re not going to build this, but here is a complete notebook that shows how one might use a technique called ‘retrieval augmented generation’ to explore complex archaeological data link. Feel free to experiment with it though.

Homework By Friday at noon, have your research compendium for this week complete and in github. Think about the ethical issues that these approaches might raise, the questions of labour. Think about the friction you encounter using these.