Researchers have created an AI model that uses sequences of life events—such as health history, education, job, and income—to predict everything from a person’s personality to their own mortality—for an entire country.
Built using transformer models, which power large language models (LLMs) like ChatGPT, the new tool, life2vec, is trained on a data set pulled from the entire population of Denmark—6 million people. The data set was made available to the researchers by the Danish government.
The tool is capable of predicting the future, with an accuracy that exceeds state-of-the-art models, says Tina Eliassi-Rad, professor of computer science and the inaugural President Joseph E. Aoun Professor at Northeastern University in Boston.
Unique human-centered model
“Even though we’re using prediction to evaluate how good these models are, the tool shouldn’t be used for prediction on real people,” says Chenru Duan, lead author of a paper recently published in Nature Computational Science.
“It is a prediction model based on a specific data set of a specific population. These tools allow you to see into your society in a different way: the policies, rules and regulations. You can think of it as a scan of what is happening on the ground.”
By involving social scientists in the process of building this tool, the team hopes it brings a human-centered approach to AI development, one that doesn’t lose sight of the humans amid the massive data set their tool has been trained on.
Confidential training data
A massive data set was used to train the life2vec model. The data is held by Statistics Denmark, the central authority on Danish statistics, and is tightly controlled because it includes a detailed registry of every Danish citizen. Although tightly regulated, it can be accessed by some members of the public, including researchers, according to the researchers.
The researchers hope the model can kickstart a public conversation about the power of these tools and how they should (and shouldn’t) be used.
Citation: Duan, C., Du, Y., Jia, H. et al. Accurate transition state generation with an object-aware equivariant elementary reaction diffusion model. Nat Comput Sci 3, 1045–1055 (2023). Also available: open-access draft arXiv version
