Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networks

Akshay Kumar Jagadish, Julian Coda-Forno, Mirko Thalmann, Eric Schulz, Marcel Binz
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:21121-21147, 2024.

Abstract

Ecological rationality refers to the notion that humans are rational agents adapted to their environment. However, testing this theory remains challenging due to two reasons: the difficulty in defining what tasks are ecologically valid and building rational models for these tasks. In this work, we demonstrate that large language models can generate cognitive tasks, specifically category learning tasks, that match the statistics of real-world tasks, thereby addressing the first challenge. We tackle the second challenge by deriving rational agents adapted to these tasks using the framework of meta-learning, leading to a class of models called ecologically rational meta-learned inference (ERMI). ERMI quantitatively explains human data better than seven other cognitive models in two different experiments. It additionally matches human behavior on a qualitative level: (1) it finds the same tasks difficult that humans find difficult, (2) it becomes more reliant on an exemplar-based strategy for assigning categories with learning, and (3) it generalizes to unseen stimuli in a human-like way. Furthermore, we show that ERMI’s ecologically valid priors allow it to achieve state-of-the-art performance on the OpenML-CC18 classification benchmark.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-jagadish24a, title = {Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networks}, author = {Jagadish, Akshay Kumar and Coda-Forno, Julian and Thalmann, Mirko and Schulz, Eric and Binz, Marcel}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {21121--21147}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/jagadish24a/jagadish24a.pdf}, url = {https://proceedings.mlr.press/v235/jagadish24a.html}, abstract = {Ecological rationality refers to the notion that humans are rational agents adapted to their environment. However, testing this theory remains challenging due to two reasons: the difficulty in defining what tasks are ecologically valid and building rational models for these tasks. In this work, we demonstrate that large language models can generate cognitive tasks, specifically category learning tasks, that match the statistics of real-world tasks, thereby addressing the first challenge. We tackle the second challenge by deriving rational agents adapted to these tasks using the framework of meta-learning, leading to a class of models called ecologically rational meta-learned inference (ERMI). ERMI quantitatively explains human data better than seven other cognitive models in two different experiments. It additionally matches human behavior on a qualitative level: (1) it finds the same tasks difficult that humans find difficult, (2) it becomes more reliant on an exemplar-based strategy for assigning categories with learning, and (3) it generalizes to unseen stimuli in a human-like way. Furthermore, we show that ERMI’s ecologically valid priors allow it to achieve state-of-the-art performance on the OpenML-CC18 classification benchmark.} }
Endnote
%0 Conference Paper %T Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networks %A Akshay Kumar Jagadish %A Julian Coda-Forno %A Mirko Thalmann %A Eric Schulz %A Marcel Binz %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-jagadish24a %I PMLR %P 21121--21147 %U https://proceedings.mlr.press/v235/jagadish24a.html %V 235 %X Ecological rationality refers to the notion that humans are rational agents adapted to their environment. However, testing this theory remains challenging due to two reasons: the difficulty in defining what tasks are ecologically valid and building rational models for these tasks. In this work, we demonstrate that large language models can generate cognitive tasks, specifically category learning tasks, that match the statistics of real-world tasks, thereby addressing the first challenge. We tackle the second challenge by deriving rational agents adapted to these tasks using the framework of meta-learning, leading to a class of models called ecologically rational meta-learned inference (ERMI). ERMI quantitatively explains human data better than seven other cognitive models in two different experiments. It additionally matches human behavior on a qualitative level: (1) it finds the same tasks difficult that humans find difficult, (2) it becomes more reliant on an exemplar-based strategy for assigning categories with learning, and (3) it generalizes to unseen stimuli in a human-like way. Furthermore, we show that ERMI’s ecologically valid priors allow it to achieve state-of-the-art performance on the OpenML-CC18 classification benchmark.
APA
Jagadish, A.K., Coda-Forno, J., Thalmann, M., Schulz, E. & Binz, M.. (2024). Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networks. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:21121-21147 Available from https://proceedings.mlr.press/v235/jagadish24a.html.

Related Material