PNAS

Curriculum learning for human compositional generalization

In novel situations, people need to repurpose past knowledge to guide behavior (generalization). How they do this remains a mystery in cognitive science. Moreover, building machines that can achieve ...
Frontiers

Leveraging Prior Concept Learning Improves Generalization From Few Examples in ...

1 Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA, United States 2 Department of Neuroscience, Georgetown University Medical Center, Washington, DC, ...
GitHub

Generating Less Certain Adversarial Examples Improves Robust Generalization

This paper revisits the robust overfitting phenomenon of adversarial training. Observing that models with better robust generalization performance are less certain in predicting adversarially ...
IEEE

Tightening Mutual Information-Based Bounds on Generalization Error

Abstract: An information-theoretic upper bound on the generalization error of supervised learning algorithms is derived. The bound is constructed in terms of the ...
Microsoft

How Do In-Context Examples Affect Compositional Generalization?

Compositional generalization–understanding unseen combinations of seen primitives–is an essential reasoning capability in human intelligence. The AI community mainly studies this capability by ...