Výsledky vyhledávání - "Singh, Pulkit"

Report

On the Informativeness of Supervision Signals

Autor: Sucholutsky, Ilia, Battleday, Ruairidh M., Collins, Katherine M., Marjieh, Raja, Peterson, Joshua C., Singh, Pulkit, Bhatt, Umang, Jacoby, Nori, Weller, Adrian, Griffiths, Thomas L.

Supervised learning typically focuses on learning transferable representations from training examples annotated by humans. While rich annotations (like soft labels) carry more information than sparse annotations (like hard labels), they are also more

Externí odkaz: http://arxiv.org/abs/2211.01407

Zobrazit plný text záznamu

Report

End-to-end Deep Prototype and Exemplar Models for Predicting Human Behavior

Autor: Singh, Pulkit, Peterson, Joshua C., Battleday, Ruairidh M., Griffiths, Thomas L.

Traditional models of category learning in psychology focus on representation at the category level as opposed to the stimulus level, even though the two are likely to interact. The stimulus representations employed in such models are either hand-des

Externí odkaz: http://arxiv.org/abs/2007.08723

Zobrazit plný text záznamu