Co-occurring keywords · papers mentioning both
Papers
Why are Sensitive Functions Hard for Transformers?
ACL 2024
A Mathematical Model for Curriculum Learning for Parities
ICML 2023
When Hardness of Approximation Meets Hardness of Learning
JMLR 2022
Sparse Polynomial Learning and Graph Sketching
NeurIPS 2014
Attribute-Efficient and Non-adaptive Learning of Parities and DNF Expressions
JMLR 2007
Toward Attribute Efficient Learning of Decision Lists and Parities
JMLR 2006