TY  - RPRT
TI  - Training Large Neural Networks With Low-Dimensional Error Feedback
AU  - Maher Hanut
AU  - Jonathan Kadmon
PY  - 2026
UR  - https://arxiv.org/abs/2502.20580
ID  - 2502.20580
ER  -