Stress testing MLE

Maximum likelihood estimation under different conditions

54 min read   ·   June 18, 2026

2026   ·   deep-learning   statistics   ·   blogpost

Tanh Function as Drop-In Replacement for Layernorm

Why LayerNorm is not without alternatives

13 min read   ·   March 25, 2025

2025

Towards Group Equivariant Self-Attention

9 min read   ·   January 10, 2024   ·   medium.com

2024

Group Equivariant Self-Attention

1 min read   ·   January 09, 2024   ·   medium.com

2024

Stand-Alone Self-Attention in Vision From Scratch

17 min read   ·   April 28, 2023   ·   medium.com

2023