Archive

Papers

Notes written down while reading and trying to understand random ML/SWE/CS papers. Originally not intended for the eyes of other people and therefore possibly incomprehensible and/or flat out wrong.

Mar, 2022
- Deep learning notes 11: All is Kernels - but not those in ring0
Feb, 2022
- Deep learning notes 10: Diffusion models - noise to nice in few steps
Nov, 2021
- Deep learning notes 09: Grokking - overfit into generalization
Oct, 2021
- Deep learning notes 08: Fastformer - additive or static (self)attention?
Aug, 2021
- Deep learning notes 07: PonderNet - Learn when to stop thinking
Jul, 2021
- Deep learning notes 06: Part-whole hierarchies with GLOM
May, 2021
- Deep learning notes 05: Unsupervised vision with DINO
Apr, 2021
- Deep learning notes 04: Perceivers and Branch specialization
Feb, 2021
- Deep learning notes 03: Switch, Feedback, and generally transformers
- Deep learning papers/notes 02: GPT3, data extraction, Dall-E
Jan, 2021
- Hypersimensional computing: wide embedding meets Hopfield
- Deep learning papers/notes 01: Performers, Lookahead, rADAM
Dec, 2020
- The Simple Essence of Automatic Differentiation
- Fast weights: attend to the recent past
Jun, 2020
- Gated linear networks

Misc

Various random content I thought might be interesting.

Nov, 2020
- Overleaf: how to maintain separate git repo with nice history
Jun, 2020
- Managed personal VPN between your devices through Azure

Swe

Software engineering of any kind.