Aaron Leslie — Software Engineer & Co-founder
aaronleslie.dev · 4,063 words · saved by 1 readers
Aaron Leslie — Software Engineer & Co-founder of Glyph
1. Challenge and result comma.ai's challenge: shrink a 37 MB dashcam video, with the constraint that two frozen neural networks (a SegNet and a PoseNet) must produce the same outputs on the reconstructed video as they would on the original. The score: score=100⋅Dseg+10⋅Dpose+25⋅∣archive∣∣source∣\text{score} = 100 \cdot D_\text{seg} + \sqrt{10 \cdot D_\text{pose}} + 25 \cdot \frac{|\text{archive}|}{|\text{source}|} Lowest wins. Our final number is 0.1988, against a field clustered between 0.32 (top neural) and 2.0 (most codec submissions). At the time of writing, we're #1 by 38%. 2. The…
saved by
related reading
- Aman's AI Journal • Primers • Ilya Sutskever's Top 30aman.ai
- Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnettransformer-circuits.pub
- Composer2.pdfcursor.com
- Generative modelling in latent space – Sander Dielemansander.ai
- Compression and Intelligencegreene.sh
- Image formats: Pixel data from encoders to decodersdeveloper.mozilla.org
- H-Nets - the Past | Goomba Labgoombalab.github.io
- Introduction to Neural Compression | Desi R. Ivanovadesirivanova.com
- Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnettransformer-circuits.pub
- Same Image, Different Score? | Halide Compressionhalide.cx
- MuZero’s first step from research into the real worlddeepmind.com
- Speculative KV coding: losslessly compressing KV cache by up to ~4× using a predictor modelfergusfinn.com