douglasorr

4 ms·

1.
Five game jams: Lessons learned(douglasorr.github.io)
1 points by douglasorr 9mo ago|0 comments
2.
Procedural Textures with Hash Functions(douglasorr.github.io)
32 points by douglasorr 1y ago|5 comments
3.
Llama 3.2 Vision – A Deep Dive(graphcore-research.github.io)
2 points by douglasorr 2y ago|0 comments
4.
Benchmarking std:{string, string_view, reference_wrapper}(douglasorr.github.io)
1 points by douglasorr 2y ago|0 comments
5.
ML papers (June): Mamba-2 and Matmul-free Models(graphcore-research.github.io)
3 points by douglasorr 2y ago|0 comments
6.
Almost-scaled dot-product attention(graphcore-research.github.io)
2 points by douglasorr 3y ago|0 comments
7.
Metaphor Thinking in Deep Learning(douglasorr.github.io)
2 points by douglasorr 3y ago|0 comments
8.
Rasterized ray tracing of sound in JavaScript(douglasorr.github.io)
2 points by douglasorr 5y ago|0 comments
9.
GitHub Game Off – “Bug”(github.blog)
3 points by douglasorr 5y ago|0 comments
10.
Autodiff from scratch(douglasorr.github.io)
59 points by douglasorr 5y ago|1 comments
11.
How to train your classifier – deep learning training objectives(douglasorr.github.io)
1 points by douglasorr 5y ago|0 comments
12.
by douglasorr 5y ago
I think at least some of the structure is down to the update equation, roughly 'new = sign(gate * old + ...)'. Since 'gate' is always positive, this means that even when the network is randomly initialized, the state is
13.
My phone case, the hidden state of an RNN(douglasorr.github.io)
71 points by douglasorr 5y ago|36 comments
14.
Write code for humans. Design data for machines(douglasorr.github.io)
4 points by douglasorr 6y ago|0 comments
15.
Write code for humans. Design data for machines(douglasorr.github.io)
3 points by douglasorr 7y ago|0 comments
16.
Finding unique items: hash vs. sort(douglasorr.github.io)
137 points by douglasorr 7y ago|81 comments