buffering...

Open source ML pipelines, social and cultural musings...an all-around great time

The hottest Substack posts of buffering...

And their main takeaways

Three thoughts on turning policy calls for ‘interpretable’ AI systems into action

0 implied HN points • 09 Aug 23

🕹 Technology AI Policy Research Interpretability Regulation

The algorithms in deep learning systems are mostly unknown, making it challenging to assess their learning process and how they generate output.
Firms like Anthropic are investing in making AI algorithms more interpretable, but more support is needed.
To promote the development of interpretable AI systems, measures like grants, collaboration across disciplines, and improving existing techniques are crucial.

Sparse Dictionary Learning and Transformer Interpretability

0 implied HN points • 15 Aug 23

🕹 Technology Interpretability Language Models

Sparse dictionary learning helps in managing ambiguity in representing input data.
Applying sparse dictionary learning to language models like BERT and GPT reveals tiers of semantic structures.
Future research could explore improving data curation methods and adapt the technique for larger models like GPT.