2 citations · 3 across the 4 of their papers we have counts for
Showing 2026Show all
2 papers · 1 filter
q-bio.NC2026
How Optimality Structures Sparse Dictionaries: Theory for Interpreting SAE Representations
William Dorrell
Sparse Autoencoders (SAEs) have found success parsing neural network representations into interpretable concepts, providing a basis for understanding and control. However, what exa…
q-bio.NC2026
Convex Efficient Coding
William Dorrell, Peter E. Latham, James Whittington
Why do neurons encode information the way they do? Normative answers to this question model neural activity as the solution to an optimisation problem; for example, the celebrated…