500 citations · 1.4k across the 23 of their papers we have counts for
3 papers · 1 filter
Learning Compressed Transforms with Low Displacement Rank
Anna T. Thomas, Albert Gu, Tri Dao +2
The low displacement rank (LDR) framework for structured matrices represents a matrix through two displacement operators and a low-rank residual. Existing use of LDR matrices in de…
Representation Tradeoffs for Hyperbolic Embeddings
Christopher De Sa, Albert Gu, Christopher Ré +1
Hyperbolic embeddings offer excellent quality with few dimensions when embedding hierarchical data structures like synonym or type hierarchies. Given a tree, we give a combinatoria…
A Kernel Theory of Modern Data Augmentation
Tri Dao, Albert Gu, Alexander J. Ratner +3
Data augmentation, a technique in which a training set is expanded with class-preserving transformations, is ubiquitous in modern machine learning pipelines. In this paper, we seek…