Pith. sign in

Provable Meta-Learning of Linear Representations

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Meta-learning, or learning-to-learn, seeks to design algorithms that can utilize previous experience to rapidly learn new skills or adapt to new environments. Representation learning -- a key tool for performing meta-learning -- learns a data representation that can transfer knowledge across multiple tasks, which is essential in regimes where data is scarce. Despite a recent surge of interest in the practice of meta-learning, the theoretical underpinnings of meta-learning algorithms are lacking, especially in the context of learning transferable representations. In this paper, we focus on the problem of multi-task linear regression -- in which multiple linear regression models share a common, low-dimensional linear representation. Here, we provide provably fast, sample-efficient algorithms to address the dual challenges of (1) learning a common set of features from multiple, related tasks, and (2) transferring this knowledge to new, unseen tasks. Both are central to the general problem of meta-learning. Finally, we complement these results by providing information-theoretic lower bounds on the sample complexity of learning these linear features.

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2026 1

verdicts

ACCEPT 1

roles

background 1

polarities

unclear 1

representative citing papers

Attention-based representations for multi-task computation

cs.LG · 2026-08-04 · accept · novelty 7.0

For min/max readout, two attention heads beat one head by an exponential resource gap, and for n-bit parity and symmetric Boolean functions, heads times polynomial degree must reach the threshold degree, with matching constructions.

citing papers explorer

Showing 1 of 1 citing paper.

  • Attention-based representations for multi-task computation cs.LG · 2026-08-04 · accept · none · ref 27 · internal anchor

    For min/max readout, two attention heads beat one head by an exponential resource gap, and for n-bit parity and symmetric Boolean functions, heads times polynomial degree must reach the threshold degree, with matching constructions.