EXTREME TENSORING FOR LOW-MEMORY PRECONDITIONING

Xinyi Chen, Naman Agarwal, Elad Hazan, Cyril Zhang, Yi Zhang

Research output: Contribution to conferencePaperpeer-review

Abstract

State-of-the-art models are now trained with billions of parameters, reaching hardware limits in terms of memory consumption. This has created a recent demand for memory-efficient optimizers. To this end, we investigate the limits and performance tradeoffs of memory-efficient adaptively preconditioned gradient methods. We propose extreme tensoring for high-dimensional stochastic optimization, showing that an optimizer needs very little memory to benefit from adaptive preconditioning. Our technique applies to arbitrary models (not necessarily with tensor-shaped parameters), and is accompanied by regret and convergence guarantees, which shed light on the tradeoffs between preconditioner quality and expressivity. On a large-scale NLP model, we reduce the optimizer memory overhead by three orders of magnitude, without degrading performance.

Original languageAmerican English
StatePublished - 2020
Event8th International Conference on Learning Representations, ICLR 2020 - Addis Ababa, Ethiopia
Duration: Apr 30 2020 → …

Conference

Conference8th International Conference on Learning Representations, ICLR 2020
Country/TerritoryEthiopia
CityAddis Ababa
Period4/30/20 → …

ASJC Scopus subject areas

  • Education
  • Linguistics and Language
  • Language and Linguistics
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'EXTREME TENSORING FOR LOW-MEMORY PRECONDITIONING'. Together they form a unique fingerprint.

Cite this