To raise or not to raise: the autonomous learning rate question

Xiaomeng Dong, Tao Tan, Michael Potter, Yun Chan Tsai, Gaurav Kumar, V.  Ratna Saripalli, Theodore Trafalis

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)


There is a parameter ubiquitous throughout the deep learning world: learning rate. There is likewise a ubiquitous question: what should that learning rate be? The true answer to this question is often tedious and time consuming to obtain, and a great deal of arcane knowledge has accumulated in recent years over how to pick and modify learning rates to achieve optimal training performance. Moreover, the long hours spent carefully crafting the perfect learning rate can come to nothing the moment your network architecture, optimizer, dataset, or initial conditions change ever so slightly. But it need not be this way. We propose a new answer to the great learning rate question: the Autonomous Learning Rate Controller. Find it at .

Original languageEnglish
JournalAnnals of Mathematics and Artificial Intelligence
Publication statusAccepted/In press - 2023
Externally publishedYes


  • AutoML
  • Deep learning
  • Learning rate
  • Optimization


Dive into the research topics of 'To raise or not to raise: the autonomous learning rate question'. Together they form a unique fingerprint.

Cite this