Learning rate

As explained in Transformers in Action

Learning rate is an optimization hyperparameter that controls the magnitude of parameter updates during training.

The learning rate is a hyperparameter that sets the step size for updating model parameters during gradient descent. It is chosen before training, unlike learned weights and biases. section 8.1 section 8.1.1

Each update moves parameters opposite the loss gradient and scales that movement by the learning rate. A value that is too high can overshoot the minimum; a value that is too low can slow training or become stuck in a local minimum. section 8.1.1

Figure 8.1 Gradient descent with lr = 0.1, demonstrating faster convergence

Compare gradient descent with a learning rate of 0.1, which emphasizes faster convergence, against 0.01, which emphasizes stability. section 8.1.1

Configuring learning rate for translation fine-tuning

A supervised fine-tuning configuration assigns learning_rate=lr alongside a linear scheduler and a 0.03 warmup ratio. The setup combines a modest rate with gradient accumulation and a limited batch size for constrained VRAM, then evaluates translation behavior after two epochs. section 7.5

How Transformers in Action covers it

How deep it goes

Explains gradient-scaled updates and high-versus-low trade-offs, then demonstrates configuration with scheduling but stops short of tuning procedures.

Examples

Uses convergence figures at 0.1 and 0.01 plus Gemma 3 270M English–Spanish fine-tuning configuration.

What it assumes

Assumes gradient descent and loss gradients, while distinguishing learned parameters from pretraining hyperparameters.

What you can do after

Operate/configure training with a selected learning rate, scheduler, warmup ratio, and resource-aware fine-tuning setup.

How it connects

Connects gradient descent, model parameters, loss function, learning-rate schedules, batch size, epochs, and overfitting.

liveBookmanning.com

Read Transformers in Action on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Grokking Machine LearningA learning rate is a small positive training parameter that controls the size of parameter updates made during model optimization.
• 

  liveBookmanning.com
  Classic Computer Science Problems in Swift: Essential techniques for practicing programmersA learning rate is a training hyperparameter that controls the size of parameter updates made while fitting a model.
• 

  liveBookmanning.com
  Machine Learning for Tabular Data: XGBoost, Deep Learning, and AIA learning rate is a hyperparameter that scales the size of successive updates made while fitting a model.
• 

  liveBookmanning.com
  Inside Deep Learning: Math, Algorithms, ModelsThe step-size parameter controlling how far an optimization update moves in the direction suggested by the gradient.
• 

  liveBookmanning.com
  Ensemble Methods for Machine Learninglearning_rate is a boosting hyperparameter that scales the contribution of each successive weak learner and controls the rate at which ensemble complexity grows.
• 

  liveBookmanning.com
  Natural Language Processing in Action, Second EditionA learning rate is a numerical step-size setting that controls how far an optimization algorithm moves parameters in response to a gradient.
• 

  liveBookmanning.com
  Deep Learning with R, Third EditionA learning rate is the step-size hyperparameter controlling how far model parameters move during an optimization update.
• 

  liveBookmanning.com
  Deep Learning with PyTorch, Second EditionThe step-size hyperparameter that controls how much model parameters are changed per update during optimization.
• 

  liveBookmanning.com
  Machine Learning Platform EngineeringA learning rate is a training hyperparameter that controls the step size used to update model parameters during optimization.
• 

  liveBookmanning.com
  Math and Architectures of Deep LearningThe step-size setting that controls how far model parameters move on each optimization update, possibly chosen separately for each parameter dimension.
• 

  liveBookmanning.com
  Evolutionary Deep Learning: Genetic algorithms and neural networksA learning rate is the step-size parameter that determines how strongly model parameters are updated during training.
• 

  liveBookmanning.com
  Data Without LabelsA learning rate is a scalar hyperparameter that controls the size of parameter updates during iterative training.
• 

  liveBookmanning.com
  Grokking AI Algorithms, Second EditionA learning rate is a training hyperparameter that controls the step size used to update model parameters.
• 

  liveBookmanning.com
  LLMs in ProductionA learning rate is a training hyperparameter that controls the magnitude of parameter updates during optimization.
• 

  liveBookmanning.com
  Designing Deep Learning Systems: A guide for software engineersA learning rate is a hyperparameter that controls the step size used to update model parameters during optimization.

Compare how 16 books explain learning rate →

Related concepts

• hyperparameterkind of · A hyperparameter is a setting chosen outside ordinary parameter learning that controls a model, algorithm, or learning procedure.
• gradient descentconfigures · Gradient descent is an iterative optimization method that updates parameters opposite the gradient of an objective function to reduce its value.
• learning rate scheduleconfigured by · A learning rate schedule is a rule or function that determines how an optimizer’s learning rate changes during training.
• stochastic gradient descentconfigures · Stochastic gradient descent is an iterative optimization method that updates parameters using gradients estimated from individual examples or randomly selected small batches.
• cosine annealingconfigured by · Cosine annealing is a learning-rate scheduling method that varies the learning rate over training according to a cosine-shaped curve.

This explainer was generated by Manning's knowledge graph from the text of Transformers in Action. The section links point to the parts of the book it draws on.