liveBook · Manning

Chain rule

As explained in Applied Reinforcement Learning

The chain rule is a calculus rule for differentiating a composition of functions by multiplying derivatives of its successive components.

The chain rule differentiates a composition by linking partial derivatives across successive functions. In neural-network backpropagation, it carries the loss gradient backward through layers to estimate each parameter’s contribution; intermediate results are reused rather than recalculated. section 8.2

∂ℒ/∂θ i = ∂ℒ/∂y ∂y/∂f l ∂f l/∂f l-1 ⋯ ∂f i/∂θ i

The equation shows the loss gradient with respect to a parameter as a chain of partial derivatives through the network. section 8.2

How Applied Reinforcement Learning covers it

How deep it goes
Defines derivative chaining through composed layers and notes reuse of intermediate results, without a full derivation.
Examples
Uses a layered neural network of linear transformations and nonlinearities as the composed-function example.
What you can do after
Understand how backpropagation traces loss gradients through layers to estimate each parameter’s contribution to prediction error.
How it connects
Connects function composition and partial derivatives to backpropagation, neural networks, loss, gradients, and parameter optimization.

Also explained in

Compare how 7 books explain chain rule →

Related concepts

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