Discrete Mathsfor Engineers

Structures, proofs, and intuition

EN | PT

IDA*

Series overview — memory-efficient heuristic search with iterative deepening

Series: IDA* 5 parts
1 2 3 4 5

What IDA* Solves

IDA* is a classical algorithm for solving shortest-path and state-space search problems with heuristic guidance while using significantly less memory than A*.

Conceptually, it combines the evaluation function f(n) = g(n) + h(n) with iterative deepening. Instead of storing a full frontier in a priority queue, IDA* performs repeated depth-first searches under progressively increasing cost thresholds.

This makes IDA* especially useful in large search spaces where A* becomes impractical due to memory consumption, while still preserving optimality under suitable conditions on the heuristic.

Reference: the presentation follows the classical treatment used in algorithm design and artificial intelligence, connecting heuristic evaluation, iterative deepening, and memory-efficient search.

Parts