# Perceptron Branch Predictor
**Entity class:** Processor prediction method
**Domain:** Computer architecture / machine learning / microprocessors
**Doc Type:** Developed Method Node
**Maturity:** Developed
## Definition
A **perceptron branch predictor** uses a simple neural classifier to predict whether a processor branch will be taken. Daniel A. Jiménez and Calvin Lin published the method in 2001, showing that a perceptron could learn correlations across branch history that conventional predictors handled less efficiently.
## Machine-Intelligence Context
Perceptron predictors place learned statistical judgment inside the control loop of a general-purpose processor. Variants entered commercial designs associated with AMD, Oracle's SPARC T4, Samsung's Exynos M1, and IBM systems. The processor therefore uses a neural model to decide which instruction path to prepare before certainty is available.
## Evidence Ledger
- **Established:** Jiménez and Lin published the perceptron prediction method in 2001.
- **Established:** Neural branch-prediction techniques entered commercial processors from multiple vendors.
- **Strongly indicated:** Machine-learning logic became part of mainstream processor operation rather than remaining an external workload.
- **Unresolved:** Vendor-specific implementation details where public documentation is incomplete; microarchitecture manuals, patents, and design testimony would settle each attribution.
## Sources / Provenance
- Daniel A. Jiménez and Calvin Lin, [“Neural Methods for Dynamic Branch Prediction” (2001)](https://www.cs.utexas.edu/ftp/techreports/tr01-50.pdf).
- [IEEE Computer Society, Jiménez Rau Award announcement (2021)](https://computer.org/press-room/2021-news/jimenez-to-receive-ieee-cs-rau-award).
- [[articles/A History of Machine Intelligence|A History of Machine Intelligence]].
## Relationships
- **Implements:** a [[wiki/Perceptron|Perceptron]] inside processor control.
- **Demonstrates:** artifact-level [[wiki/Machine Intelligence Intentionality|Machine Intelligence Intentionality]].
- **Belongs to:** [[wiki/Substrate Intentionality|Substrate Intentionality]].