# Frank Rosenblatt **Entity class:** Person / psychologist and computer scientist **Domain:** Neural networks / machine learning history **Doc Type:** Developed Person Node **Maturity:** Developed ## Definition **Frank Rosenblatt** (1928–1971) developed the perceptron and led construction of the Mark I Perceptron at the Cornell Aeronautical Laboratory. His work turned a learning rule inspired by neural organization into a trainable physical machine. ## Historical Role Rosenblatt's 1958 perceptron program joined theory, military research funding, sensors, adjustable weights, and automated classification. The perceptron became both a technical ancestor of later neural networks and a public symbol of the thinking-machine objective. ## Evidence Ledger - **Established:** Rosenblatt published the perceptron model and led the Mark I Perceptron program. - **Established:** The system learned classification weights from examples. - **Strongly indicated:** The program transmitted formal-neuron research into trainable hardware and later connectionist work. - **Unresolved:** The degree of direct architectural inheritance from Rosenblatt into each later neural-network family; citations, personnel connections, and design documents settle individual cases. ## Sources / Provenance - Frank Rosenblatt, [“The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain” (1958)](https://doi.org/10.1037/h0042519). - [[articles/A History of Machine Intelligence|A History of Machine Intelligence]]. ## Relationships - **Developed:** [[wiki/Perceptron|Perceptron]] and the Mark I Perceptron. - **Extends:** [[wiki/Warren McCulloch|Warren McCulloch]] and [[wiki/Walter Pitts|Walter Pitts]]. - **Precedes:** modern neural learning and [[wiki/Perceptron Branch Predictor|Perceptron Branch Predictor]].