Rosenblatt’s Perceptron

Frank Rosenblatt was pioneering neural networks and connectionist machine learning in the 1950s with the Mark I Perceptron. While the term perceptron now refers primarily to a learning algorithm, Rosenblatt’s perceptron was the physical machine that executed it.

In a 1958 New York Times article (below), Rosenblatt conveys an ambitious (prescient?) vision of the future of machine learning. The article refers to Rosenblatt’s perceptron as “the embryo of an electronic computer that [the Navy] expects will be able to walk, talk, see, write, reproduce itself and be conscious of its existence.”

The strengths and weaknesses of perceptrons were reviewed in Perceptrons (1969), by Minsky and Papert, and somehow people came away with the misconception that perceptrons were severely limited. Rosenblatt died in a boating accident in 1971. A so-called “connectionist winter” that lasted until the 1980s followed.

New York Times Article, 1958 (Thanks to a long lost student for physically tracking this down at the library, scanning, and sending.)

Perceptrons Applied to Images (video)