perceptron
An early neural network with one layer of adjustable weights, introduced by Frank Rosenblatt, with the first provably effective learning algorithm.
As defined in 343 | Tom Griffiths on The Laws of Thought
Other definitions
An early (1950s) brain-inspired learning device that takes weighted inputs and fires if they exceed a threshold; the ancestor of modern neural networks, but with no hidden layer.
68 | Melanie Mitchell on Artificial Intelligence and the Challenge of Common Sense
Where it comes up · 10
- The AI Expert: "Super AI Will Be Unstoppable!" – What’s Coming NEXT | Stephen Wolfram6 claims
- Ray Kurzweil (USA) at Ci2019 - The Future of Intelligence, Artificial and Natural3 claims
- 343 | Tom Griffiths on The Laws of Thought2 claims
- 68 | Melanie Mitchell on Artificial Intelligence and the Challenge of Common Sense2 claims
- Stephen Wolfram's Take on Artificial Intelligence & The Future of Humanity1 claim
- AI: Grappling with a New Kind of Intelligence | World Science Festival1 claim
- The Future of Artificial Intelligence1 claim
- 180 | Camilla Pang on Instructions for Being Human1 claim
and 2 more
Most important claims · 10 of 19Search all claims for “perceptron” →
- Around 1970 Marvin Minsky and Seymour Papert mathematically proved that perceptrons were fundamentally limited in what they could learn no matter how much training data was provided, partly motivated by viewing neural networks as a competitor to their preferred symbolic AI approach.68 | Melanie Mitchell on Artificial Intelligence and the Challenge of Common Sense
- A neural network (such as Rosenblatt's Perceptron) can be understood precisely as a way of computing with spaces: it takes a vector of values (a point in one space) and transforms it into another vector of values (a point in another space), and adding more layers lets it perform multi-step transformations across successive spaces to solve more complex problems—resolving how to compute when concepts are points in space.343 | Tom Griffiths on The Laws of Thought
- Minsky and Papert proved that single-layer Perceptrons (with one layer of adjustable weights) are fundamentally limited to computing linear, local functions of their input and so cannot solve interesting problems; this dampened interest until Rumelhart, Hinton, and McClelland developed learning algorithms for multilayer networks that could solve those more complex problems, leading 40 years later to modern AI mainly through scale.343 | Tom Griffiths on The Laws of Thought
- Frank Rosenblatt's Mark 1 perceptron neural network in 1957 led the Navy to fund neural network research, and the New York Times reported in 1958 that the Navy revealed 'the embryo of an electronic computer today that expects will be able to walk talk see write reproduce itself and be conscious of its existence.'The Future of Artificial Intelligence
- The perceptron, created in the 1950s by psychologist Frank Rosenblatt as a hardware model of linear-threshold neurons, was one of the first brain-based systems that could learn from examples via an adjustable-weight algorithm, but it had no hidden layer.68 | Melanie Mitchell on Artificial Intelligence and the Challenge of Common Sense
- Frank Rosenblatt's 1958 perceptron built artificial brain tissue from transistors as neurons in three layers, learning by trial and error: connections were adjusted (knobs wiggled, changes kept if they helped) until the network recognized patterns on its own—establishing the basic algorithm of tinkering with weights and keeping beneficial changes that underlies all AI learning today.How LLMs Took Over The World
- Marvin Minsky, despite originally being interested in neural networks and having written his PhD thesis on them, decided that perceptrons and neural networks were fundamentally trivial and incapable of doing anything interesting, co-authoring a book with Seymour Papert that effectively ended research in neural networks until the 1980s.The AI Expert: "Super AI Will Be Unstoppable!" – What’s Coming NEXT | Stephen Wolfram
- The perceptron, a single-layer artificial neural network introduced in the 1960s, could recognize some patterns but had fundamental limitations, such as inability to solve the connectedness problem (determining whether shapes are spatially connected), which Minsky's 1969 book mathematically proved.Ray Kurzweil (USA) at Ci2019 - The Future of Intelligence, Artificial and Natural
- The history of AI consists of a repeated cycle where researchers propose a new paradigm, declare it the solution to artificial intelligence, invest heavily, then abandon it when it fails to deliver; this pattern includes the General Problem Solver of the 1950s, perceptrons, expert systems, and early neural networks.AI: Grappling with a New Kind of Intelligence | World Science Festival
- In the early 1960s, Marvin Minsky—an AI pioneer who had originally been interested in neural networks and even written his PhD thesis on them—decided that perceptrons and neural networks were fundamentally trivial and incapable of doing anything interesting. With Seymour Papert, he published a book arguing this thesis, effectively killing mainstream interest in neural networks.The AI Expert: "Super AI Will Be Unstoppable!" – What’s Coming NEXT | Stephen Wolfram