2554 ;;Quote: can regard a physical system as a content-addressable memory, if it has many locally stable states; also need to create stable states

2554 ;;Quote: Hopfield's model for content addressable memory consists of neurons that fire with some mean attempt rate

2554 ;;Quote: Hopfield's model differs from Perceptrons by strong back-coupling, emergent computational properties, and asynchronous operation

2556 ;;Quote: Hopfield's networks perform abstract calculations on coded inputs; feature extraction should be already performed

2556 ;;Quote: a neuron net has stable limit points like an Ising model; phase space is dominated by attractors that represent memories

2556 ;;Quote: by raising the threshold can make 0000 represent all unfamiliar states

17 ;;Quote: collective, computational properties can arise from a network of simple neurons and little structure

17+;;Quote: memories are gestalts recalled from a subpart; ambiguities resolved statistically

2558 ;;Quote: brains must be made from a proliferation of simple, well-defined circuits

2558+;;Quote: from large numbers of simple processing elements, get spontaneous emergence of complex computational capabilities