ConnectionismReading
William Bechtel, "Connectionism and the Philosophy of Mind", Chapter 10 in Lycan, pp252-273.BE ACTIVE: Write down from memory the headings of the key points. Heads.
Connectionism is the view that it is more promising to model cognitive powers on complex networks of computing units rather than on von Neumann computers. The von Neumann computerA central processing unit has a fixed repertoire of operations. The machine is given a program, which lists the operations to be conducted.Difficulties with thinking of the brain as essentially one or a few von Neumann computers:
A connectionist systemA connectionist system consists not of one central processing unit with access to memory but of a network of 'nodes'. Each node is capable of being in a state of activation and is linked to several other nodes. A node is switched on and off according to a rule which is characteristic of a given system. (EG 'switch on when three connections have reached strength S; otherwise switch off.) The strength of each connection can be varied as any particular run of the system proceeds.Think of the nodes as arranged in a rectangle. The a particular run of the system goes like this: The line of nodes on the bottom edge of the rectangle are subjected to a pattern of activation (i.e. some are switched on). The nodes at this edge are called the input layer. This causes certain nodes throughout the system to become active. The input stimulation will initiate activation in some of the nodes to which the input nodes are connected. A wave of stimulations will pass into the network. As a node becomes active its connections become live. Depending on what other stimulations it is getting, a node thus in receipt of stimulation will possibly be activated. If so it will pass on a stimulus to all the nodes it is connected to. But one of the nodes it is connected to will be the node that helped stimulate it. So the nodes may twinkle. Sometimes what is found experimentally is that after a time the pattern of activity across the network settles down and you get a steady state. If you consider the top edge layer of the rectangular network, that layer sometimes comes to show a steady pattern of activity. This layer is called the output layer. The pattern of activity put into the input layer is called the problem. The pattern of activity that gets established at the output layer is called the solution. It is understood that no one knows exactly what is going on in the middle.. One particular connectionist set-up: NETtalk.Written syllables ('sh') were coded in at the input end. The 'solution' was compared with a set of arbitrary codes for sounds. (Sss )When a steady state was achieved with a certain input, a sound would be emitted, depending on what the output was. At first (of course) there was no relation between input sylable and output sound. All the sounds would be 'wrong'. But each time the sound was not right, the weightings of the connections between the nodes in the middle of the net were altered. And if this produced a closer result, they were altered in the same direction some more: until the right sound was produced. The system gradually learnt to speak, (or made one or two elementary steps in this direction...). END Review Question What the point of a net? Why not just use a von Neumann machine and programme it appropriately? Saith the wise
Tester If you have to teach the net the answer, aren't you in effect programming it? A response. Poser Experiments with connectionist systems are mostly done by simulating them on von Neumann machines. What are the implications of this? A response.
STAY ACTIVE: Have another go at writing down from memory the headings of the key points. Heads
|