Modelling Brain and Mind


Вопрос 1 of 1

  Осталось Времени


Match word to definition

Сопоставьте текст (нажмите и перетащите)   


Совпадение текста


Algorithmic complexity

Logical depth


Fractal dimension

Synthetic psychology

Braitenberg vehicle

Artificial neural network


Gradient descent

Global minimum

Local minimum


Supervised learning

The Rescorla-Wagner model

The Delta-rule


Decision boundary

Linear separability

Exclusive-OR problem (XOR problem)




Нажать и перетащить

A method for finding the valleys in a function (the lowest point, where the error is lowest)

A motor whose output value is calculated by multiplying each wire by the corresponding sensor reading and adding them together

The behaviour of a circuit that controls its output to reduce the differences between its output and target ouput

Solution corresponds to drawing a straight line in a graph, that separates one category of ouputs from another

The number of iterations of the shortest algorithm required to generate a sequence.

A series of instructions that should be followed

Synthesizing the behabiourof complex systems like the brain and mind by exploring how simple models of biological processes interact with an environment

A secondary valley with a higher error where the algorithm will stop before the global minimum

A ratio comparing how detail ina pattern (a fractal pattern) changes with the scale at which it is measured

This measures the length of the shortest computer program required to generate a particular sequence of symbols.

A property of complex systems, such as a recurrent neural network. It's a set of states that a system can adopt where one state always leadds to a second state, to a third state etc.

A straight line in the input space, separated positive outputs from negative outputs of a neuron

Comes from neural network research and was discovered as a way of modifying the synaptic weights of an artificial neuron

A branch of maths concerned with finding and using the derivative of a function

Spontaneous emergence of global order from a system of individuals interacting according to simple rules in the absence of a supervisor

The value of the parameter that generates the lowest possible error

The order in which associations are learn between stimulus (cue) and a reinforcer (reward/punishment) matters

Presenting a network with examples of input patterns along with information about what the correct outputs should be in response to those inputs

A supervised learning algorithm for updating the synaptic weights in a neural network

Discovered as a way of describing how associative strengths should change during a classical conditioning task

How difficult it is to deduce the shortest possbile algorithm required to generate a sequence.

A set of artificial neurons whose outputs are connected to the inputs of others

Should generate an output of 1 when either the first or the second input is 1 and it should generate an output of 0 when both inputs are 1 or 0