Artificial Neural Networks

1. 1That the energy cannot increase under the deterministic Hopfield dynamics is a consequence of the fact that the diagonal weights are set to zero.

   1. AVerdadero
   2. BFalso
2. 2The detailed balance condition is a necessary condition for the Markov-Chain Monte-Carlo algorithm to converge.

   1. AVerdadero
   2. BFalso
3. 3In the limit of alpha →0 the one-step error probability for mixed states in the deterministic Hopfield network vanishes.

   1. AVerdadero
   2. BFalso
4. 4One feeds a distorted pattern into the network by setting the neuron states equal to the bits in the pattern.

   1. AVerdadero
   2. BFalso
5. 5For a given alpha, the one-step error probability for the deterministic Hopfield network is higher when the diagonal weights are set to zero.

   1. AVerdadero
   2. BFalso
6. 6In the limit of N→∞ the order parameter m_μ​​ can have at most one component of order unity, the other components are small.

   1. AVerdadero
   2. BFalso
7. 7Nesterov`s accelerated gradient scheme is often more efficient than the simple momentum scheme because the weighting factor of the momentum term increases as a function of iteration number.

   1. AVerdadero
   2. BFalso
8. 8A perceptron that solves the parity problem with N inputs contains at least 2^N​ hidden neurons.

   1. AVerdadero
   2. BFalso
9. 9Weight decay helps against overfitting.

   1. AVerdadero
   2. BFalso
10. 10Different layers of a deep network learn at different speeds because their effects on the output are different.

    1. AVerdadero
    2. BFalso

11. 11L_​1​​ -regularisation reduces small weights more than L_2​-regularisation.

    1. AVerdadero
    2. BFalso

Descripción

Artificial Neural Networks Test sobre Artificial Neural Networks, creado por David Larsson el 25/10/2019.

Autor

David Larsson

Idioma

English

Visitas

3

Etiquetas

artificial neural networks

Valoración

Aún sin valoraciones

Valora este test

Valora este test

Más sobre Neural Networks and Deep Learning