Q. Asynchronous update ensures that the next state is atmost unit hamming distance from current state, is that true?

View solution

Q. If pattern is to be stored, then what does stable state should have updated value of?

View solution

Q. For symmetric weights there exist?

View solution

Q. If connections are not symmetric then basins of attraction may correspond to?

View solution

Q. For analysis of storage capacity what are the conditions imposed on hopfield model?

View solution

Q. What is gradient descent?

View solution

Q. If connections are not symmetric then basins of attraction may correspond to oscillatory or stable regions, is that true?

View solution

Q. In hopfield network with symmetric weights, energy at each state may?

View solution

Q. In hopfield model with symmetric weights, network can move to?

View solution

Q. Can error in recall due to false minima be reduced?

View solution

Q. How can error in recall due to false minima be reduced?

View solution

Q. Energy at each state in hopfield with symmetric weights network may increase or decrease?

View solution

Q. Pattern storage problem which cannot be represented by a feedback network of given size can be called as?

View solution

Q. What is the other way to reduce error in recall due to false minima apart from stochastic update?

View solution

Q. How can error in recall due to false minima be further reduced?

View solution

Q. As temperature increase, what happens to stochastic update?

View solution

Q. Why does change in temperature doesn’t effect stochastic update?

View solution

Q. p(s=1|x) = 1/(1+exp(-x/T))) ,where ‘s’ is the output given the activation ‘x’ is a?

View solution

Q. Does a stochastic network will evolve differently each time it is run?

View solution

Q. In case of deterministic update, what kind of equilibrium is reached?

View solution