Artificial Neural Network - Hopfield Networks

Hopfield neural network was invented by Dr. John J. Hopfield in 1982. It consists of a single layer which contains one or more fully connected recurrent neurons. The Hopfield network is commonly used for auto-association and optimization tasks.

Discrete Hopfield Network

A Hopfield network which operates in a discrete line fashion or in other words, it can be said the input and output patterns are discrete vector, which can be either binary (0,1) or bipolar (+1, -1) in nature. The network has symmetrical weights with no self-connections i.e., w_ij = w_ji and w_ii = 0.

Architecture

Following are some important points to keep in mind about discrete Hopfield network −

· This model consists of neurons with one inverting and one non-inverting output.

· The output of each neuron should be the input of other neurons but not the input of self.

· Weight/connection strength is represented by w_ij.

· Connections can be excitatory as well as inhibitory. It would be excitatory, if the output of the neuron is same as the input, otherwise inhibitory.

· Weights should be symmetrical, i.e. w_ij = w_ji

The output from Y₁ going to Y₂, Y_i and Y_n have the weights w₁₂, w_1i and w_1nrespectively. Similarly, other arcs have the weights on them.

Training Algorithm

During training of discrete Hopfield network, weights will be updated. As we know that we can have the binary input vectors as well as bipolar input vectors. Hence, in both the cases, weight updates can be done with the following relation

Case 1 − Binary input patterns

For a set of binary patterns s(p), p = 1 to P

Here, s(p) = s₁(p), s₂(p),..., s_i(p),..., s_n(p)

Weight Matrix is given by

wij=∑p=1P[2si(p)−1][2sj(p)−1]fori≠jwij=∑p=1P[2si(p)−1][2sj(p)−1]fori≠j

Case 2 − Bipolar input patterns

For a set of binary patterns s(p), p = 1 to P

Here, s(p) = s₁(p), s₂(p),..., s_i(p),..., s_n(p)

Weight Matrix is given by

wij=∑p=1P[si(p)][sj(p)]fori≠jwij=∑p=1P[si(p)][sj(p)]fori≠j

Testing Algorithm

Step 1 − Initialize the weights, which are obtained from training algorithm by using Hebbian principle.

Step 2 − Perform steps 3-9, if the activations of the network is not consolidated.

Step 3 − For each input vector X, perform steps 4-8.

Step 4 − Make initial activation of the network equal to the external input vector X as follows −

yi=xifori=1tonyi=xifori=1ton

Step 5 − For each unit Y_i, perform steps 6-9.

Step 6 − Calculate the net input of the network as follows −

yini=xi+∑jyjwjiyini=xi+∑jyjwji

Step 7 − Apply the activation as follows over the net input to calculate the output −

yi=⎧⎩⎨1yi0ifyini>θiifyini=θiifyini<θiyi={1ifyini>θiyiifyini=θi0ifyini<θi

Here θiθi is the threshold.

Step 8 − Broadcast this output y_i to all other units.

Step 9 − Test the network for conjunction.

Energy Function Evaluation

An energy function is defined as a function that is bonded and non-increasing function of the state of the system.

Energy function E_f⁡, ⁡also called Lyapunov function determines the stability of discrete Hopfield network, and is characterized as follows −

Ef=−12∑i=1n∑j=1nyiyjwij−∑i=1nxiyi+∑i=1nθiyiEf=−12∑i=1n∑j=1nyiyjwij−∑i=1nxiyi+∑i=1nθiyi

Condition − In a stable network, whenever the state of node changes, the above energy function will decrease.

Suppose when node i has changed state from y(k)iyi(k) to y(k+1)iyi(k+1) ⁡then the Energy change ΔEfΔEf is given by the following relation

ΔEf=Ef(y(k+1)i)−Ef(y(k)i)ΔEf=Ef(yi(k+1))−Ef(yi(k))

=−(∑j=1nwijy(k)i+xi−θi)(y(k+1)i−y(k)i)=−(∑j=1nwijyi(k)+xi−θi)(yi(k+1)−yi(k))

=−(neti)Δyi=−(neti)Δyi

Here Δyi=y(k+1)i−y(k)iΔyi=yi(k+1)−yi(k)

The change in energy depends on the fact that only one unit can update its activation at a time.

Continuous Hopfield Network

In comparison with Discrete Hopfield network, continuous network has time as a continuous variable. It is also used in auto association and optimization problems such as travelling salesman problem.

Model − The model or architecture can be build up by adding electrical components such as amplifiers which can map the input voltage to the output voltage over a sigmoid activation function.

Energy Function Evaluation

Ef=12∑i=1n∑j=1j≠inyiyjwij−∑i=1nxiyi+1λ∑i=1n∑j=1j≠inwijgri∫yi0a−1(y)dyEf=12∑i=1n∑j=1j≠inyiyjwij−∑i=1nxiyi+1λ∑i=1n∑j=1j≠inwijgri∫0yia−1(y)dy

Here λ is gain parameter and g_ri input conductance.