Adaptive Learning in Neural Networks

Gradient descent is one of the most powerful optimizing method. However, learning time is a challange, too. Standard version of gradient descent learns slowly. That’s why, some modifications are included into the gradient descent in real world applications. These approaches would be applied to converge faster. In previous post, incorporating momentum is already mentioned to work for same purpose. Now, we will focus on applying adaptive learning to learn faster.

rango
Chameleons are famous for adaptation, Rango (2011)

As you might remember, weights are updated by the following formula in back propagation.

wi = wi – α . (∂Error / ∂wi)

Alpha refers to learning rate in the formula. Applying adaptive learning rate proposes to increase / decrease alpha based on cost changes. The following code block would realize this process.

a = 0.1, b = 0.5; //adaptive learning params

if(currentCost < previousCost){

 learningRate = learningRate + a;

}
else{

 learningRate = learningRate - b * learningRate;

}

previousCost = currentCost * 1;

To sum up, learning rate should be increased if cost decreases. In contrast, alpha should be decreased if cost increases. This procedure should be applied in each gradient descent iteration.

Let’s transform the theory to the practice. If adaptive learning rate is applied on classical xor dataset, cost value decreases dramatically whereas standard gradient descent decreases cost stable. Thus, network could converge faster as illustrated below. Of course, momentum incorporation and adaptive learning rate are applied both.

adaptive-learning-rate
Applying Adaptive Learning Rate

So, we’ve focused on a method that improve gradient descent performance. The experiment shows that this approach improves performance considerably. Moreover, it is easy to implement. Only a few code lines could perform this action. Finally, adaptive learning capability is added to neural-networks repository shared on my GitHub profile.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s