Optimal Connectivity through Network Gradients for the Restricted Boltzmann Machine
share
Leveraging sparse networks to connect successive layers in deep neural networks has recently been shown to provide benefits to large scale state-of-the-art models. However, network connectivity also plays a significant role on the learning curves of shallow networks, such as the classic Restricted Boltzmann Machines (RBM). A fundamental problem is efficiently finding connectivity patterns that improve the learning curve. Recent principled approaches explicitly include network connections as parameters that must be optimized in the model, but often rely on continuous functions to represent connections and on explicit penalization. This work presents a method to find optimal connectivity patterns for RBMs based on the idea of network gradients: computing the gradient of every possible connection, given a specific connection pattern, and using the gradient to drive a continuous connection strength parameter that in turn is used to determine the connection pattern. Thus, learning RBM parameters and learning network connections is truly jointly performed, albeit with different learning rates, and without changes to the objective function. The method is applied to the MNIST data set showing that better RBM models are found for the benchmark tasks of sample generation and input classification.
READ FULL TEXT