Predicting Parameters in Deep Learning
This reduces computational and memory costs for training deep learning models, though it appears incremental as it builds on existing parameter efficiency methods.
The paper tackles the problem of parameter redundancy in deep learning models by demonstrating that most weights can be accurately predicted from a few values, achieving over 95% prediction without accuracy loss.
We demonstrate that there is significant redundancy in the parameterization of several deep learning models. Given only a few weight values for each feature it is possible to accurately predict the remaining values. Moreover, we show that not only can the parameter values be predicted, but many of them need not be learned at all. We train several different architectures by learning only a small number of weights and predicting the rest. In the best case we are able to predict more than 95% of the weights of a network without any drop in accuracy.