Learning Sparse Structured Ensembles with SG-MCMC and Network Pruning
This work addresses the problem of computational efficiency for practitioners using neural network ensembles, representing an incremental improvement by integrating existing techniques.
The paper tackles the high computational cost of neural network ensembles by proposing a two-stage method combining SG-MCMC with group sparse priors and network pruning to create Sparse Structured Ensembles (SSEs), achieving a 21% relative reduction in perplexity for LSTM language modeling with only 30% of parameters and 70% of computations compared to a baseline.
An ensemble of neural networks is known to be more robust and accurate than an individual network, however usually with linearly-increased cost in both training and testing. In this work, we propose a two-stage method to learn Sparse Structured Ensembles (SSEs) for neural networks. In the first stage, we run SG-MCMC with group sparse priors to draw an ensemble of samples from the posterior distribution of network parameters. In the second stage, we apply weight-pruning to each sampled network and then perform retraining over the remained connections. In this way of learning SSEs with SG-MCMC and pruning, we not only achieve high prediction accuracy since SG-MCMC enhances exploration of the model-parameter space, but also reduce memory and computation cost significantly in both training and testing of NN ensembles. This is thoroughly evaluated in the experiments of learning SSE ensembles of both FNNs and LSTMs. For example, in LSTM based language modeling (LM), we obtain 21% relative reduction in LM perplexity by learning a SSE of 4 large LSTM models, which has only 30% of model parameters and 70% of computations in total, as compared to the baseline large LSTM LM. To the best of our knowledge, this work represents the first methodology and empirical study of integrating SG-MCMC, group sparse prior and network pruning together for learning NN ensembles.