Skip to yearly menu bar Skip to main content


Nonlinear Sufficient Dimension Reduction with a Stochastic Neural Network

SIQI LIANG · Yan Sun · Faming Liang

Hall J (level 1) #530

Keywords: [ Deep Learning ] [ Stochastic neural network ] [ Adaptive Stochastic Gradient MCMC ] [ Big Data ] [ Sufficient Dimension Reduction ]


Sufficient dimension reduction is a powerful tool to extract core information hidden in the high-dimensional data and has potentially many important applications in machine learning tasks. However, the existing nonlinear sufficient dimension reduction methods often lack the scalability necessary for dealing with large-scale data. We propose a new type of stochastic neural network under a rigorous probabilistic framework and show that it can be used for sufficient dimension reduction for large-scale data. The proposed stochastic neural network is trained using an adaptive stochastic gradient Markov chain Monte Carlo algorithm, whose convergence is rigorously studied in the paper as well. Through extensive experiments on real-world classification and regression problems, we show that the proposed method compares favorably with the existing state-of-the-art sufficient dimension reduction methods and is computationally more efficient for large-scale data.

Chat is not available.