Global information is essential for dense prediction problems, whose goal is to compute a discrete or continuous label for each pixel in the images. Traditional convolutional layers in neural networks, initially designed for image classification, are restrictive in these problems since the filter size limits their receptive fields. In this work, we propose to replace any traditional convolutional layer with an autoregressive moving-average (ARMA) layer, a novel module with an adjustable receptive field controlled by the learnable autoregressive coefficients. Compared with traditional convolutional layers, our ARMA layer enables explicit interconnections of the output neurons and learns its receptive field by adapting the autoregressive coefficients of the interconnections. ARMA layer is adjustable to different types of tasks: for tasks where global information is crucial, it is capable of learning relatively large autoregressive coefficients to allow for an output neuron's receptive field covering the entire input; for tasks where only local information is required, it can learn small or near zero autoregressive coefficients and automatically reduces to a traditional convolutional layer. We show both theoretically and empirically that the effective receptive field of networks with ARMA layers (named ARMA networks) expands with larger autoregressive coefficients. We also provably solve the instability problem of learning and prediction in the ARMA layer through a re-parameterization mechanism. Additionally, we demonstrate that ARMA networks substantially improve their baselines on challenging dense prediction tasks, including video prediction and semantic segmentation.
Jiahao Su (University of Maryland)
Shiqi Wang (Nanjing University )
Furong Huang (University of Maryland)
Furong Huang is an assistant professor of computer science. Huang’s research focuses on machine learning, high-dimensional statistics and distributed algorithms—both the theoretical analysis and practical implementation of parallel spectral methods for latent variable graphical models. Some applications of her research include developing fast detection algorithms to discover hidden and overlapping user communities in social networks, learning convolutional sparse coding models for understanding semantic meanings of sentences and object recognition in images, healthcare analytics by learning a hierarchy on human diseases for guiding doctors to identify potential diseases afflicting patients, and more. Huang recently completed a postdoctoral position at Microsoft Research in New York.
More from the Same Authors
2020 Poster: Convolutional Tensor-Train LSTM for Spatio-Temporal Learning »
Jiahao Su · Wonmin Byeon · Jean Kossaifi · Furong Huang · Jan Kautz · Anima Anandkumar
2012 Poster: Learning Mixtures of Tree Graphical Models »
Anima Anandkumar · Daniel Hsu · Furong Huang · Sham M Kakade