DANA: Dimension-Adaptive Neural Architecture for Multivariate Sensor Data

Motion sensors embedded in wearable and mobile devices allow for dynamic selection of sensor streams and sampling rates, enabling useful applications, e.g. for power management or control of data sharing. While deep neural networks (DNNs) achieve competitive accuracy in sensor data classification, current DNN architectures only process data coming from a fixed set of sensors with a fixed sampling rate, and changes in the dimensions of their inputs cause considerable accuracy loss, unnecessary computations, or failure in operation. To address this problem, we introduce a dimension-adaptive pooling (DAP) layer that makes DNNs robust to temporal changes in sampling rate and in sensor availability. DAP operates on convolutional filter maps of variable dimensions and produces an input of fixed dimensions suitable for feedforward and recurrent layers. Building on this architectural improvement, we propose a dimension-adaptive training (DAT) procedure to generalize over the entire space of feasible data dimensions at the inference time. DAT comprises the random selection of dimensions during the forward passes and optimization with accumulated gradients of several backward passes. We then combine DAP and DAT to transform existing non-adaptive DNNs into a Dimension-Adaptive Neural Architecture (DANA) without altering other architectural aspects. Our solution does not need up-sampling or imputation, thus reduces unnecessary computations at inference time. Experimental results, on four benchmark datasets of human activity recognition, show that DANA prevents losses in classification accuracy of the state-of-the-art DNNs, under dynamic sensor availability and varying sampling rates.