Deep Differential Recurrent Neural Networks

TL;DR

The paper introduces deep differential recurrent neural networks (d2RNN) that enhance LSTM's ability to detect salient spatio-temporal dynamics in sequential data by stacking multiple layers with derivative-based gating.

Contribution

It proposes a novel deep LSTM architecture with individual derivative of states gating, improving detection of complex dynamical patterns in sequential data.

Findings

Outperforms LSTM and other state-of-the-art methods on human activity datasets.

Effectively models higher-order spatio-temporal dynamics.

Demonstrates improved accuracy in activity recognition tasks.

Abstract

Due to the special gating schemes of Long Short-Term Memory (LSTM), LSTMs have shown greater potential to process complex sequential information than the traditional Recurrent Neural Network (RNN). The conventional LSTM, however, fails to take into consideration the impact of salient spatio-temporal dynamics present in the sequential input data. This problem was first addressed by the differential Recurrent Neural Network (dRNN), which uses a differential gating scheme known as Derivative of States (DoS). DoS uses higher orders of internal state derivatives to analyze the change in information gain caused by the salient motions between the successive frames. The weighted combination of several orders of DoS is then used to modulate the gates in dRNN. While each individual order of DoS is good at modeling a certain level of salient spatio-temporal sequences, the sum of all the orders of DoS could distort the detected motion patterns. To address this problem, we propose to control the LSTM gates via individual orders of DoS and stack multiple levels of LSTM cells in an increasing order of state derivatives. The proposed model progressively builds up the ability of the LSTM gates to detect salient dynamical patterns in deeper stacked layers modeling higher orders of DoS, and thus the proposed LSTM model is termed deep differential Recurrent Neural Network (d2RNN). The effectiveness of the proposed model is demonstrated on two publicly available human activity datasets: NUS-HGA and Violent-Flows. The proposed model outperforms both LSTM and non-LSTM based state-of-the-art algorithms.