TY - CONF
AB - We introduce a new class of time-continuous recurrent neural network models. Instead of declaring a learning systemâ€™s dynamics by implicit nonlinearities, we construct networks of linear first-order dynamical systems modulated via nonlinear interlinked gates. The resulting models represent dynamical systems with varying (i.e., liquid) time-constants coupled to their hidden state, with outputs being computed by numerical differential equation solvers. These neural networks exhibit stable and bounded behavior, yield superior expressivity within the family of neural ordinary differential equations, and give rise to improved performance on time-series prediction tasks. To demonstrate these properties, we first take a theoretical approach to find bounds over their dynamics, and compute their expressive power by the trajectory length measure in a latent trajectory space. We then conduct a series of time-series prediction experiments to manifest the approximation capability of Liquid Time-Constant Networks (LTCs) compared to classical and modern RNNs.
AU - Hasani, Ramin
AU - Lechner, Mathias
AU - Amini, Alexander
AU - Rus, Daniela
AU - Grosu, Radu
ID - 10671
IS - 9
SN - 2159-5399
T2 - Proceedings of the AAAI Conference on Artificial Intelligence
TI - Liquid time-constant networks
VL - 35
ER -