Gated Recurrent Neural Networks with Weighted Time-Delay Feedback
This addresses a known bottleneck in recurrent neural networks for handling long-term dependencies, offering an incremental improvement over existing gated architectures.
The paper tackled the problem of modeling long-term dependencies in sequential data by introducing a gated recurrent unit with a weighted time-delay feedback mechanism, resulting in outperforming state-of-the-art recurrent units on various tasks with faster convergence and better generalization.
In this paper, we present a novel approach to modeling long-term dependencies in sequential data by introducing a gated recurrent unit (GRU) with a weighted time-delay feedback mechanism. Our proposed model, named $τ$-GRU, is a discretized version of a continuous-time formulation of a recurrent unit, where the dynamics are governed by delay differential equations (DDEs). We prove the existence and uniqueness of solutions for the continuous-time model and show that the proposed feedback mechanism can significantly improve the modeling of long-term dependencies. Our empirical results indicate that $τ$-GRU outperforms state-of-the-art recurrent units and gated recurrent architectures on a range of tasks, achieving faster convergence and better generalization.