Two papers accepted in NeurIPS 2021
Noisy Recurrent Neural Networks ( preprint)
- Abstract: We provide a general framework for studying recurrent neural networks (RNNs) trained by injecting noise into hidden states. Specifically, we consider RNNs that can be viewed as discretizations of stochastic differential equations driven by input data. This framework allows us to study the implicit regularization effect of general noise injection schemes by deriving an approximate explicit regularizer in the small noise regime. We find that, under reasonable assumptions, this implicit regularization promotes flatter minima; it biases towards models with more stable dynamics; and, in classification tasks, it favors models with larger classification margin. Sufficient conditions for global stability are obtained, highlighting the phenomenon of stochastic stabilization, where noise injection can improve stability during training. Our theory is supported by empirical results which demonstrate improved robustness with respect to various input perturbations, while maintaining state-of-the-art performance
Compressing Deep ODE-Nets using Basis Function Expansions ( preprint)
- Abstract: The recently-introduced class of ordinary differential equation networks (ODE-Nets) establishes a fruitful connection between deep learning and dynamical systems. In this work, we reconsider formulations of the weights as continuous-depth functions using linear combinations of basis functions. This perspective allows us to compress the weights through a change of basis, without retraining, while maintaining near state-of-the-art performance. In turn, both inference time and the memory footprint are reduced, enabling quick and rigorous adaptation between computational environments. Furthermore, our framework enables meaningful continuous-in-time batch normalization layers using function projections. The performance of basis function compression is demonstrated by applying continuous-depth models to (a) image classification tasks using convolutional units and (b) sentence-tagging tasks using transformer encoder units.