On Multiplicative Integration with Recurrent Neural NetworksOn Multiplicative Integration with Recurrent Neural NetworksWu, Yuhuai and Zhang, Saizheng and Zhang, Ying and Bengio, Yoshua and Salakhutdinov, Ruslan2016
Paper summarynipsreviewsThis paper has a simple premise: that the, say, LSTM cell works better with multiplicative updates (equation 2) rather than additive ones (equation 1). This additive update is used in various places in lieu of additive ones, in various places in the LSTM recurrence equations (the exact formulation is in the supplementary material). A slightly hand wavy argument is made in favour of the multiplicative update, on the grounds of superior gradient flow (section 2.2). Mainly however, the authors make a rather thorough empirical investigation which shows remarkably good performance of their new architectures, on a range of real problems. Figure 1(a) is nice, showing an apparent greater information flow (as defined by a particular gradient) through time for the new scheme, as well as faster convergence and less saturated hidden unit activations. Overall, the experimental results appear thorough and convincing, although I am not a specialist in this area.
This model presents a multiplicative alternative (with an additive component) to the additive update which happens at the core of various RNNs (Simple RNNs, GRUs, LSTMs). The multiplicative component, without introducing a significant change in the number of parameters, yields better gradient passing properties which enable the learning of better models, as shown in experiments.
This paper has a simple premise: that the, say, LSTM cell works better with multiplicative updates (equation 2) rather than additive ones (equation 1). This additive update is used in various places in lieu of additive ones, in various places in the LSTM recurrence equations (the exact formulation is in the supplementary material). A slightly hand wavy argument is made in favour of the multiplicative update, on the grounds of superior gradient flow (section 2.2). Mainly however, the authors make a rather thorough empirical investigation which shows remarkably good performance of their new architectures, on a range of real problems. Figure 1(a) is nice, showing an apparent greater information flow (as defined by a particular gradient) through time for the new scheme, as well as faster convergence and less saturated hidden unit activations. Overall, the experimental results appear thorough and convincing, although I am not a specialist in this area.
This model presents a multiplicative alternative (with an additive component) to the additive update which happens at the core of various RNNs (Simple RNNs, GRUs, LSTMs). The multiplicative component, without introducing a significant change in the number of parameters, yields better gradient passing properties which enable the learning of better models, as shown in experiments.