Adaptive Computation Time for Recurrent Neural Networks
This addresses the inefficiency of fixed-step RNNs for sequence processing, offering a deterministic and differentiable method that could infer segment boundaries in data, though it is incremental as it builds on existing RNN architectures.
The paper tackles the problem of recurrent neural networks (RNNs) needing fixed computational steps by introducing Adaptive Computation Time (ACT), which allows RNNs to dynamically adjust steps per input, resulting in dramatic performance improvements on synthetic tasks like parity determination and integer addition, though gains were limited on a language modeling dataset.
This paper introduces Adaptive Computation Time (ACT), an algorithm that allows recurrent neural networks to learn how many computational steps to take between receiving an input and emitting an output. ACT requires minimal changes to the network architecture, is deterministic and differentiable, and does not add any noise to the parameter gradients. Experimental results are provided for four synthetic problems: determining the parity of binary vectors, applying binary logic operations, adding integers, and sorting real numbers. Overall, performance is dramatically improved by the use of ACT, which successfully adapts the number of computational steps to the requirements of the problem. We also present character-level language modelling results on the Hutter prize Wikipedia dataset. In this case ACT does not yield large gains in performance; however it does provide intriguing insight into the structure of the data, with more computation allocated to harder-to-predict transitions, such as spaces between words and ends of sentences. This suggests that ACT or other adaptive computation methods could provide a generic method for inferring segment boundaries in sequence data.