Encoding Recurrence into TransformersDownload PDF

Published: 01 Feb 2023, Last Modified: 28 Feb 2023ICLR 2023 notable top 5%Readers: Everyone
Keywords: Recurrent models, Transformers, sample efficiency, gated mechanism
TL;DR: We propose a new module to encode the recurrent dynamics of an RNN layer into Transformers and higher sample efficiency can be achieved.
Abstract: This paper novelly breaks down with ignorable loss an RNN layer into a sequence of simple RNNs, each of which can be further rewritten into a lightweight positional encoding matrix of a self-attention, named the Recurrence Encoding Matrix (REM). Thus, recurrent dynamics introduced by the RNN layer can be encapsulated into the positional encodings of a multihead self-attention, and this makes it possible to seamlessly incorporate these recurrent dynamics into a Transformer, leading to a new module, Self-Attention with Recurrence (RSA). The proposed module can leverage the recurrent inductive bias of REMs to achieve a better sample efficiency than its corresponding baseline Transformer, while the self-attention is used to model the remaining non-recurrent signals. The relative proportions of these two components are controlled by a data-driven gated mechanism, and the effectiveness of RSA modules are demonstrated by four sequential learning tasks.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Submission Guidelines: Yes
Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning
Supplementary Material: zip
28 Replies