Recurrent Deep Differentiable Logic Gate Networks

ICLR 2026 Conference Submission16553 Authors

Keywords: Recurrent Neural Networks, Differentiable Logic Gates, Machine Translation, Sequence-to-Sequence Learning, Hardware Acceleration, Efficient Deep Learning, FPGA, Edge device

Abstract:

While differentiable logic gates have shown promise in feedforward networks, their application to sequential modeling remains unexplored.
This paper presents the first implementation of Recurrent Deep Differentiable Logic Gate Networks (RDDLGN), combining Boolean operations with recurrent architectures for sequence-to-sequence learning.
Evaluated on WMT'14 English-German translation, RDDLGN achieves 5.00 BLEU and 30.9\% accuracy during training, approaching GRU performance (5.41 BLEU) and graceful degradation (4.39 BLEU) during inference.
This work establishes recurrent logic-based neural computation as viable, opening research directions for FPGA acceleration in sequential modeling and other recursive network architectures.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 16553