Bitformer: An efficient Transformer with bitwise operation-based attention for Big Data Analytics at low-cost low-precision devices

TL;DR

Bitformer introduces a novel Transformer variant utilizing bitwise operations in its attention mechanism, significantly reducing computational complexity and enabling efficient deployment on low-resource edge devices without sacrificing performance.

Contribution

The paper presents a new attention mechanism based on bitwise operations that replaces floating-point matrix multiplication, reducing complexity and making Transformers suitable for low-power, low-precision devices.

Findings

Maintains ability to model long-range dependencies.

Reduces attention computation complexity from O(n^2d) to O(n^2T).

Enables efficient edge computing deployment.

Abstract

In the current landscape of large models, the Transformer stands as a cornerstone, playing a pivotal role in shaping the trajectory of modern models. However, its application encounters challenges attributed to the substantial computational intricacies intrinsic to its attention mechanism. Moreover, its reliance on high-precision floating-point operations presents specific hurdles, particularly evident in computation-intensive scenarios such as edge computing environments. These environments, characterized by resource-constrained devices and a preference for lower precision, necessitate innovative solutions. To tackle the exacting data processing demands posed by edge devices, we introduce the Bitformer model, an inventive extension of the Transformer paradigm. Central to this innovation is a novel attention mechanism that adeptly replaces conventional floating-point matrix multiplication with bitwise operations. This strategic substitution yields dual advantages. Not only does it maintain the attention mechanism's prowess in capturing intricate long-range information dependencies, but it also orchestrates a profound reduction in the computational complexity inherent in the attention operation. The transition from an $O(n^2d)$ complexity, typical of floating-point operations, to an $O(n^2T)$ complexity characterizing bitwise operations, substantiates this advantage. Notably, in this context, the parameter $T$ remains markedly smaller than the conventional dimensionality parameter $d$. The Bitformer model in essence endeavors to reconcile the indomitable requirements of modern computing landscapes with the constraints posed by edge computing scenarios. By forging this innovative path, we bridge the gap between high-performing models and resource-scarce environments, thus unveiling a promising trajectory for further advancements in the field.