TreeGRNG: Binary Tree Gaussian Random Number Generator for Efficient Probabilistic AI Hardware

2026-06-15

AI Summary

Deploying Bayesian neural networks on edge devices is hindered by the high energy consumption and area overhead of Gaussian random number generators (GRNGs). This work proposes TreeGRNG, the first GRNG architecture based on a binary tree structure that enables flexible and tunable Gaussian sampling. By replacing complex arithmetic units with an array of constant comparators and incorporating hardware-aware optimizations, TreeGRNG significantly reduces resource usage while maintaining or even improving distributional accuracy. Compared to the state-of-the-art, the proposed approach achieves a 3.7× reduction in energy per sample and a 5.8× increase in throughput per unit area, while supporting a broader range of adjustable distribution shapes.

Abstract

Bayesian Neural Networks (BNNs) offer opportunities for greatly enhancing the trustworthiness of conventional neural networks by monitoring the uncertainties in decision-making. A significant drawback for BNN inference at the extreme edge, however, is the imperative need to incorporate Gaussian Random Number Generators (GRNG) within each neuron. State-of-the-art GRNG algorithms heavily depend on multiple arithmetic operations and the use of extensive look-up tables, posing significant implementation challenges for ultra-low power hardware implementations. To overcome this, this paper presents an innovative binary tree random number generator (TreeGRNG) allowing the use of ultra-low-cost constant comparators instead of arithmetic units. We further enhance the TreeGRNG proposal with a set of hardware-aware optimizations exploiting the Gaussian properties. The optimized TreeGRNG surpasses the State-of-the-Art (SoTA) in terms of distribution accuracy while achieving a 3.7$\times$ reduction in energy per sample and boosting the throughput per unit area by 5.8$\times$. Moreover, our TreeGRNG proposal possesses a distinct advantage over the current SoTA in terms of flexibility, as it easily enables designers to adjust the shape of the sampled probability distribution, extending beyond the capabilities of traditional GRNGs, opening the horizon towards future probabilistic AI designs. The TreeGRNG design is available open-source in the link

Problem

• Bayesian Neural Networks
• Gaussian Random Number Generator
• Edge AI
• Ultra-low Power Hardware
• Probabilistic AI

Innovation

• TreeGRNG
• Gaussian Random Number Generator
• Bayesian Neural Networks
• Probabilistic AI Hardware
• Energy-Efficient Design