A 65 nm Bayesian Neural Network Accelerator with 360 fJ/Sample In-Word GRNG for AI Uncertainty Estimation

2025-01-08 · Zephan M. Enciso, Boyang Cheng, Likai Pei, Jianbo Liu, Steven Davis, Michael Niemier, Ningyuan Cao

Uncertainty estimation is an indispensable capability for AI-enabled, safety-critical applications, e.g. autonomous vehicles or medical diagnosis. Bayesian neural networks (BNNs) use Bayesian statistics to provide both classification predictions and uncertainty estimation, but they suffer from high computational overhead associated with random number generation and repeated sample iterations. Furthermore, BNNs are not immediately amenable to acceleration through compute-in-memory architectures due to the frequent memory writes necessary after each RNG operation. To address these challenges, we present an ASIC that integrates 360 fJ/Sample Gaussian RNG directly into the SRAM memory words. This integration reduces RNG overhead and enables fully-parallel compute-in-memory operations for BNNs. The prototype chip achieves 5.12 GSa/s RNG throughput and 102 GOp/s neural network throughput while occupying 0.45 mm2, bringing AI uncertainty estimation to edge computation.

📄 PDF Abstract BibTeX arXiv:2501.04577

Code (0)

등록된 구현이 없습니다.

Tasks

Autonomous Vehicles Medical Diagnosis

VIBNN: Hardware Acceleration of Bayesian Neural Networks

2018-02-02 · Ruizhe Cai, Ao Ren, Ning Liu, Caiwen Ding 외

Bayesian Neural Networks (BNNs) have been proposed to address the problem of model uncertainty in training and inference. By introducing weights associated with conditioned probability distributions, BNNs are capable of …

Small Data Image Classification Variational Inference

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

2026-06-15 · Jonas Crols, Guilherme Paim, Shirui Zhao, Marian Verhelst arxiv

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 a…

Probabilistic Tree Inference Enabled by FDSOI Ferroelectric FETs

2026-04-06 · Pengyu Ren, Xingtian Wang, Boyang Cheng, Jiahui Duan 외 arxiv

Artificial intelligence applications in autonomous driving, medical diagnostics, and financial systems increasingly demand machine learning models that can provide robust uncertainty quantification, interpretability, and…

Autonomous Driving

Gradient Regularized Natural Gradients

2026-01-26 · Satya Prakash Dash, Hossein Abdi, Wei Pan, Samuel Kaski 외 arxiv

Gradient regularization (GR) has been shown to improve the generalizability of trained models. While Natural Gradient Descent has been shown to accelerate optimization in the initial phase of training, little attention h…

Model-free and Bayesian Ensembling Model-based Deep Reinforcement Learning for Particle Accelerator Control Demonstrated on the FERMI FEL

2020-12-17 · Simon Hirlaender, Niky Bruchon

Reinforcement learning holds tremendous promise in accelerator controls. The primary goal of this paper is to show how this approach can be utilised on an operational level on accelerator physics problems. Despite the su…

Deep Reinforcement Learning model Policy Gradient Methods Q-Learning+3