ACM
Research
AiTPO: KAN-UNet Heterogeneous Network for Timing Prediction and Optimization at Global Routing
Abstract
AiTPO presents a novel framework utilizing a heterogeneous KAN-UNet network architecture designed for highly accurate timing prediction and optimization during the critical global routing phase of IC design. By integrating the localized
Research
Advanced Side-Channel Evaluation Using Contextual Deep Learning-Based Leakage Modeling
Abstract
This work introduces an advanced methodology for side-channel evaluation by leveraging contextual deep learning techniques for leakage modeling. The approach utilizes neural networks to automatically learn and analyze complex, high-dimensional side-channel traces,
Research
DeepVerifier: Learning to Update Test Sequences for Coverage-Guided Verification
Abstract
DeepVerifier introduces a novel machine learning framework designed to optimize hardware verification efficiency by intelligently updating existing test sequences for coverage closure. The system utilizes deep reinforcement learning (DRL) to guide the
Research
A Novel Approach to Reducing Testing Costs and Minimizing Defect Escapes Using Dynamic Neighborhood Range and Shapley Values
Abstract
This article presents a novel verification methodology aimed at substantially reducing hardware testing costs while simultaneously minimizing defect escapes. The approach integrates Dynamic Neighborhood Range (DNR) algorithms for efficient adaptive test generation
Research
Mixed-Level Modeling and Evaluation of a Cache-less Grid of Processing Cells
Abstract
This article introduces a novel architecture based on a cache-less grid composed of numerous processing cells, designed to maximize parallelism and efficiency for embedded applications. The core innovation lies in utilizing a
Research
TreeHouse: An MLIR-based Compilation Flow for Real-Time Tree-based Inference
Abstract
TreeHouse presents a novel MLIR-based compilation flow specifically optimized for achieving real-time, low-latency inference of tree-based machine learning models (like Random Forests and GBTs). By leveraging MLIR's retargetability and modular