期刊
IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
卷 41, 期 6, 页码 1636-1648出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCAD.2021.3114617
关键词
Codes; Kernel; Feature extraction; Computational modeling; Benchmark testing; Deep learning; Analytical models; Computation mapping; deep learning (DL); heterogeneous platform; machine learning; source code analysis
This article discusses the application of deep learning in source code analysis, introducing techniques that utilize different networks and input information. The impact on the accuracy of DL methods and how to extract effective information are also explored. By studying, it was found that normalizing auxiliary information can improve accuracy, and a new method is proposed to enhance mapping accuracy by increasing the dataset's cardinality.
Despite its relatively recent history, deep learning (DL)-based source code analysis is already a cornerstone in machine learning for compiler optimization. When applied to the classification of pieces of code to identify the best computational unit in a heterogeneous Systems-on-Chip, it can be effective in supporting decisions that a programmer has otherwise to take manually. Several techniques have been proposed exploiting different networks and input information, prominently sequence-based and graph-based representations, complemented by auxiliary information typically related to payload and device configuration. While the accuracy of DL methods strongly depends on the training and test datasets, so far no exhaustive and statistically meaningful analysis has been done on its impact on the results and on how to effectively extract the available information. This is relevant also considering the scarce availability of source code datasets that can be labeled by profiling on heterogeneous compute units. In this article, we first present such a study, which leads us to devise the contribution of code sequences and auxiliary inputs separately. Starting from this analysis, we then demonstrate that by using the normalization of auxiliary information, it is possible to improve state-of-the-art results in terms of accuracy. Finally, we propose a novel approach exploiting Siamese networks that further improve mapping accuracy by increasing the cardinality of the dataset, thus compensating for its relatively small size.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据