Vector-based Dynamic IR-drop Prediction Using Machine Learning

Chen, Jiaxian; Liu, Shi-Tang; Wu, Yu-Tsung; Wu, Mengfan; Li, Chieo-Mo; Chang, Norman; Li, Ying-Shiun; Chuang, Wen-Tze

doi:10.1109/asp-dac52403.2022.9712489

Cited by 5 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fang et al [3] use a convolutional neural network (CNN) to reduce the dimension of neighbouring cell features and feed them together with all target cell features into the extreme gradient boosting (XGBoost) model. Density map features were proposed by [4] with instance-level prediction results from XGBoost-based models. These two decision tree based methods [3,4] give each decision depending on all features.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Attention‐based model for dynamic IR drop prediction with multi‐view features

Zhu

Liu

2023

Electronics Letters

View full text Add to dashboard Cite

Dynamic IR drop prediction based on machine learning has been studied in recent years. However, most proposed models used all input features extracted from circuits or manually selected parts of raw features as inputs, which failed to differentiate the order of priority among input features in a flexible manner. In this paper, QuantumForest to vector‐based dynamic IR drop prediction is introduced. With the sparse attention mechanism brought by QuantumForest, important attributes of circuits are weighed more heavily than others. A new multi‐view feature creation method is also proposed and a novel regional distance feature is built up subsequently. The performance is evaluated on two chip designs with real simulation vectors. The experiment results indicate that the prediction result of the method outperforms other prominent methods for dealing with machine learning based IR drop analysis, reaching an average MAE of only 1.457 mV$\text{mV}$ on two designs.

show abstract

mentioning

confidence: 99%

“…Density map features were proposed by [4] with instance-level prediction results from XGBoost-based models. These two decision tree based methods [3,4] give each decision depending on all features. A maximum CNN architecture is proposed by [5].…”

mentioning

confidence: 99%