Lixin Liu

Lixin Liu

Auto-generated profile, based on publicly available information

Lixin Liu

The Chinese University of Hong Kong, Hong Kong, Hong Kong, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 December 2023 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
LLLixin Liu
Lixin Liu
BFBangqi Fu
Bangqi Fu
SLShiju Lin
Shiju Lin
JLJinwei Liu
Jinwei Liu
EYEvangeline F. Y. Young
Evangeline F. Y. Young
MWMartin D. F. Wong
Martin D. F. Wong
Placement serves as a fundamental step in VLSI physical design. Recently, GPU-based placer DREAMPlace [1] demonstrated its superiority over CPU-based placers. In this work, we develop an extremely fast GPU-accelerated placer Xplace which considers factors at operator-level optimization. Xplace achieves around 2x speedup with better-solution quality compared to DREAMPlace. We also plug a novel Fourier neural network into Xplace as an extension. Besides, we enable Xplace to handle the detailed-routability-driven placement problem and demonstrate its superiority in terms of quality and performance. We believe this work not only proposes an extremely fast and extensible placement framework but also illustrates a possibility of incorporating a neural network component into a GPU-accelerated analytical placer. The source code of Xplace is released on GitHub.

31 August 2020 in Ionics
Verified
Journal Article
Closed access
PDF file
LLLixin Liu
Lixin Liu
ZWZenghao Wang
Zenghao Wang
YXYongshuai Xie
Yongshuai Xie
YPYing Peng
Ying Peng
JDJianhong Dong
Jianhong Dong
DXDong Xu
Dong Xu

27 October 2024
Verified
Conference Paper
Open access
BFBangqi Fu
Bangqi Fu
LLLixin Liu
Lixin Liu
MWMartin D. F. Wong
Martin D. F. Wong
EYEvangeline F. Y. Young
Evangeline F. Y. Young

1 November 2021
Verified
Conference Paper
Closed access
FWFangzhou Wang
Fangzhou Wang
LLLixin Liu
Lixin Liu
JCJingsong Chen
Jingsong Chen
JLJinwei Liu
Jinwei Liu
XZXinshi Zang
Xinshi Zang
MWMartin D.F. Wong
Martin D.F. Wong
Placement and routing (P&R) are two important stages in the physical design flow. After circuit components are assigned locations by a placer, routing will take place to make the connections. Defined as two separate problems, placement and routing aim to optimize different objectives. For instance, placement usually focuses on optimizing the half-perimeter wire length (HPWL) and estimated congestion while routing will try to minimize the routed wire length and the number of overflows. The misalignment between the objectives will inevitably lead to a significant degradation in solution quality. Therefore, in this paper, we present Starfish, an efficient P&R co-optimization engine that bridges the gap between placement and routing. To incrementally optimize the routed wire length, Starfish conducts cell movements and reconnects broken nets by A*-based partial rerouting. Experimental results on the ICCAD 2020 contest benchmark suites [1] show that our co-optimizer outperforms all the contestants with better solution quality and much shorter runtime.

27 January 2022 in PLOS ONE
Verified
Journal Article
Open access
PDF file
YCYanyan Cui
Yanyan Cui
LLLixin Liu
Lixin Liu
In recent years, online lending has created many risks while providing lending convenience to Chinese individuals and small and medium-sized enterprises. The timely assessment and prediction of the status of industry indicators is an important prerequisite for effectively preventing the spread of risks in China’s new financial formats. The role of investor sentiment should not be underestimated. We first use the BERT model to divide investor sentiment in the review information of China’s online lending third-party information website into three categories and analyze the relationship between investor sentiment and quantitative indicators of online lending product transactions. The results show that the percentage of positive comments has a positive relationship to the borrowing interest rate of P2P platforms that investors are willing to participate in for bidding projects. The percentage of negative comments has an inverse relationship to the borrowing period. Second, after introducing investor sentiment into the long short-term memory (LSTM) model, the average RMSE of the three forecast periods for borrowing interest rates is 0.373, and that of the borrowing period is 0.262, which are better than the values of other control models. Corresponding suggestions for the risk prevention of China’s new financial formats are made.

21 March 2021
Verified
Conference Paper
Open access
BJBentian Jiang
Bentian Jiang
XZXiaopeng ZHANG
Xiaopeng ZHANG
LLLixin Liu
Lixin Liu
EYEvangeline F.Y. Young
Evangeline F.Y. Young
With the continuous shrinkage of device technology node, the tremendously increasing demands for resolution enhancement technologies (RETs) have created severe concerns over the balance between computational affordability and model accuracy. Having realized the analogies between computational lithography tasks and deep learning-based computer vision applications (e.g., medical image analysis), both industry and academia start gradually migrating various RETs to deep learning-enabled platforms. In this paper, we propose a unified self-training paradigm for building up an end-to-end mask optimization framework from undisclosable layout patterns. Our proposed flow comprises (1) a learning-based pattern generation stage to massively synthesize diverse and realistic layout patterns following the distribution of the undisclosable target layouts, while keeping these confidential layouts blind for any successive training stage, and (2) a complete self-training stage for building up an end-to-end on-neural-network mask optimization framework from scratch, which only requires the aforementioned generated patterns and a compact lithography simulation model as the inputs. Quantitative results demonstrate that our proposed flow achieves comparable state-of-the-art (SOTA) performance in terms of both mask printability and mask correction time while reducing 66% of the turn around time for flow construction.