Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis
High level synthesis (HLS) tools enable the use of high level languages such as C, C++ and SystemC for VLSI design. This simplifies the programming task and also allows the programmers to apply various pragmas or synthesis directives for controlling the hardware design parameters. Since these direct...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-10-01
|
| Series: | Engineering Science and Technology, an International Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S221509862100210X |
| _version_ | 1828344035554623488 |
|---|---|
| author | Meena Belwal T.K. Ramesh |
| author_facet | Meena Belwal T.K. Ramesh |
| author_sort | Meena Belwal |
| collection | DOAJ |
| description | High level synthesis (HLS) tools enable the use of high level languages such as C, C++ and SystemC for VLSI design. This simplifies the programming task and also allows the programmers to apply various pragmas or synthesis directives for controlling the hardware design parameters. Since these directives can take multiple values and also can be applied in many places for ASIC and FPGA designs, the design space grows exponentially making the design space exploration time consuming. Predicting Pareto optimal designs by performing HLS for minimum possible designs has been a driving force to bring in the learning techniques such as Random forests and Gaussian Process models. However, these techniques suffer from scalability issues in large design space or are ineffective in utilizing the prediction uncertainty information for model refinement. We propose a novel active learning approach for design space exploration (Q-PIR) based on the theory of Quantile Regression Forests. Our technique uses the conditional quantiles and prediction intervals to build the region of prediction uncertainty for model refinement and Pareto front discovery in the objective space of area and latency. Through experimental evidence across HLS specific benchmarks, our approach demonstrates better performance in Pareto front discovery than the state-of-the art approaches. |
| first_indexed | 2024-04-13T23:50:43Z |
| format | Article |
| id | doaj.art-535c9af4600947f389cf8a5bd2c24975 |
| institution | Directory Open Access Journal |
| issn | 2215-0986 |
| language | English |
| last_indexed | 2024-04-13T23:50:43Z |
| publishDate | 2022-10-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Engineering Science and Technology, an International Journal |
| spelling | doaj.art-535c9af4600947f389cf8a5bd2c249752022-12-22T02:24:05ZengElsevierEngineering Science and Technology, an International Journal2215-09862022-10-0134101078Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesisMeena Belwal0T.K. Ramesh1Department of Computer Science and Engineering, Amrita School of Engineering, Bengaluru, Amrita Vishwa Vidyapeetham, India; Corresponding author.Department of Electronics and Communication Engineering, Amrita School of Engineering, Bengaluru, Amrita Vishwa Vidyapeetham, IndiaHigh level synthesis (HLS) tools enable the use of high level languages such as C, C++ and SystemC for VLSI design. This simplifies the programming task and also allows the programmers to apply various pragmas or synthesis directives for controlling the hardware design parameters. Since these directives can take multiple values and also can be applied in many places for ASIC and FPGA designs, the design space grows exponentially making the design space exploration time consuming. Predicting Pareto optimal designs by performing HLS for minimum possible designs has been a driving force to bring in the learning techniques such as Random forests and Gaussian Process models. However, these techniques suffer from scalability issues in large design space or are ineffective in utilizing the prediction uncertainty information for model refinement. We propose a novel active learning approach for design space exploration (Q-PIR) based on the theory of Quantile Regression Forests. Our technique uses the conditional quantiles and prediction intervals to build the region of prediction uncertainty for model refinement and Pareto front discovery in the objective space of area and latency. Through experimental evidence across HLS specific benchmarks, our approach demonstrates better performance in Pareto front discovery than the state-of-the art approaches.http://www.sciencedirect.com/science/article/pii/S221509862100210XActive learningDesign space explorationHigh level synthesisMulti-objective optimizationQuantile regression forests |
| spellingShingle | Meena Belwal T.K. Ramesh Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis Engineering Science and Technology, an International Journal Active learning Design space exploration High level synthesis Multi-objective optimization Quantile regression forests |
| title | Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis |
| title_full | Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis |
| title_fullStr | Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis |
| title_full_unstemmed | Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis |
| title_short | Q-PIR: A quantile based Pareto iterative refinement approach for high-level synthesis |
| title_sort | q pir a quantile based pareto iterative refinement approach for high level synthesis |
| topic | Active learning Design space exploration High level synthesis Multi-objective optimization Quantile regression forests |
| url | http://www.sciencedirect.com/science/article/pii/S221509862100210X |
| work_keys_str_mv | AT meenabelwal qpiraquantilebasedparetoiterativerefinementapproachforhighlevelsynthesis AT tkramesh qpiraquantilebasedparetoiterativerefinementapproachforhighlevelsynthesis |