Work-in-Progress: Toward a Robust, Reconfigurable Hardware Accelerator for Tree-Based Genetic Programming

Crary, Christopher; Piard, Wesley; Chesley, Britton; Stitt, Greg

doi:10.1109/cases55004.2022.00015

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It even outperformed Operon, the most efficient CPU-based approach. This is a notable result, since Operon is the most accurate method in SRBench and has been shown to be more efficient than previous GPU-based implementations of GP [49]. Indeed, performance on the SRBench benchmark suite shows that runtimes are largely robust to the size of the learning problem.…”

Section: Discussionmentioning

confidence: 87%

“…While it is natural to expect that a GPU-based implementation will require less wall-clock time than a CPU implementation, this is not necessarily the case. Indeed, recent works have shown that CPU-based GP systems, such as Operon, for instance, can outperform GPU and FPGA implementations of [49,63]. Concerning model size, for the black-box problems, the methods in SRBench have not been configured to perform symbolic simplification of the models before computing model size.…”

Section: Srbench Resultsmentioning

confidence: 99%

“…For instance, Operon, which produces the most accurate and relatively compact models, implemented in a highly efficient manner [7]. In fact, while Operon runs on a CPU, its multithreaded implementation allows it to be more efficient than some FPGA-and GPUbased implementations [49].…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

M5GP: Parallel Multidimensional Genetic Programming with Multidimensional Populations for Symbolic Regression

Cárdenas Florido,

Trujillo,

Hernandez

et al. 2024

MCA

View full text Add to dashboard Cite

Machine learning and artificial intelligence are growing in popularity thanks to their ability to produce models that exhibit unprecedented performance in domains that include computer vision, natural language processing and code generation. However, such models tend to be very large and complex and impossible to understand using traditional analysis or human scrutiny. Conversely, Symbolic Regression methods attempt to produce models that are relatively small and (potentially) human-readable. In this domain, Genetic Programming (GP) has proven to be a powerful search strategy that achieves state-of-the-art performance. This paper presents a new GP-based feature transformation method called M5GP, which is hybridized with multiple linear regression to produce linear models, implemented to exploit parallel processing on graphical processing units for efficient computation. M5GP is the most recent variant from a family of feature transformation methods (M2GP, M3GP and M4GP) that have proven to be powerful tools for both classification and regression tasks applied to tabular data. The proposed method was evaluated on SRBench v2.0, the current standard benchmarking suite for Symbolic Regression. Results show that M5GP achieves performance that is competitive with the state-of-the-art, achieving a top-three rank on the most difficult subset of black-box problems. Moreover, it achieves the lowest computation time when compared to other GP-based methods that have similar accuracy scores.

show abstract

Section: Discussionmentioning

confidence: 87%

Section: Srbench Resultsmentioning

confidence: 99%