Large-scale regression and classification datasets usually used for comparing Training times of our algorithm with other standard symbolic regression Polynomial (ranging in size from $4096$ to $16$ million points), profiling We evaluate our algorithm on synthetic datasets for the Pagie Vector operations are also used for computation of the fitness of population We introduce representing candidate solution expressions as prefix lists, whichĮnables evaluation using a fixed-length stack in GPU memory. The selection andĮvaluation steps of the generational GP algorithm are parallelized using CUDA. Used for symbolic regression and binary classification. This paper describes a GPUĪccelerated stack-based variant of the generational GP algorithm which can be It an ideal candidate for GPU based parallelization. GP has several inherent parallel steps, making Authors: Vimarsh Sathia (1), Venkataramana Ganesh (2), Shankara Rao Thejaswi Nanditale (2) ((1) Indian Institute of Technology Madras, (2) NVIDIA Corporation) Download PDF Abstract: Genetic Programming (GP), an evolutionary learning technique, has multipleĪpplications in machine learning such as curve fitting, data modelling, feature
