Geometric semantic genetic programming (GSGP) is a wellknown variant of genetic programming (GP) where recombination and mutation operators have a clear semantic effect. Both kind of operators have randomly selected parameters that are not optimized by the search process. In this paper we combine GSGP with a well-known gradientbased optimizer, Adam, in order to leverage the ability of GP to operate structural changes of the individuals with the ability of gradient-based methods to optimize the parameters of a given structure. Two methods, named HYB-GSGP and HeH-GSGP, are defined and compared with GSGP on a large set of regression problems, showing that the use of Adam can improve the performance on the test set. The idea of merging evolutionary computation and gradient-based optimization is a promising way of combining two methods with very different -and complementary -strengths.

No abstract

Observing the ocean provides us with essential information necessary to study and understand marine ecosystem dynamics, its evolution and the impact of human activities. However, observations are sparse, limited in time and space coverage, and unevenly collected among variables. Our work aims to develop an improved deep-learning technique for predicting relationships between high-frequency and low-frequency sampled variables. Specifically, we use a larger dataset, EMODnet, and train our model for predicting nutrient concentrations and carbonate system variables (low-frequency sampled variables) starting from information such as sampling time and geolocation, temperature, salinity and oxygen (high-frequency sampled variables). Novel elements of our application include (i) the calculation of a confidence interval for prediction based on deep ensembles of neural networks, and (ii) a two-step analysis for the quality check of the input data. The proposed method proves capable of predicting the desired variables with relatively small errors, outperforming the results obtained by the current state-of-the-art models.

Among the evolutionary methods, one that is quite prominent is Genetic Programming, and, in recent years, a variant called Geometric Semantic Genetic Programming (GSGP) has shown to be successfully applicable to many real-world problems. Due to a peculiarity in its implementation, GSGP needs to store all the evolutionary history, i.e., all populations from the first one. We exploit this stored information to define a multi-generational selection scheme that is able to use individuals from older populations. We show that a limited ability to use "old" generations is actually useful for the search process, thus showing a zero-cost way of improving the performances of GSGP.