Bayesian optimisation is a popular, surrogate model-based approach for optimising expensive black-box functions. Given a surrogate model, the next location to expensively evaluate is chosen via maximisation of a cheap-to-query acquisition function. We present an $\epsilon$-greedy procedure for Bayesian optimisation in batch settings in which the black-box function can be evaluated multiple times in parallel. Our $\epsilon$-shotgun algorithm leverages the model's prediction, uncertainty, and the approximated rate of change of the landscape to determine the spread of batch solutions to be distributed around a putative location. The initial target location is selected either in an exploitative fashion on the mean prediction, or -- with probability $\epsilon$ -- from elsewhere in the design space. This results in locations that are more densely sampled in regions where the function is changing rapidly and in locations predicted to be good (i.e close to predicted optima), with more scattered samples in regions where the function is flatter and/or of poorer quality. We empirically evaluate the $\epsilon$-shotgun methods on a range of synthetic functions and two real-world problems, finding that they perform at least as well as state-of-the-art batch methods and in many cases exceed their performance.

The ultimate aim of artificial intelligence research is to develop agents with truly intelligent behaviors, akin to those found in humans and animals. To this end, a number of tools and techniques have been developed. In recent years, two approaches in particular - deep learning (DL) and reinforcement learning (RL), seem to have made considerable progress towards this goal. Both these fields have been widely studied, with numerous successful examples [22, 29, 42, 25, 40] reported, particularly in recent years. However, even with the unprecedented success of recent approaches such as deep RL [28, 27, 36], poor sample efficiency and limited generalization remain major concerns to be addressed, keeping in view the ultimate goal of developing general purpose agents. The poor generalization capability of DL is exposed by its liability to deception when presented with adversarial examples [30, 39]. Recent work [38], showed that it was possible to hurt the performance of DLbased image recognition systems by carefully altering just a single pixel.

Most decision tree induction algorithms are based on a greedy top-down recursive partitioning strategy for tree growth. In this paper, we propose several methods for induction of decision trees and their ensembles based on evolutionary algorithms. The main difference of our approach is using real-valued vector representation of decision tree that allows to use a large number of different optimization algorithms, as well as optimize the whole tree or ensemble for avoiding local optima. Differential evolution and evolution strategies were chosen as optimization algorithms, as they have good results in reinforcement learning problems. We test the predictive performance of this methods using several public UCI data sets, and the proposed methods show better quality than classical methods.

We present the implementation of a new stochastic Kuramoto-Vicsek-type model for global optimization of nonconvex functions on the sphere. This model belongs to the class of Consensus-Based Optimization. In fact, particles move on the sphere driven by a drift towards an instantaneous consensus point, which is computed as a convex combination of particle locations, weighted by the cost function according to Laplace's principle, and it represents an approximation to a global minimizer. The dynamics is further perturbed by a random vector field to favor exploration, whose variance is a function of the distance of the particles to the consensus point. In particular, as soon as the consensus is reached the stochastic component vanishes. The main results of this paper are about the proof of convergence of the numerical scheme to global minimizers provided conditions of well-preparation of the initial datum. The proof combines previous results of mean-field limit with a novel asymptotic analysis, and classical convergence results of numerical methods for SDE. We present several numerical experiments, which show that the algorithm proposed in the present paper scales well with the dimension and is extremely versatile. To quantify the performances of the new approach, we show that the algorithm is able to perform essentially as good as ad hoc state of the art methods in challenging problems in signal processing and machine learning, namely the phase retrieval problem and the robust subspace detection.

Dubins tours represent a solution of the Dubins Traveling Salesman Problem (DTSP) that is a variant of the optimization routing problem to determine a curvature-constrained shortest path to visit a set of locations such that the path is feasible for Dubins vehicle, which moves only forward and has a limited turning radius. The DTSP combines the NP-hard combinatorial optimization to determine the optimal sequence of visits to the locations, as in the regular TSP, with the continuous optimization of the heading angles at the locations, where the optimal heading values depend on the sequence of visits and vice versa. We address the computationally challenging DTSP by fast evaluation of the sequence of visits by the proposed Windowing Surrogate Model (WiSM) which estimates the length of the optimal Dubins path connecting a sequence of locations in a Dubins tour. The estimation is sped up by a regression model trained using close to optimum solutions of small Dubins tours that are generalized for large-scale instances of the addressed DTSP utilizing the sliding window technique and a cache for already computed results. The reported results support that the proposed WiSM enables a fast convergence of a relatively simple evolutionary algorithm to high-quality solutions of the DTSP. We show that with an increasing number of locations, our algorithm scales significantly better than other state-of-the-art DTSP solvers.

NeuroEvolution is one of the most competitive evolutionary learning frameworks for designing novel neural networks for use in specific tasks, such as logic circuit design and digital gaming. However, the application of benchmark methods such as the NeuroEvolution of Augmenting Topologies (NEAT) remains a challenge, in terms of their computational cost and search time inefficiency. This paper advances a method which incorporates a type of topological edge coding, named Reverse Encoding Tree (RET), for evolving scalable neural networks efficiently. Using RET, two types of approaches -- NEAT with Binary search encoding (Bi-NEAT) and NEAT with Golden-Section search encoding (GS-NEAT) -- have been designed to solve problems in benchmark continuous learning environments such as logic gates, Cartpole, and Lunar Lander, and tested against classical NEAT and FS-NEAT as baselines. Additionally, we conduct a robustness test to evaluate the resilience of the proposed NEAT algorithms. The results show that the two proposed strategies deliver an improved performance, characterized by (1) a higher accumulated reward within a finite number of time steps; (2) using fewer episodes to solve problems in targeted environments, and (3) maintaining adaptive robustness under noisy perturbations, which outperform the baselines in all tested cases. Our analysis also demonstrates that RET expends potential future research directions in dynamic environments. Code is available from https://github.com/HaolingZHANG/ReverseEncodingTree.

Loss function optimization for neural networks has recently emerged as a new direction for metalearning, with Genetic Loss Optimization (GLO) providing a general approach for the discovery and optimization of such functions. GLO represents loss functions as trees that are evolved and further optimized using evolutionary strategies. However, searching in this space is difficult because most candidates are not valid loss functions. In this paper, a new technique, Multivariate Taylor expansion-based genetic loss-function optimization (TaylorGLO), is introduced to solve this problem. It represents functions using a novel parameterization based on Taylor expansions, making the search more effective. TaylorGLO is able to find new loss functions that outperform those found by GLO in many fewer generations, demonstrating that loss function optimization is a productive avenue for metalearning.

--One problem found when working with satellite images is the radiometric variations across the image and different images. Intending to improve remote sensing models for the classification of burnt areas, we set two objectives. The first is to understand the relationship between feature spaces and the predictive ability of the models, allowing us to explain the differences between learning and generalization when training and testing in different datasets. We find that training on datasets built from more than one image provides models that generalize better . These results are explained by visualizing the dispersion of values on the feature space. The second objective is to evolve hyper-features that improve the performance of different classifiers on a variety of test sets. We find the hyper-features to be beneficial, and obtain the best models with XGBoost, even if the hyper-features are optimized for a different method. Deforestation has serious implications on biodiversity, on rural communities that depend on forests for survival, and on greenhouse gas emissions that drive the global climate. The machine learning (ML) community can help by providing predictive models that, after learning from a small sample of an image, can automatically classify the whole image. Although previous ML work in forest monitoring has shown good results, the predictive models are often applied on the same location where they were learnt, i.e., the models are trained and tested in samples from the same dataset (e.g., [1]) or time series from the same area (e.g., [2]).

Feature selection, as a critical pre-processing step for machine learning, aims at determining representative predictors from a high-dimensional feature space dataset to improve the prediction accuracy. However, the increase in feature space dimensionality, comparing to the number of observations, poses a severe challenge to many existing feature selection methods with respect to computational efficiency and prediction performance. This paper presents a new hybrid two-layer feature selection approach that combines a wrapper and an embedded method in constructing an appropriate subset of predictors. In the first layer of the proposed method, the Genetic Algorithm(GA) has been adopted as a wrapper to search for the optimal subset of predictors, which aims to reduce the number of predictors and the prediction error. As one of the meta-heuristic approaches, GA is selected due to its computational efficiency; however, GAs do not guarantee the optimality. To address this issue, a second layer is added to the proposed method to eliminate any remaining redundant/irrelevant predictors to improve the prediction accuracy. Elastic Net(EN) has been selected as the embedded method in the second layer because of its flexibility in adjusting the penalty terms in regularization process and time efficiency. This hybrid two-layer approach has been applied on a Maize genetic dataset from NAM population, which consists of multiple subsets of datasets with different ratio of the number of predictors to the number of observations. The numerical results confirm the superiority of the proposed model.

Lindenmayer systems (L-systems) are a grammar system that consist of string rewriting rules. The rules replace every symbol in a string in parallel with a successor to produce the next string, and this procedure iterates. In a stochastic context-free L-system (S0L-system), every symbol may have one or more rewriting rule, each with an associated probability of selection. Properly constructed rewriting rules have been found to be useful for modeling and simulating some natural and human engineered processes where each derived string describes a step in the simulation. Typically, processes are modeled by experts who meticulously construct the rules based on measurements or domain knowledge of the process. This paper presents an automated approach to finding stochastic L-systems, given a set of string sequences as input. The implemented tool is called the Plant Model Inference Tool for S0L-systems (PMIT-S0L). PMIT-S0L is evaluated using 960 procedurally generated S0L-systems in a test suite, which are each used to generate input strings, and PMIT-S0L is then used to infer the system from only the sequences. The evaluation shows that PMIT-S0L infers S0L-systems with up to 9 rewriting rules each in under 12 hours. Additionally, it is found that 3 sequences of strings is sufficient to find the correct original rewriting rules in 100% of the cases in the test suite, and 6 sequences of strings reduces the difference in the associated probabilities to approximately 1% or less.