Neural networks are trained by minimizing a loss function that defines the discrepancy between the predicted model output and the target value. The selection of the loss function is crucial to achieve task-specific behaviour and highly influences the capability of the model. A variety of loss functions have been proposed for a wide range of tasks affecting training and model performance. For classification tasks, the cross entropy is the de-facto standard and usually the first choice. Here, we try to experimentally challenge the well-known loss functions, including cross entropy (CE) loss, by utilizing the genetic programming (GP) approach, a population-based evolutionary algorithm. GP constructs loss functions from a set of operators and leaf nodes and these functions are repeatedly recombined and mutated to find an optimal structure. Experiments were carried out on different small-sized datasets CIFAR-10, CIFAR-100 and Fashion-MNIST using an Inception model. The 5 best functions found were evaluated for different model architectures on a set of standard datasets ranging from 2 to 102 classes and very different sizes. One function, denoted as Next Generation Loss (NGL), clearly stood out showing same or better performance for all tested datasets compared to CE. To evaluate the NGL function on a large-scale dataset, we tested its performance on the Imagenet-1k dataset where it showed improved top-1 accuracy compared to models trained with identical settings and other losses. Finally, the NGL was trained on a segmentation downstream task for Pascal VOC 2012 and COCO-Stuff164k datasets improving the underlying model performance.

Advanced biological intelligence learns efficiently from an information-rich stream of stimulus information, even when feedback on behaviour quality is sparse or absent. Such learning exploits implicit assumptions about task domains. We refer to such learning as Domain-Adapted Learning (DAL). In contrast, AI learning algorithms rely on explicit externally provided measures of behaviour quality to acquire fit behaviour. This imposes an information bottleneck that precludes learning from diverse non-reward stimulus information, limiting learning efficiency. We consider the question of how biological evolution circumvents this bottleneck to produce DAL. We propose that species first evolve the ability to learn from reward signals, providing inefficient (bottlenecked) but broad adaptivity. From there, integration of non-reward information into the learning process can proceed via gradual accumulation of biases induced by such information on specific task domains. This scenario provides a biologically plausible pathway towards bottleneck-free, domain-adapted learning. Focusing on the second phase of this scenario, we set up a population of NNs with reward-driven learning modelled as Reinforcement Learning (A2C), and allow evolution to improve learning efficiency by integrating non-reward information into the learning process using a neuromodulatory update mechanism. On a navigation task in continuous 2D space, evolved DAL agents show a 300-fold increase in learning speed compared to pure RL agents. Evolution is found to eliminate reliance on reward information altogether, allowing DAL agents to learn from non-reward information exclusively, using local neuromodulation-based connection weight updates only.

This paper introduces a novel approach, evolutionary multi-objective optimisation for fairness-aware self-adjusting memory classifiers, designed to enhance fairness in machine learning algorithms applied to data stream classification. With the growing concern over discrimination in algorithmic decision-making, particularly in dynamic data stream environments, there is a need for methods that ensure fair treatment of individuals across sensitive attributes like race or gender. The proposed approach addresses this challenge by integrating the strengths of the self-adjusting memory K-Nearest-Neighbour algorithm with evolutionary multi-objective optimisation. This combination allows the new approach to efficiently manage concept drift in streaming data and leverage the flexibility of evolutionary multi-objective optimisation to maximise accuracy and minimise discrimination simultaneously. We demonstrate the effectiveness of the proposed approach through extensive experiments on various datasets, comparing its performance against several baseline methods in terms of accuracy and fairness metrics. Our results show that the proposed approach maintains competitive accuracy and significantly reduces discrimination, highlighting its potential as a robust solution for fairness-aware data stream classification. Further analyses also confirm the effectiveness of the strategies to trigger evolutionary multi-objective optimisation and adapt classifiers in the proposed approach.

The diversity optimization is the class of optimization problems, in which we aim at finding a diverse set of good solutions. One of the frequently used approaches to solve such problems is to use evolutionary algorithms which evolve a desired diverse population. This approach is called evolutionary diversity optimization (EDO). In this paper, we analyse EDO on a 3-objective function LOTZ$_k$, which is a modification of the 2-objective benchmark function (LeadingOnes, TrailingZeros). We prove that the GSEMO computes a set of all Pareto-optimal solutions in $O(kn^3)$ expected iterations. We also analyze the runtime of the GSEMO$_D$ (a modification of the GSEMO for diversity optimization) until it finds a population with the best possible diversity for two different diversity measures, the total imbalance and the sorted imbalances vector. For the first measure we show that the GSEMO$_D$ optimizes it asymptotically faster than it finds a Pareto-optimal population, in $O(kn^2\log(n))$ expected iterations, and for the second measure we show an upper bound of $O(k^2n^3\log(n))$ expected iterations. We complement our theoretical analysis with an empirical study, which shows a very similar behavior for both diversity measures that is close to the theory predictions.

Recently surrogate functions based on the tail inequalities were developed to evaluate the chance constraints in the context of evolutionary computation and several Pareto optimization algorithms using these surrogates were successfully applied in optimizing chance-constrained monotone submodular problems. However, the difference in performance between algorithms using the surrogates and those employing the direct sampling-based evaluation remains unclear. Within the paper, a sampling-based method is proposed to directly evaluate the chance constraint. Furthermore, to address the problems with more challenging settings, an enhanced GSEMO algorithm integrated with an adaptive sliding window, called ASW-GSEMO, is introduced. In the experiments, the ASW-GSEMO employing the sampling-based approach is tested on the chance-constrained version of the maximum coverage problem with different settings. Its results are compared with those from other algorithms using different surrogate functions. The experimental findings indicate that the ASW-GSEMO with the sampling-based evaluation approach outperforms other algorithms, highlighting that the performances of algorithms using different evaluation methods are comparable. Additionally, the behaviors of ASW-GSEMO are visualized to explain the distinctions between it and the algorithms utilizing the surrogate functions.

The rapidly changing landscapes of modern optimization problems require algorithms that can be adapted in real-time. This paper introduces an Adaptive Metaheuristic Framework (AMF) designed for dynamic environments. It is capable of intelligently adapting to changes in the problem parameters. The AMF combines a dynamic representation of problems, a real-time sensing system, and adaptive techniques to navigate continuously changing optimization environments. Through a simulated dynamic optimization problem, the AMF's capability is demonstrated to detect environmental changes and proactively adjust its search strategy. This framework utilizes a differential evolution algorithm that is improved with an adaptation module that adjusts solutions in response to detected changes. The capability of the AMF to adjust is tested through a series of iterations, demonstrating its resilience and robustness in sustaining solution quality despite the problem's development. The effectiveness of AMF is demonstrated through a series of simulations on a dynamic optimization problem. Robustness and agility characterize the algorithm's performance, as evidenced by the presented fitness evolution and solution path visualizations. The findings show that AMF is a practical solution to dynamic optimization and a major step forward in the creation of algorithms that can handle the unpredictability of real-world problems.

Choosing an appropriate optimization algorithm is essential to achieving success in optimization challenges. Here we present a new evolutionary algorithm structure that utilizes a reinforcement learning-based agent aimed at addressing these issues. The agent employs a double deep q-network to choose a specific evolutionary operator based on feedback it receives from the environment during optimization. The algorithm's structure contains five single-objective evolutionary algorithm operators. This single-objective structure is transformed into a multi-objective one using the R2 indicator. This indicator serves two purposes within our structure: first, it renders the algorithm multi-objective, and second, provides a means to evaluate each algorithm's performance in each generation to facilitate constructing the reinforcement learning-based reward function. The proposed R2-reinforcement learning multi-objective evolutionary algorithm (R2-RLMOEA) is compared with six other multi-objective algorithms that are based on R2 indicators. These six algorithms include the operators used in R2-RLMOEA as well as an R2 indicator-based algorithm that randomly selects operators during optimization. We benchmark performance using the CEC09 functions, with performance measured by inverted generational distance and spacing. The R2-RLMOEA algorithm outperforms all other algorithms with strong statistical significance (p<0.001) when compared with the average spacing metric across all ten benchmarks.

Unmanned boats, while navigating at sea, utilize active compensation systems to mitigate wave disturbances experienced by onboard instruments and equipment. However, there exists a lag in the measurement of unmanned boat attitudes, thus introducing unmanned boat motion attitude prediction to compensate for the lag in the signal acquisition process. This paper, based on the basic principles of waves, derives the disturbance patterns of waves on unmanned boats from the wave energy spectrum. Through simulation analysis of unmanned boat motion attitudes, motion attitude data is obtained, providing experimental data for subsequent work. A combined prediction model based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Particle Swarm Optimization (PSO), and Support Vector Machine (SVM) is designed to predict the motion attitude of unmanned boats. Simulation results validate its superior prediction accuracy compared to traditional prediction models. For example, in terms of mean absolute error, it improves by 17% compared to the EMD-PSO-SVM model.

One of the ways to make artificial intelligence more natural is to give it some room for doubt. Two main questions should be resolved in that way. First, how to train a model to estimate uncertainties of its own predictions? And then, what to do with the uncertain predictions if they appear? First, we proposed an uncertainty-aware negative log-likelihood loss for the case of N-dimensional multivariate normal distribution with spherical variance matrix to the solution of N-classes classification tasks. The loss is similar to the heteroscedastic regression loss. The proposed model regularizes uncertain predictions, and trains to calculate both the predictions and their uncertainty estimations. The model fits well with the label smoothing technique. Second, we expanded the limits of data augmentation at the training and test stages, and made the trained model to give multiple predictions for a given number of augmented versions of each test sample. Given the multi-view predictions together with their uncertainties and confidences, we proposed several methods to calculate final predictions, including mode values and bin counts with soft and hard weights. For the latter method, we formalized the model tuning task in the form of multimodal optimization with non-differentiable criteria of maximum accuracy, and applied particle swarm optimization to solve the tuning task. The proposed methodology was tested using CIFAR-10 dataset with clean and noisy labels and demonstrated good results in comparison with other uncertainty estimation methods related to sample selection, co-teaching, and label smoothing.

Multimodal pre-trained models, such as CLIP, are popular for zero-shot classification due to their open-vocabulary flexibility and high performance. However, vision-language models, which compute similarity scores between images and class labels, are largely black-box, with limited interpretability, risk for bias, and inability to discover new visual concepts not written down. Moreover, in practical settings, the vocabulary for class names and attributes of specialized concepts will not be known, preventing these methods from performing well on images uncommon in large-scale vision-language datasets. To address these limitations, we present a novel method that discovers interpretable yet discriminative sets of attributes for visual recognition. We introduce an evolutionary search algorithm that uses a large language model and its in-context learning abilities to iteratively mutate a concept bottleneck of attributes for classification. Our method produces state-of-the-art, interpretable fine-grained classifiers. We outperform the latest baselines by 18.4% on five fine-grained iNaturalist datasets and by 22.2% on two KikiBouba datasets, despite the baselines having access to privileged information about class names.