Reproducibility in Evolutionary Computation

Experimental studies are prevalent in Evolutionary Computation ( EC ), and concerns about the reproducibility and replicability of such studies have increased in recent times, reflecting similar concerns in other scientific fields. In this article, we discuss, within the context of EC, the different types of reproducibility and suggest a classification that refines the badge system of the Association of Computing Machinery ( ACM ) adopted by ACM Transactions on Evolutionary Learning and Optimization ( TELO ). We identify cultural and technical obstacles to reproducibility in the EC field. Finally, we provide guidelines and suggest tools that may help to overcome some of these reproducibility obstacles.

Evolutionary learning in the brain by heterosynaptic plasticity

The way in which the brain modifies synapses to improve the performance of complicated networks remains one of the biggest mysteries in neuroscience. Existing proposals lack sufficient experimental support, and neglect inter-cellular signaling pathways ubiquitous in the brain. Here we show that the heterosynaptic plasticity between hippocampal or cortical pyramidal cells mediated by diffusive nitric oxide and astrocyte calcium wave, together with flexible dendritic gating of somatostatin interneurons, implies an evolutionary algorithm (EA). In simulation, this EA is able to train deep networks with biologically plausible binary weights in MNIST classification and Atari-game playing tasks up to performance comparable with continuous-weight networks trained by gradient-based methods. Our work leads paradigmatically fresh understanding of the brain learning mechanism.

Supervised evolutionary learning: Use of gradient histogram & particle swarm algorithm to detection & tracking pedestrian in sequence of infrared images

Recently, tracking and pedestrian detection from various images have become one of the major issues in the field of image processing and statistical identification. In this regard, using evolutionary learning-based approaches to improve performance in different contexts can greatly influence the appropriate response. There are problems with pedestrian tracking/identification, such as low accuracy for detection, high processing time, and uncertainty in response to answers. Researchers are looking for new processing models that can accurately monitor one's position on the move. In this study, a hybrid algorithm for the automatic detection of pedestrian position is presented. It is worth noting that this method, contrary to the analysis of visible images, examines pedestrians' thermal and infrared components while walking and combines a neural network with maximum learning capability, wavelet kernel (Wavelet transform), and particle swarm optimization (PSO) to find parameters of learner model. Gradient histograms have a high effect on extracting features in infrared images. As well, the neural network algorithm can achieve its goal (pedestrian detection and tracking) by maximizing learning. The proposed method, despite the possibility of maximum learning, has a high speed in education, and results of various data sets in this field have been analyzed. The result indicates a negligible error in observing the infrared sequence of pedestrian movements, and it is suggested to use neural networks because of their precision and trying to boost the selection of their hyperparameters based on evolutionary algorithms.

Evolutionary Learning-Derived Clinical-Radiomic Models for Predicting Early Recurrence of Hepatocellular Carcinoma after Resection

<b><i>Background and Aims:</i></b> Current prediction models for early recurrence of hepatocellular carcinoma (HCC) after surgical resection remain unsatisfactory. The aim of this study was to develop evolutionary learning-derived prediction models with interpretability using both clinical and radiomic features to predict early recurrence of HCC after surgical resection. <b><i>Methods:</i></b> Consecutive 517 HCC patients receiving surgical resection with available contrast-enhanced computed tomography (CECT) images before resection were retrospectively enrolled. Patients were randomly assigned to a training set (<i>n</i> = 362) and a test set (<i>n</i> = 155) in a ratio of 7:3. Tumor segmentation of all CECT images including noncontrast phase, arterial phase, and portal venous phase was manually performed for radiomic feature extraction. A novel evolutionary learning-derived method called genetic algorithm for predicting recurrence after surgery of liver cancer (GARSL) was proposed to design prediction models for early recurrence of HCC within 2 years after surgery. <b><i>Results:</i></b> A total of 143 features, including 26 preoperative clinical features, 5 postoperative pathological features, and 112 radiomic features were used to develop GARSL preoperative and postoperative models. The area under the receiver operating characteristic curves (AUCs) for early recurrence of HCC within 2 years were 0.781 and 0.767, respectively, in the training set, and 0.739 and 0.741, respectively, in the test set. The accuracy of GARSL models derived from the evolutionary learning method was significantly better than models derived from other well-known machine learning methods or the early recurrence after surgery for liver tumor (ERASL) preoperative (AUC = 0.687, <i>p</i> &#x3c; 0.001 vs. GARSL preoperative) and ERASL postoperative (AUC = 0.688, <i>p</i> &#x3c; 0.001 vs. GARSL postoperative) models using clinical features only. <b><i>Conclusion:</i></b> The GARSL models using both clinical and radiomic features significantly improved the accuracy to predict early recurrence of HCC after surgical resection, which was significantly better than other well-known machine learning-derived models and currently available clinical models.

Binary Black-Box Adversarial Attacks with Evolutionary Learning against IoT Malware Detection

5G is about to open Pandora’s box of security threats to the Internet of Things (IoT). Key technologies, such as network function virtualization and edge computing introduced by the 5G network, bring new security threats and risks to the Internet infrastructure. Therefore, higher detection and defense against malware are required. Nowadays, deep learning (DL) is widely used in malware detection. Recently, research has demonstrated that adversarial attacks have posed a hazard to DL-based models. The key issue of enhancing the antiattack performance of malware detection systems that are used to detect adversarial attacks is to generate effective adversarial samples. However, numerous existing methods to generate adversarial samples are manual feature extraction or using white-box models, which makes it not applicable in the actual scenarios. This paper presents an effective binary manipulation-based attack framework, which generates adversarial samples with an evolutionary learning algorithm. The framework chooses some appropriate action sequences to modify malicious samples. Thus, the modified malware can successfully circumvent the detection system. The evolutionary algorithm can adaptively simplify the modification actions and make the adversarial sample more targeted. Our approach can efficiently generate adversarial samples without human intervention. The generated adversarial samples can effectively combat DL-based malware detection models while preserving the consistency of the executable and malicious behavior of the original malware samples. We apply the generated adversarial samples to attack the detection engines of VirusTotal. Experimental results illustrate that the adversarial samples generated by our method reach an evasion success rate of 47.8%, which outperforms other attack methods. By adding adversarial samples in the training process, the MalConv network is retrained. We show that the detection accuracy is improved by 10.3%.

Adverse Selection, Heterogeneous Beliefs, and Evolutionary Learning

We relax the common assumption of homogeneous beliefs in principal-agent relationships with adverse selection. Principals are competitors in the product market and write contracts also on the base of an expected aggregate. The model is a version of a cobweb model. In an evolutionary learning set-up, which is imitative, principals can have different beliefs about the distribution of agents’ types in the population. The resulting nonlinear dynamic system is studied. Convergence to a uniform belief depends on the relative size of the bias in beliefs.