The ability to predict, anticipate and reason about future outcomes is a key component of intelligent decision-making systems. In light of the success of deep learning in computer vision, deep-learning-based video prediction emerged as a promising research direction. Defined as a self-supervised learning task, video prediction represents a suitable framework for representation learning, as it demonstrated potential capabilities for extracting meaningful representations of the underlying patterns in natural videos. Motivated by the increasing interest in this task, we provide a review on the deep learning methods for prediction in video sequences. We first define the video prediction fundamentals, as well as mandatory background concepts and the most used datasets. Next, we carefully analyze existing video prediction models organized according to a proposed taxonomy, highlighting their contributions and their significance in the field. The summary of the datasets and methods is accompanied with experimental results that facilitate the assessment of the state of the art on a quantitative basis. The paper is summarized by drawing some general conclusions, identifying open research challenges and by pointing out future research directions.

The optimal selection of chemical features (molecular descriptors) is an essential pre-processing step for the efficient application of computational intelligence techniques in virtual screening for identification of bioactive molecules in drug discovery. The selection of molecular descriptors has key influence in the accuracy of affinity prediction. In order to improve this prediction, we examined a Random Forest (RF)-based approach to automatically select molecular descriptors of training data for ligands of kinases, nuclear hormone receptors, and other enzymes. The reduction of features to use during prediction dramatically reduces the computing time over existing approaches and consequently permits the exploration of much larger sets of experimental data. To test the validity of the method, we compared the results of our approach with the ones obtained using manual feature selection in our previous study (Perez-Sanchez, Cano, and Garcia-Rodriguez, 2014).The main novelty of this work in the field of drug discovery is the use of RF in two different ways: feature ranking and dimensionality reduction, and classification using the automatically selected feature subset. Our RF-based method outperforms classification results provided by Support Vector Machine (SVM) and Neural Networks (NN) approaches.This work was partially supported by the Fundación Séneca del Centro de Coordinación de la Investigación de la Región de Murcia under Project 18946/JLI/13. This work has been funded by the Nils Coordinated Mobility under grant 012-ABEL-CM-2014A, in part financed by the European Regional Development Fund (ERDF)

Digital twins (DTs) represent a transformative technology in manufacturing, facilitating significant advancements in monitoring, simulation, and optimization. This paper offers an extensive bibliographic review of AI-Based DT applications, categorized into three principal dimensions: operator, process, and product. The operator dimension focuses on enhancing safety and ergonomics through intelligent assistance, utilizing real-time monitoring and artificial intelligence, notably in human–robot collaboration contexts. The process application concerns itself with optimizing production flows, identifying bottlenecks, and dynamically reconfiguring systems through predictive models and real-time simulations. Lastly, the product dimension emphasizes the applications focused on the improvements in product design and quality, employing lifecycle and historical data to satisfy evolving market requirements. This categorization provides a structured framework for analyzing the specific capabilities and trends of DTs, while also identifying knowledge gaps in contemporary research. This review highlights the key challenges of technological interoperability, data integration, and high implementation costs while emphasizing how digital twins, supported by AI, can drive the transition toward sustainable, human-centered manufacturing systems in line with Industry 5.0. The findings provide valuable insights for advancing the state of the art and exploring future opportunities in digital twin applications.

Object pose estimation is a core perception task that enables, for example, object manipulation and scene understanding. The widely available, inexpensive and high-resolution RGB sensors and CNNs that allow for fast inference make monocular approaches especially well suited for robotics applications. We observe that previous surveys on establish the state of the art for varying modalities, single- and multi-view settings, and datasets and metrics that consider a multitude of applications. We argue, however, that those works' broad scope hinders the identification of open challenges that are specific to monocular approaches and the derivation of promising future challenges for their application in robotics. By providing a unified view on recent publications from both robotics and computer vision, we find that occlusion handling, pose representations, and formalizing and improving category-level pose estimation are still fundamental challenges that are highly relevant for robotics. Moreover, to further improve robotic performance, large object sets, novel objects, refractive materials, and uncertainty estimates are central, largely unsolved open challenges. In order to address them, ontological reasoning, deformability handling, scene-level reasoning, realistic datasets, and the ecological footprint of algorithms need to be improved.

The kNN (k-nearest neighbors) classification algorithm is one of the most widely used non-parametric classification methods, however it is limited due to memory consumption related to the size of the dataset, which makes them impractical to apply to large volumes of data. Variations of this method have been proposed, such as condensed KNN which divides the training dataset into clusters to be classified, other variations reduce the input dataset in order to apply the algorithm. This paper presents a variation of the kNN algorithm, of the type structure less NN, to work with categorical data. Categorical data, due to their nature, can be compressed in order to decrease the memory requirements at the time of executing the classification. The method proposes a previous phase of compression of the data to then apply the algorithm on the compressed data. This allows us to maintain the whole dataset in memory which leads to a considerable reduction of the amount of memory required. Experiments and tests carried out on known datasets show the reduction in the volume of information stored in memory and maintain the accuracy of the classification. They also show a slight decrease in processing time because the information is decompressed in real time (on-the-fly) while the algorithm is running.