Artificial neural networks
Investigations into the limitations of spiking neural networks have led researchers to explore innovative solutions for improving their performance. A recent study published in the Frontiers in Neuroscience journal presents a novel algorithm-hardware co-design framework centered on a Ternary-8-bit Hybrid Weight Quantization (T8HWQ) scheme. This approach aims to bridge the
Machine learning
Signal Advance's patented Analog Guard encryption system has demonstrated significant resistance to decryption by advanced artificial intelligence (AI) models, offering a new layer of security against next-generation AI threats. The system's analog-domain design allows it to provide intrinsic defense against AI attacks, making it a promising
Technical institutes
Researchers from the South China University of Technology have made a breakthrough in robotics by developing a novel data-driven distributed recurrent neural network (DDD-RNN) that enables precise collaborative motion generation in multimanipulator systems (MMCs) with unknown structural parameters. This innovative approach overcomes traditional methods that rely on precise models and
Machine learning
Researchers from the Iran University of Science and Technology have developed an innovative hybrid control approach to improve the precision and reliability of ferromagnetic continuum robots (FCRs) for medical applications. The approach combines neural network-based modeling with feedback linearization to mitigate the complex nonlinearities of FCRs, ensuring precise trajectory tracking
Machine learning
Research conducted at Northeastern University in Shenyang, People's Republic of China, has led to a groundbreaking study on machine learning and support vector machines. The study proposes an asymmetrical optical fiber Mach-Zehnder interferometer (MZI) that utilizes support vector machine (SVM) and artificial neural network (ANN) to sense vector
Technical institutes
Researchers at Technical University Munich have developed a new method for rotational equivariance in graph neural networks that achieves high predictive accuracy while maintaining a smaller computational footprint. The method, which transforms vector features and spatial neighborhood information into local, node-specific vector bases, is designed to address the need for
Technical institutes
Researchers from Shanghai Institute of Technology have made significant breakthroughs in sensor research, introducing a novel framework for energy-efficient sensing applications. The study focuses on dynamic tracking with event attention spiking networks, which have the potential to revolutionize event-driven processing of asynchronous event streams from Dynamic Vision Sensors (DVSs). The
Artificial neural networks
Researchers at the PSG Institute of Technology and Applied Research have made a groundbreaking discovery in the field of artificial neural networks, successfully developing a model to predict the functional fatigue behavior of shape memory alloys (SMAs). The novel approach utilizes a feed-forward backpropagation artificial neural network (ANN) to capture
Technical institutes
Researchers at Kielce University of Technology have made a significant breakthrough in understanding the applications of machine learning (ML) and neural network (NN) techniques in computational mechanics. Their comprehensive review paper, published in Applied Sciences, provides an in-depth analysis of the recent advancements in the field. The study highlights the
Artificial neural networks
Researchers from Racing 92 in Le Plessis-Robinson, France, have developed a machine learning model to predict players' readiness for training and improve injury prevention strategies in professional rugby. The study, published in Clinical Biomechanics, used daily data from a professional rugby club to train and evaluate various machine learning
Algorithms
Researchers at the Chinese Academy of Sciences have made a significant advancement in the field of machine learning by developing a novel automated graph neural network (GNN) architecture search framework, dubbed AutoGNP, designed to tackle nonlinear optimization problems. This framework represents a major breakthrough in the application of GNNs to
Machine learning
Researchers from the University of Blida 1 have proposed a novel combination of neural network model predictive control with active disturbance rejection for trajectory tracking of a 4 degrees of freedom robot manipulator. The method leverages a neural network for accurate prediction of robot dynamics, enabling effective trajectory tracking while
