Articles tagged with Neural Networks
Researchers have identified three cell types in the median raphe nucleus that control decisions on perseverance, exploration, and disengagement. These findings may help understand neuropsychiatric conditions such as OCD, autism, and major depressive disorder.
Researchers discovered a horizontally distributed and modular organization of cortical movement units, with different types of neurons forming functional clusters in distinct regions. The study also found that the brain re-networks and adapts to learn new motor skills.
Researchers develop a novel adaptive nonlinear PID controller integrated with radial basis function neural network for enhanced ballbot functionality. The proposed NPID-RBFNN controller demonstrates superior stability and robustness, outperforming traditional PID and NPID controllers.
Researchers created a computational method to track brain cell development over time, capturing unlabeled cells and fine structures in live cultures. The algorithm achieved high precision rates for detecting individual neurons, paving the way for studying neurological diseases and developing therapies.
A study by Nagoya University researchers found that excessive neuronal activation over time leads to brain function decline, contradicting previous theories. Interventions targeting reduced neuronal hyperactivation, such as dietary changes, may mitigate age-related cognitive decline in humans.
Harvard researchers have developed a silicon chip capable of recording small yet telltale synaptic signals from a large number of neurons. The chip has successfully mapped over 70,000 synaptic connections from approximately 2,000 rat neurons.
A new study led by researchers at Mass General Brigham suggests that different brain regions activated by creative tasks are part of one common brain circuit. People with brain injuries or neurodegenerative diseases may have increased creativity due to changes in this circuit.
The study compared the performance of seven AI models with that of 400 humans in comprehension tasks, revealing a significant difference in accuracy. Human accuracy reached 89%, while AI models struggled to achieve more than 70% correct answers.
A new optical encryption system uses holograms and neural networks to encode information, making it virtually unbreakable. The system achieves an exceptional level of encryption by utilizing a neural network to generate the decryption key.
A recent study published in JAMA Network Open found that heavy cannabis users exhibited reduced brain activity during working memory tasks, associated with worse performance. Abstaining from cannabis before cognitive tasks may help improve performance.
A new tool called EpiScalp uses algorithms trained on dynamic network models to map brainwave patterns and identify hidden signs of epilepsy from a single routine EEG. This tool has ruled out 96% of false positives, cutting potential misdiagnoses among cases by nearly 70%, according to a Johns Hopkins University study.
The new model, based on a PV-RNN framework, achieves compositionality by combining language with vision, proprioception, working memory, and attention. It requires less computing power than large language models (LLMs) and makes mistakes similar to humans.
Researchers introduced a novel approach to enhance reservoir computing, incorporating a generalized readout that offers improved accuracy and robustness compared to conventional methods. The new method uses a nonlinear combination of reservoir variables to uncover deeper patterns in input data.
Researchers at the University of Bonn have developed a new training technique for highly efficient AI methods, inspired by biological neurons that use short voltage pulses to communicate. This approach enables spiking neural networks to be trained using conventional methods, resulting in improved accuracy and reduced energy consumption.
Researchers developed a cutting-edge method leveraging Graph Neural Networks (GNNs) to predict mesozooplankton community dynamics and visualize their interactions. The study achieved remarkable improvements in forecasting accuracy by integrating inter-series relationships and temporal dependencies among input-variables.
The study reveals that directional connections propagate signals in a downstream flow, leading to more complex activity patterns. Mathematical models also suggest that modularity and connectivity interact to foster dynamical complexity.
A recent study demonstrates how DNNs can predict fragrance profiles from essential oil chemical compositions, validating sensory evaluations. The model achieved high accuracy in predicting floral scents and showed promise for generating new and unique combinations.
Researchers developed a laser-based artificial neuron that emulates biological graded neuron functions, achieving a signal processing speed of 10 GBaud. This enables fast AI decision-making in time-critical applications with high accuracy.
A new study has identified a specific connectivity pattern of brain atrophy in schizophrenia, distinct from brain networks associated with other psychiatric disorders. The findings suggest that this network may be a core characteristic of schizophrenia and could inform treatment plans.
Researchers developed a new method, k* distribution method, to visualize and assess how well deep neural networks categorize related items together. The model reveals clustered, fractured or overlapping arrangements of data points, indicating accuracy and reliability issues.
Studies have shown that the lateral habenula regulates stress-related respiratory responses via the monoaminergic system, which includes dopaminergic and serotoninergic pathways. Researchers found that electrical stimulation of the LHb mimicked a stress state in rats, significantly increasing respiratory frequency.
Researchers at Pusan National University developed a hybrid model to predict metal wear in magnesium alloys, enabling safer, lighter designs. The model combines machine learning and physics to improve fatigue life prediction, offering greater predictive reliability for enhanced safety and longevity.
Researchers at the University of Tokyo discover that the patterns of spontaneous activity and stimulus-evoked response are similar in lower visual areas of the cerebral cortex but gradually become independent as one moves to higher visual areas. This orthogonal relationship helps explain how sensory perception remains stable despite co...
Researchers at Karolinska Institutet and Columbia University identified a mini-brain within the heart with its own nervous system that controls the heartbeat. This discovery challenges current views on how the heartbeat is controlled and may lead to new insights into heart diseases and treatments.
Scientists at MIT developed a fully integrated photonic processor that can perform all key computations of a deep neural network optically on the chip. The device completed machine-learning classification tasks in under half a nanosecond while achieving over 92% accuracy, similar to traditional hardware.
A new computational model called Multi-Stage Residual-BCR Net (m-rBCR) uses a unique frequency representation to solve deconvolution tasks with fewer parameters and faster processing times. The model demonstrates high performance on various microscopy datasets, outperforming traditional methods.
Researchers discovered that NMDA receptors set the baseline level for neural network activity, helping maintain stable brain function. The study's findings suggest potential innovative treatments for diseases linked to disrupted neural stability.
Researchers at Tel Aviv University found that a special protocol of hyperbaric oxygen therapy can improve the condition of PTSD sufferers, reducing typical symptoms such as flashbacks, hypervigilance, and irritability. The study showed improvements in brain connectivity and clinical symptoms, offering new hope for millions of PTSD suff...
Researchers at Linköping University have developed a new version of AlphaFold that can predict the shape of very large and complex protein structures, integrating experimental data. This breakthrough aims to improve the development of new proteins for medical drugs.
Researchers developed a novel neural network model to reconstruct 3D digital images of relief-type cultural heritage objects from old photos. The model improves the accuracy of depth estimation and soft-edge detection, enhancing the preservation of cultural heritage.
A new training algorithm called ternarized gradient BNN (TGBNN) enables learning capabilities for binarized neural networks (BNNs) on IoT edge devices. The proposed MRAM-based CiM architecture achieves faster convergence and matching accuracy with regular BNNs.
Researchers develop Knowledge-enhanced Bottlenecks (KnoBo) method to emulate human physicians' education, resulting in more accurate and interpretable AI models for medical image recognition. KnoBo-based models outperform existing best-in-class models on accuracy and robustness, especially in handling confounded data.
Researchers found that positive expectations lead to increased activity in pleasure-related brain regions, while negative expectations prime pain processing. The study suggests a dissociable impact of hedonic information, with positive expectations facilitating reward processing and negative expectations heightening anxiety.
A Princeton-led research team has built the first neuron-by-neuron and synapse-by-synapse roadmap through the brain of an adult fruit fly. The map reveals connections within the brain at every scale, enabling researchers to better understand its underlying logic and potentially develop tailored treatments for brain diseases.
A new method called Clio allows robots to make task-relevant decisions by identifying the parts of a scene that matter. In real experiments, Clio successfully mapped scenes at different levels of granularity based on natural-language prompts and enabled robots to grasp objects of interest.
A new study reveals how psilocybin alters brain connectivity to alleviate symptoms of body dysmorphic disorder, potentially aiding treatment. Psilocybin strengthens neural connections between executive functions and emotionally salient stimuli, leading to improved BDD symptoms.
Researchers at Brigham and Women's Hospital have identified a specific brain circuit that may protect against post-traumatic stress disorder (PTSD) in veterans with traumatic brain injury. The study suggests using neurostimulation therapies on this circuit could treat PTSD, offering a new potential non-invasive treatment option.
Researchers have developed a novel platform for testing radiation therapy effects on brain cells, providing insights into how the brain reacts to cancer treatment. The study aims to improve treatment strategies by understanding the impact of radiation on neural circuits, ultimately leading to personalized care for patients.
Researchers at PNNL create a uniform two-dimensional layer of silk protein fragments on graphene, enabling the design and fabrication of silk-based electronics. This biocompatible system has potential applications in wearable and implantable health sensors, as well as computing neural networks.
A Concordia-led team developed a framework that enables crowdsourced deep reinforcement learning as a service, using blockchain technology. This allows smaller organizations to access complex AI tasks previously out of reach, reducing costs and risk.
A study at the University of Bonn has revealed that fly larvae have special sensors triggered by swallowing, releasing serotonin to continue eating. This control circuit may also exist in humans and could help understand eating disorders such as anorexia or binge eating.
Researchers at Peking University developed a dual-IMC scheme to accelerate machine learning and improve energy efficiency. The new computing scheme stores both neural network weights and inputs in memory, reducing data movement and power consumption.
A new study using advanced MRI techniques has discovered a functional backbone formed by robust, delay-free interactions that serve as the core of communication in the brain. Weaker connections amplify the system's potential functional states, providing flexibility.
A deep-learning algorithm developed by astronomer David Harvey can untangle the complex signals of self-interacting dark matter and AGN feedback in galaxy cluster images. The Inception model achieved an accuracy of 80% under ideal conditions, showcasing its potential for analyzing vast amounts of space data.
A novel approach to overcome limitations of traditional methods, NeuPh uses local conditional neural fields to reconstruct high-resolution phase information from low-resolution measurements. It provides robust resolution enhancement and outperforms existing models in accuracy.
A new study reveals that more than half of strokes causing ataxia are located outside the cerebellum but affect a specific brain network. This discovery changes our understanding of ataxia's neural mechanisms and may lead to safer treatments for patients.
A new 'deep scanning' approach reveals that individuals with depression have a larger salience network in the frontal cortex and striatum, increasing their risk for depression. This finding suggests that people may be pre-wired for depression if they have this brain feature in childhood.
Researchers at the University of Birmingham have discovered that male fruit flies ignore physical threats as they pursue a female fly, driven by the neurotransmitter dopamine. As courtship advances, dopamine levels increase, blocking sensory pathways and reducing the ability to respond to threats.
Researchers used neural networks to solve fundamental equations in complex molecular systems, achieving promising results in simulating excited states of molecules. This breakthrough could lead to practical uses in materials science and chemical synthesis.
Researchers developed a new technique to study charge density waves in materials, revealing two previously unobserved ways electricity can manipulate their state. The method allows for the observation of nanoscale lengths and nanosecond speeds, with potential applications in energy-efficient microelectronics.
Researchers used advanced imaging techniques to create detailed maps of mouse brains, identifying areas vulnerable to blood vessel degeneration. These changes can lead to cognitive decline and neurodegenerative disorders like Alzheimer's disease.
A novel, fast and high-quality neural text-to-speech model was successfully developed using a Transformer encoder + ConvNeXt decoder and MS-FC-HiFi-GAN. The model can synthesize one second of speech at high speed in just 0.1 seconds using a single CPU core, achieving eight times faster synthesis than conventional methods.
A new study finds that psilocybin temporarily desynchronizes the default mode network in the brain, creating a psychedelic experience. The study provides a neurobiological explanation for the drug's effects and lays groundwork for its potential use as a therapy for mental illnesses.
Researchers have developed Nano-MIND technology, which uses magnetism to selectively activate specific deep brain neural circuits, modulating complex brain functions such as cognition and emotion. The technology has been successfully tested in animals, demonstrating its potential to regulate feeding behaviors and maternal instincts.
Researchers found that small regions of the brain can momentarily 'flicker' awake while the rest of the brain remains asleep, and vice versa from wake to sleep. This challenges traditional understanding of sleep and wake patterns, which have been defined by slow, long-lasting waves.
Scientists from HZDR, TU Chemnitz, TU Dresden, and Forschungszentrum Jülich have demonstrated the storage of entire bit sequences in cylindrical domains. The team's findings could lead to novel types of data storage and sensors, including magnetic variants of neural networks.
Researchers discovered that propofol, a commonly used anesthesia drug, induces unconsciousness by causing the brain to become increasingly unstable. This instability leads to chaotic brain activity, resulting in loss of consciousness. The study's findings could help develop better tools for monitoring patients during general anesthesia.
A study published in Nature Medicine found that off-the-shelf wearable trackers can monitor the response to two treatments for atrial fibrillation and heart failure. The devices provided clinically useful information similar to in-person hospital assessments, with a neural network helping to analyze missing data.
Researchers at Max Planck Institute propose a new method for implementing neural networks with optical systems, which could lead to faster and more energy-efficient alternatives. The approach allows for parallel computations in high speeds limited by the speed of light, and can be applied to various physically different systems.
A recent study found that the cerebellum plays a crucial role in regulating thirst, with the hormone asprosin activating Purkinje neurons to enhance water intake. This discovery has significant implications for managing thirst disorders such as polydipsia and hypodipsia, for which current treatments are scarce.