Articles tagged with Learning Processes
A new theorem limits the use of quantum machine learning to learn processes like black holes, highlighting its fundamental limits. However, most physical processes are structured enough that they don't resemble random ones, making the impact less severe.
Assistant Professor Aaron Maxwell uses big data analytics and deep learning to map West Virginia's surface over the last 60 years. The research aims to understand natural landscapes and predict future changes such as landslides and river erosion.
A collaborative study identified strengths and challenges autistic teens face while learning to drive. Researchers recommend scaffolded instruction and caregiver-supervised practice to overcome obstacles. Autistic drivers show similar crash rates to non-autistic peers, with lower license suspension rates.
Researchers from Nagoya University developed a two-step labeling technique to analyze the dynamics of neurotransmitter receptor proteins. The study revealed key findings on AMPA-type glutamate receptors, including their localization at synapses and half-life in neurons.
New research from Carnegie Mellon University and the University of Pittsburgh examines how changes in internal states like engagement affect learning. The study found that fluctuations in neural activity are influenced by shifts in internal states, suggesting a key role for engagement in learning.
A team of scientists from GIST developed an AI-based strategy combining diffuse reflectance spectroscopy with deep learning to detect spoiled meat. The technique accurately classifies beef samples in seconds, offering a fast, affordable, and non-invasive solution for checking freshness.
Researchers developed microswimmers that can change direction by heating tiny gold particles, then learned to navigate through a virtual environment via external control and virtual rewards. The findings suggest an optimal speed is key to navigation, with implications for autonomous tasks and collective behavior in biological systems.
A team of researchers from UNIST developed a new learning method for PCM-based memristor neural networks, improving their learning ability by about 3% in handwriting classification tasks. The approach leverages the 'resistance drift' property of phase-change memory to update synapses and associate patterns with data.
A graduate student used a class assignment to apply linear algebra techniques to analyze flow field data, identifying dominant modes that capture the most important characteristics of the flow. This technique, called proper orthogonal decomposition, allows for efficient study of unsteady processes with minimal noise and uncertainty.
A new method developed by Salk researchers enables scientists to improve microscope image quality using artificial intelligence and artificially degraded images. The 'crappifier' tool can help democratize microscopy, making it accessible to those without powerful microscopes.
A study published in PNAS found that theta oscillations in the human brain, generated by the hippocampus, play a crucial role in active learning and memory formation. Active navigation improves learning efficiency, while two consecutive phenomena, encoding and retrieval, are facilitated by theta phase code.
Researchers developed a machine learning-powered platform, OC-SMART, to process ocean color in satellite images, solving a 30-year-old problem in retrieving data from coastal regions and open ocean areas. The tool provides useful data on chlorophyll concentrations and phytoplankton presence, helping to understand the ocean's state.
A recent paper from McGill University suggests that a one-size-fits-all approach to sustainability is not feasible due to regional complexities. The researchers propose envisioning diverse desirable futures, nurturing seeds of sustainability, and navigating emerging pathways to create a more sustainable future. This approach acknowledg...
A new framework combines DRL and curriculum learning to enable mobile robots to navigate quickly while avoiding collisions. The framework is proven to achieve the highest success rate in training and validation processes, making it valuable for various fields requiring fast but safe navigation.
Researchers developed photonic processors to process information rapidly and in parallel, enabling complex mathematical tasks at fast speeds. These chips use wavelength multiplexing for highly parallel data processing.
Developing a new era of optical signal processing, researchers created an optical convolutional neural network accelerator capable of processing large amounts of information per second. This innovation harnesses the massive parallelism of light to outperform top-of-the-line graphics processing units by over one order of magnitude.
A study published in Science found that the perirhinal cortex accumulates information from multiple senses and sends it back to the rest of the cortex, enabling coordinated learning. Without this connection, animals struggle to retain information, resembling patients with anterograde amnesia like H.M.
The Parker Solar Probe has discovered that the densities of dust particles around the sun vary wildly over months, contradicting scientists' expectations. This finding suggests a more complex and fast-shifting dusty environment than previously thought, with implications for understanding planet formation.
By analyzing crystal orientations from a 1959 Kilauea eruption, researchers inferred magma flow patterns prior to eruption, shedding light on past and potential future eruptions. The study's findings offer new insights into conduit dynamics, aiding in hazard prediction and monitoring.
A protein called CatSper1 acts as a molecular 'barcode', helping determine which sperm cells survive to fertilize an egg. The study's findings may enable new approaches to treating infertility by understanding the selection process in the female reproductive tract.
Flies sleep longer and are less responsive to light pulses during vibration, suggesting they accumulate 'sleep credit.' The ability to go to sleep improves with repeated exposure, implying habituation. Vibration-induced sleep serves vital functions and may offer insights into human sleep regulation.
Adults can acquire a complex language through cross-situational statistical learning, solving the chicken-and-egg problem of word meaning and sentence structure. By keeping track of associations between words and scenes across hundreds of trials, learners can figure out word meanings and roles in sentences.
The University of the Basque Country has developed a system using natural language processing techniques to represent sensors and human activity in smart environments. By giving sensors meaningful names, the algorithm can transfer knowledge from one environment to another, overcoming a major challenge in smart homes.
A study on language aptitude identified key components of language acquisition as cognitive, emotional, and personality-related characteristics. Effective language learning depends on thought process, emotional experience, motivation, and personal traits such as self-esteem and anxiety levels.
A Rutgers University study examines how high-income countries regulate health care costs using a fee-for-service model, finding that structuring negotiations and setting standardized prices can lower US healthcare spending. The study highlights the need for policymakers to address healthcare prices, not just volume of services.
Researchers from the University of Tsukuba found that praising participants with robots and virtual agents improved offline motor task performance. Two agents outperformed one agent in terms of skill consolidation.
Researchers study brain mechanisms in mice using a custom-made puzzle task to disentangle cognitive schemes. They discover that the anterior cingulate cortex (ACC) creates a neural map representing various aspects of behavior, including current location and predicted consequences of actions.
The article discusses the benefits of combining intensive lab experiences with online learning for students studying ecology and evolution. A short but intensive field period can enhance student interpersonal skills, offer more flexibility, and provide a better testing experience.
Researchers trained an AI to rapidly assess post-disaster building damage using convolutional neural network (CNN) technology. The model achieved accuracy rates of approximately 94% in classifying building damage levels, making it a valuable tool for crisis responders.
A recent study published in Neuron reveals that the brain's signal transmission is regulated by the environment, with certain processes allowing for more open communication between neurons. The researchers found that astrocytes, specialized cells in the brain, play a crucial role in shielding communication to some extent.
The University of Göttingen and the Lower Saxony State Service for Cultural Heritage are conducting a research project to analyze the oldest wooden objects from the Schöningen site. The spears, dated at 300,000 years old, provide valuable insights into the technical and intellectual abilities of early European hominins.
A new paper suggests that COVID-19 policy makers can improve transparency and accountability by learning from the aviation industry. The research emphasizes the importance of assurance cases to explain model assumptions and their limitations, enabling policymakers to critically review decisions.
The LEARNER project features a mixed-reality learning environment with physical, augmented, and virtual reality components to augment responders' capabilities. The platform aims to increase the speed of integration and adoption of human augmentation technologies in emergency response and other industrial sectors.
Auralee Edelen's work uses machine learning to streamline particle accelerator operations, while Wai Ling Wu explores mysteries in astrophysics and cosmology. The Panofsky Fellowship provides funding for five years of research.
A research team developed an AI system that recommends process conditions for plastic injection molding, enabling the creation of various shapes in real-time. The system achieved an average relative error of 0.66% and can be used by non-experts to produce uniform results.
Researchers identified nine clusters of journalists, including an 'elite/legacy' cluster dominated by The Washington Post and others, which showed high levels of insularity. In contrast, a long-form/enterprise cluster fostered exchange among journalists with diverse perspectives.
Researchers at Ruhr-University Bochum used EEG to study the impact of repetitive tactile stimulation on brain activity. They found that neuronal responses adapted to the frequency of stimulation and showed changes over time, potentially illustrating a learning process.
New research reveals that indigenous Tsimane' people, American adults, preschoolers, and macaque monkeys exhibit similar recursion skills, indicating a shared cognitive process for conveying complex commands and ideas.
Liang Zhao at George Mason University has been awarded a $102,873 NSF CAREER Award to develop transformative frameworks for spatial network generative modeling. The project aims to learn complex generation processes from massive datasets and create more interpretable models.
Researchers at Stevens Institute of Technology reveal how bombardier beetles biosynthesize chemicals to create fuel for their explosions. The study shows the beetles' biochemistry is intricate, with toxic chemicals derived from coal tar and benzene-like compounds metabolized from hydroquinone.
A new study found that ability-based classes severely affect pupils' self-confidence, particularly in maths. The research suggests that students in top sets have higher self-confidence in maths and English, while those in bottom sets have lower self-confidence, widening the gap over time.
New modern school spaces need active negotiations between all stakeholders, including teachers and students, to be transformed into learning environments. This process involves cyclic processes of interaction and meaning-making, which can lead to the development of authentic and functional learning environments.
Stroke survivors who receive early visual training can recover more of their lost vision than those trained later. Researchers discovered that the brain's visual structures degenerate over time after an occipital stroke, but early intervention can halt this loss.
Carnegie Mellon University researchers have created a method to build accurate 3D reconstructions of faces from smartphone videos, outperforming other camera-based processes with sub-millimeter accuracy. The digital face might be used in gaming, virtual or augmented reality, animation, biometric identification, and medical procedures.
A study found that Russian social media platforms like Facebook and VKontakte exhibit characteristics of McDonaldization, a form of rationality that prioritizes efficiency and profit over human values. This process can lead to the displacement of individuals from meaningful relationships and the loss of personal autonomy.
The Learning to Synthesize (LS-DNN) approach splits input signals into low and high spatial frequency bands, enabling deep neural networks to process and synthesize them. The algorithm is robust in handling noisy intensity signals, making it suitable for applications like x-rays and sonograms.
Researchers developed a more efficient automated parking guidance control strategy that mimics the approach to parallel parking commonly used by human drivers. This new method simplifies control rules and strategies, reducing computing and storage resources required in vehicles.
A team from Deakin University in Australia developed an improved sight-correcting system for self-driving vehicles. By watching human operators complete tasks, the vehicles can learn to make decisions based on visual information, reducing the need for extensive training data.
A new study by Northwestern University found that a simulation-based training program can help medical students develop the skills to break bad news to patients in a clear and compassionate way. The training, which uses a mastery learning approach, has been shown to improve patient outcomes and increase student satisfaction.
Army researchers developed a new algorithm that enables collaborative and communication-efficient deep learning, reducing the need for centralized data pooling. The algorithm decreases communication overhead by up to 70% without sacrificing performance accuracy or learning rate.
Researchers found that neurons in the hippocampus become more synchronized when a memory is formed, enabling animals to form and retrieve episodic knowledge. This finding provides clues for the development of future therapies for diseases like dementia, autism, and PTSD.
Effective leadership is driven by four distinct mindsets: growth, learning, deliberative, and promotion. Leaders with these mindsets are more likely to engage, navigate change, and lead effectively. Prioritizing mindset development can significantly impact leadership effectiveness.
A role-playing game called Citizenship Quest helped students build trust and empathize with immigrants from China, India, and Mexico. The exercise increased understanding among students who had little direct contact with immigrants.
Researchers at KAIST discovered a computational and neural mechanism for human meta reinforcement learning, enabling the porting of key human intelligence elements into AI algorithms. The study highlights how humans adapt to complexity and uncertainty in decision-making.
Researchers developed an autonomous machine learning-based method to determine crystal structure from EBSD data with high accuracy. The platform enables high-throughput evaluation of material properties, opening doors for rapid discovery and analysis.
A high-resolution catalog of earthquakes from the 2019 Ridgecrest sequence reveals complex rupturing and crosscutting fault structures. The catalog helps researchers understand the triggering and evolution of the sequence, shedding light on the physics and processes involved.
Researchers from UMA developed a method using deep learning to increase brain image quality and detection of pathologies with increased accuracy. This breakthrough technique enables specialists to identify brain-related disorders more accurately and definitionally, reducing the need for additional tests.
Researchers analyzed how juvenile male zebra finches learn to sing using a novel framework that revealed multilayered changes in brain wiring. Good songs improve slowly over time, while bad ones change quickly but are often forgotten the next day.
Scientists at Max Planck Florida Institute for Neuroscience have developed novel biosensors to study CREB dynamics in living brains. They found that sensory experiences shape CREB activity over hours to days after a stimulus, and this effect can be observed even when visual stimuli are absent.
Researchers have found that the cerebral cortex can directly control the activities of globus pallidus neurons in a manner that depends on the type of neuron. This discovery could improve our understanding of why certain symptoms occur in movement and learning behavioural disorders, such as Parkinson's disease.