There has long been a problem of “weak supervision” in the field of artificial intelligence, where only part of the labels are observable in the training data, while the remaining labels are unknown. Multiple Instance Learning (MIL) is a paradigm to address this issue. In MIL, the training data is divided into multiple “bags”, each containing multi...

Traditional representation learning methods usually design a fixed basis function architecture to achieve desired properties such as orthogonality and sparsity. In contrast, the idea of deep reinforcement learning is that the designer should not encode the properties of the representation, but instead let the data flow determine the properties of t...

In model-based fault diagnosis, the ability to identify the key observational data that leads to system abnormalities is highly valuable. This paper introduces a framework and algorithm for identifying key observational data. The framework determines which observations are crucial for diagnosis by abstracting the raw observational data into “sub-ob...

With the widespread application of graph-structured data in various fields, Graph Neural Networks (GNNs) have attracted significant attention as a powerful tool for analyzing graph data. However, existing GNNs primarily rely on neighborhood node similarity information when learning node representations, overlooking the potential of node differences...

Circuit Mechanisms of Hue Selectivity in the Fruit Fly Visual System Color perception is a key aspect of visual experience, playing an important role in the interaction between organisms and their environment. In primates and other animals with three types of photoreceptor cells, neurons in the visual cortex have been found to selectively respond t...

This paper reports on research investigating how zebra finches recover their complex learned behaviors following large-scale neuronal perturbations. The researchers selectively disrupted the activity of the projection neurons in the HVC (hyperpallium ventralis) region critical for song sequence generation in zebra finches using genetic tools, leadi...

Brain Clearance Mechanism Suppressed During Sleep and Anesthesia Background Clearing metabolic wastes and toxic substances from the brain is a crucial process for maintaining a healthy nervous system. However, the specific mechanism of this process has been debated. A widely discussed viewpoint suggests that the sleep state promotes the function of...

Obsessive-compulsive behaviors have long been associated with overactivation of the striatum. The GABA-ergic parvalbumin-positive interneurons (PVIs) in the striatum play a crucial role in regulating striatal activity and inhibiting inappropriate motivated behaviors. To investigate the potential role of striatal PVIs in regulating obsessive-compuls...

This paper reports a study on the cognitive representation of intracranial self-stimulation (ICSS) of midbrain dopamine neurons. The authors aimed to answer why rats work to obtain ICSS, i.e., how the pulsatile activation of dopamine neurons is represented in the brain. Research Background: The pulsatile activation of dopamine neurons is typically ...

I. Research Background and Purpose The subcommissural organ (SCO) is a glandular structure located at the entrance of the ventricular system, which is present in humans and other vertebrates, but its function has not been fully elucidated. Previous studies have suggested that the SCO may be involved in the regulation of neural development and the m...