Search

Abstract The phenomenon of semantic satiation, which refers to the loss of meaning of a word or phrase after being repeated many times, is a well-known psychological phenomenon. However, the microscopic neural computational principles responsible for these mechanisms remain unknown. In this study, we use a deep learning model of continuous coupled neural networks to investigate the mechanism underlying semantic satiation and precisely describe this process with neuronal components. Our results suggest that, from a mesoscopic perspective, semantic satiation may be a bottom-up process. Unlike existing macroscopic psychological studies that suggest that semantic satiation is a top-down process, our simulations use a similar experimental paradigm as classical psychology experiments and observe similar results. Satiation of semantic objectives, similar to the learning process of our network model used for object recognition, relies on continuous learning and switching between objects. The underlying neural coupling strengthens or weakens satiation. Taken together, both neural and network mechanisms play a role in controlling semantic satiation.

In digital cameras, we find a major limitation: the image and video form inherited from a film camera obstructs it from capturing the rapidly changing photonic world. Here, we present vform, a bit sequence array where each bit represents whether the accumulation of photons has reached a threshold, to record and reconstruct the scene radiance at any moment. By employing only consumer-level complementary metal–oxide semiconductor (CMOS) sensors and integrated circuits, we have developed a spike camera that is 1000× faster than conventional cameras. By treating vform as spike trains in biological vision, we have further developed a spiking neural network (SNN)-based machine vision system that combines the speed of the machine and the mechanism of biological vision, achieving high-speed object detection and tracking 1000× faster than human vision. We demonstrate the utility of the spike camera and the super vision system in an assistant referee and target pointing system. Our study is expected to fundamentally revolutionize the image and video concepts and related industries, including photography, movies, and visual media, and to unseal a new SNN-enabled speed-free machine vision era.

Summary: Finding the complete functional circuits of neurons is a challenging problem in brain research. Here, we present a protocol, based on visual stimuli and spikes, for obtaining the complete circuit of recorded neurons using spike-triggered nonnegative matrix factorization. We describe steps for data preprocessing, inferring the spatial receptive field of the subunits, and analyzing the module matrix. This approach identifies computational components of the feedforward network of retinal ganglion cells and dissects the network structure based on natural image stimuli.For complete details on the use and execution of this protocol, please refer to Jia et al. (2021).1 : Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Neurons, represented as a tree structure of morphology, have various distinguished branches of dendrites. Different types of synaptic receptors distributed over dendrites are responsible for receiving inputs from other neurons. NMDA receptors (NMDARs) are expressed as excitatory units, and play a key physiological role in synaptic function. Although NMDARs are widely expressed in most types of neurons, they play a different role in the cerebellar Purkinje cells (PCs). Utilizing a computational PC model with detailed dendritic morphology, we explored the role of NMDARs at different parts of dendritic branches and regions. We found somatic responses can switch from silent, to simple spikes and complex spikes, depending on specific dendritic branches. Detailed examination of the dendrites regarding their diameters and distance to soma revealed diverse response patterns, yet explain two firing modes, simple and complex spike. Taken together, these results suggest that NMDARs play an important role in controlling excitability sensitivity while taking into account the factor of dendritic properties. Given the complexity of neural morphology varying in cell types, our work suggests that the functional role of NMDARs is not stereotyped but highly interwoven with local properties of neuronal structure.

A neuroprosthesis is a type of precision medical device that is intended to manipulate the neuronal signals of the brain in a closed-loop fashion, while simultaneously receiving stimuli from the environment and controlling some part of a human brain or body. Incoming visual information can be processed by the brain in millisecond intervals. The retina computes visual scenes and sends its output to the cortex in the form of neuronal spikes for further computation. Thus, the neuronal signal of interest for a retinal neuroprosthesis is the neuronal spike. Closed-loop computation in a neuroprosthesis includes two stages: encoding a stimulus as a neuronal signal, and decoding it back into a stimulus. In this paper, we review some of the recent progress that has been achieved in visual computation models that use spikes to analyze natural scenes that include static images and dynamic videos. We hypothesize that in order to obtain a better understanding of the computational principles in the retina, a hypercircuit view of the retina is necessary, in which the different functional network motifs that have been revealed in the cortex neuronal network are taken into consideration when interacting with the retina. The different building blocks of the retina, which include a diversity of cell types and synaptic connections—both chemical synapses and electrical synapses (gap junctions)—make the retina an ideal neuronal network for adapting the computational techniques that have been developed in artificial intelligence to model the encoding and decoding of visual scenes. An overall systems approach to visual computation with neuronal spikes is necessary in order to advance the next generation of retinal neuroprosthesis as an artificial visual system.

Finding out the physical structure of neuronal circuits that governs neuronal responses is an important goal for brain research. With fast advances for large-scale recording techniques, identification of a neuronal circuit with multiple neurons and stages or layers becomes possible and highly demanding. Although methods for mapping the connection structure of circuits have been greatly developed in recent years, they are mostly limited to simple scenarios of a few neurons in a pairwise fashion; and dissecting dynamical circuits, particularly mapping out a complete functional circuit that converges to a single neuron, is still a challenging question. Here, we show that a recent method, termed spike-triggered non-negative matrix factorization (STNMF), can address these issues. By simulating different scenarios of spiking neural networks with various connections between neurons and stages, we demonstrate that STNMF is a persuasive method to dissect functional connections within a circuit. Using spiking activities recorded at neurons of the output layer, STNMF can obtain a complete circuit consisting of all cascade computational components of presynaptic neurons, as well as their spiking activities. For simulated simple and complex cells of the primary visual cortex, STNMF allows us to dissect the pathway of visual computation. Taken together, these results suggest that STNMF could provide a useful approach for investigating neuronal systems leveraging recorded functional neuronal activity.

Skeleton-based 3D object retrieval is a very efficient method to query the sketch databases in numerous applications. However, few skeleton images are found so far in existing sketch benchmarks. In this paper, we provide an initial benchmark dataset consisting of skeleton sketches, including hand-drawn skeletons and skeletons extracted from 3D objects, and both of them are used to form a generic object class. Then we present a method for skeleton-based 3D object retrieval using a retina-like feature descriptor (S3DOR-RFD) based on the structural property of the human retina for processing complex visual information in a very efficient way. As part of the S3DOR-RFD algorithm, we combine artificial bee colony (ABC) in support vector machine (SVM) so as to improve the performance with automatic parameter selection, where one can make full use of the advantages of ABC and SVM to further improve the accuracy rate of 3D object retrieval. Experimental results indicate that skeleton sketches can be automatically distinguished from perspective sketches, and that the proposed S3DOR-RFD method works efficiently for selected object classes.

Numerous experimental studies suggest that noise is inherent in the human brain. However, the functional importance of noise remains unknown. n particular, from a computational perspective, such stochasticity is potentially harmful to brain function. In machine learning, a large number of saddle points are surrounded by high error plateaus and give the illusion of the existence of local minimum. As a result, being trapped in the saddle points can dramatically impair learning and adding noise will attack such saddle point problems in high-dimensional optimization, especially under the strict saddle condition. Motivated by these arguments, we propose one biologically plausible noise structure and demonstrate that noise can efficiently improve the optimization performance of spiking neural networks based on stochastic gradient descent. The strict saddle condition for synaptic plasticity is deduced, and under such conditions, noise can help optimization escape from saddle points on high dimensional domains. The theoretical results explain the stochasticity of synapses and guide us on how to make use of noise. In addition, we provide biological interpretations of proposed noise structures from two points: one based on the free energy principle in neuroscience and another based on observations of in vivo experiments. Our simulation results manifest that in the learning and test phase, the accuracy of synaptic sampling with noise is almost 20% higher than that without noise for synthesis dataset, and the gain in accuracy with/without noise is at least 10% for the MNIST and CIFAR-10 dataset. Our study provides a new learning framework for the brain and sheds new light on deep noisy spiking neural networks.

The majority of neurons in the neuronal system of the brain have a complex morphological structure, which diversifies the dynamics of neurons. In the granular layer of the cerebellum, there exists a unique cell type, the unipolar brush cell (UBC), that serves as an important relay cell for transferring information from outside mossy fibers to downstream granule cells. The distinguishing feature of the UBC is that it has a simple morphology, with only one short dendritic brush connected to its soma. Based on experimental evidence showing that UBCs exhibit a variety of dynamic behaviors, here we develop two simple models, one with a few detailed ion channels for simulation and the other one as a two-variable dynamical system for theoretical analysis, to characterize the intrinsic dynamics of UBCs. The reasonable values of the key channel parameters of the models can be determined by analysis of the stability of the resting membrane potential and the rebound firing properties of UBCs. Considered together with a large variety of synaptic dynamics installed on UBCs, we show that the simple-structured UBCs, as relay cells, can extend the range of dynamics and information from input mossy fibers to granule cells with low-frequency resonance and transfer stereotyped inputs to diverse amplitudes and phases of the output for downstream granule cells. These results suggest that neuronal computation, embedded within intrinsic ion channels and the diverse synaptic properties of single neurons without sophisticated morphology, can shape a large variety of dynamic behaviors to enhance the computational ability of local neuronal circuits.

Traffic sensing is one of the promising applications to guarantee safe and efficient traffic systems in vehicular networks. However, due to the unique characteristics of vehicular networks, such as limited wireless bandwidth and dynamic mobility of vehicles, traffic sensing always faces high estimation error based on collected traffic data with missing elements and over-high communication cost between terminal users and central server. Hence, this paper investigates the traffic sensing system in vehicular networks with mobile edge computing (MEC), where each MEC server enables traffic data collection and recovery in its local server. On this basis, we formulate the bandwidth-constrained traffic sensing (BCTS) problem, aiming at minimizing the estimation error based on the collected traffic data. To tackle the BCTS problem, we first propose the bandwidth-aware data collection (BDC) algorithm to select the optimal uploaded traffic data by evaluating the priority of each road segment covered by the MEC server. Then, we propose the convex-based data recovery (CDR) algorithm to minimize estimation error by transforming the BCTS into an l 2 -norm minimization problem. Last but not the least, we implement the simulation model and conduct performance evaluation. The comprehensive simulation results verify the superiority of the proposed algorithm.