arXiv subjects

Zhaofei Yu

Publications and source records attributed to Zhaofei Yu.

83 records · Page 5Linked to original sources

arXiv2021

Pruning of Deep Spiking Neural Networks through Gradient Rewiring

Spiking Neural Networks (SNNs) have been attached great importance due to their biological plausibility and high energy-efficiency on neuromorphic chips. As these chips are usually resource-constrained, the compression of SNNs is thus crucial along the road of practical use of SNNs. Most existing methods directly apply pruning approaches in artificial neural networks (ANNs) to SNNs, which ignore the difference between ANNs and SNNs, thus limiting the performance of the pruned SNNs. Besides, these methods are only suitable for shallow SNNs. In this paper, inspired by synaptogenesis and synapse elimination in the neural system, we propose gradient rewiring (Grad R), a joint learning algorithm of connectivity and weight for SNNs, that enables us to seamlessly optimize network structure without retraining. Our key innovation is to redefine the gradient to a new synaptic parameter, allowing better exploration of network structures by taking full advantage of the competition between pruning and regrowth of connections. The experimental results show that the proposed method achieves minimal loss of SNNs' performance on MNIST and CIFAR-10 dataset so far. Moreover, it reaches a $\sim$3.5% accuracy loss under unprecedented 0.73% connectivity, which reveals remarkable structure refining capability in SNNs. Our work suggests that there exists extremely high redundancy in deep SNNs. Our codes are available at https://github.com/Yanqi-Chen/Gradient-Rewiring.

cs.NE↗

arXiv2021

Optimal ANN-SNN Conversion for Fast and Accurate Inference in Deep Spiking Neural Networks

Spiking Neural Networks (SNNs), as bio-inspired energy-efficient neural networks, have attracted great attentions from researchers and industry. The most efficient way to train deep SNNs is through ANN-SNN conversion. However, the conversion usually suffers from accuracy loss and long inference time, which impede the practical application of SNN. In this paper, we theoretically analyze ANN-SNN conversion and derive sufficient conditions of the optimal conversion. To better correlate ANN-SNN and get greater accuracy, we propose Rate Norm Layer to replace the ReLU activation function in source ANN training, enabling direct conversion from a trained ANN to an SNN. Moreover, we propose an optimal fit curve to quantify the fit between the activation value of source ANN and the actual firing rate of target SNN. We show that the inference time can be reduced by optimizing the upper bound of the fit curve in the revised ANN to achieve fast inference. Our theory can explain the existing work on fast reasoning and get better results. The experimental results show that the proposed method achieves near loss less conversion with VGG-16, PreActResNet-18, and deeper structures. Moreover, it can reach 8.6x faster reasoning performance under 0.265x energy consumption of the typical method. The code is available at https://github.com/DingJianhao/OptSNNConvertion-RNL-RIL.

cs.NE↗

arXiv2020

Probabilistic Inference of Binary Markov Random Fields in Spiking Neural Networks through Mean-field Approximation

Recent studies have suggested that the cognitive process of the human brain is realized as probabilistic inference and can be further modeled by probabilistic graphical models like Markov random fields. Nevertheless, it remains unclear how probabilistic inference can be implemented by a network of spiking neurons in the brain. Previous studies have tried to relate the inference equation of binary Markov random fields to the dynamic equation of spiking neural networks through belief propagation algorithm and reparameterization, but they are valid only for Markov random fields with limited network structure. In this paper, we propose a spiking neural network model that can implement inference of arbitrary binary Markov random fields. Specifically, we design a spiking recurrent neural network and prove that its neuronal dynamics are mathematically equivalent to the inference process of Markov random fields by adopting mean-field theory. Furthermore, our mean-field approach unifies previous works. Theoretical analysis and experimental results, together with the application to image denoising, demonstrate that our proposed spiking neural network can get comparable results to that of mean-field inference.

q-bio.NC↗

arXiv2020

Neural System Identification with Spike-triggered Non-negative Matrix Factorization

Neuronal circuits formed in the brain are complex with intricate connection patterns. Such complexity is also observed in the retina as a relatively simple neuronal circuit. A retinal ganglion cell receives excitatory inputs from neurons in previous layers as driving forces to fire spikes. Analytical methods are required that can decipher these components in a systematic manner. Recently a method termed spike-triggered non-negative matrix factorization (STNMF) has been proposed for this purpose. In this study, we extend the scope of the STNMF method. By using the retinal ganglion cell as a model system, we show that STNMF can detect various computational properties of upstream bipolar cells, including spatial receptive field, temporal filter, and transfer nonlinearity. In addition, we recover synaptic connection strengths from the weight matrix of STNMF. Furthermore, we show that STNMF can separate spikes of a ganglion cell into a few subsets of spikes where each subset is contributed by one presynaptic bipolar cell. Taken together, these results corroborate that STNMF is a useful method for deciphering the structure of neuronal circuits.

q-bio.NC↗

arXiv2020

Revealing Fine Structures of the Retinal Receptive Field by Deep Learning Networks

Deep convolutional neural networks (CNNs) have demonstrated impressive performance on many visual tasks. Recently, they became useful models for the visual system in neuroscience. However, it is still not clear what are learned by CNNs in terms of neuronal circuits. When a deep CNN with many layers is used for the visual system, it is not easy to compare the structure components of CNNs with possible neuroscience underpinnings due to highly complex circuits from the retina to higher visual cortex. Here we address this issue by focusing on single retinal ganglion cells with biophysical models and recording data from animals. By training CNNs with white noise images to predict neuronal responses, we found that fine structures of the retinal receptive field can be revealed. Specifically, convolutional filters learned are resembling biological components of the retinal circuit. This suggests that a CNN learning from one single retinal cell reveals a minimal neural network carried out in this cell. Furthermore, when CNNs learned from different cells are transferred between cells, there is a diversity of transfer learning performance, which indicates that CNNs are cell-specific. Moreover, when CNNs are transferred between different types of input images, here white noise v.s. natural images, transfer learning shows a good performance, which implies that CNNs indeed capture the full computational ability of a single retinal cell for different inputs. Taken together, these results suggest that CNNs could be used to reveal structure components of neuronal circuits, and provide a powerful model for neural system identification.

q-bio.NC↗

arXiv2020

Reconstruction of Natural Visual Scenes from Neural Spikes with Deep Neural Networks

Neural coding is one of the central questions in systems neuroscience for understanding how the brain processes stimulus from the environment, moreover, it is also a cornerstone for designing algorithms of brain-machine interface, where decoding incoming stimulus is highly demanded for better performance of physical devices. Traditionally researchers have focused on functional magnetic resonance imaging (fMRI) data as the neural signals of interest for decoding visual scenes. However, our visual perception operates in a fast time scale of millisecond in terms of an event termed neural spike. There are few studies of decoding by using spikes. Here we fulfill this aim by developing a novel decoding framework based on deep neural networks, named spike-image decoder (SID), for reconstructing natural visual scenes, including static images and dynamic videos, from experimentally recorded spikes of a population of retinal ganglion cells. The SID is an end-to-end decoder with one end as neural spikes and the other end as images, which can be trained directly such that visual scenes are reconstructed from spikes in a highly accurate fashion. Our SID also outperforms on the reconstruction of visual stimulus compared to existing fMRI decoding models. In addition, with the aid of a spike encoder, we show that SID can be generalized to arbitrary visual scenes by using the image datasets of MNIST, CIFAR10, and CIFAR100. Furthermore, with a pre-trained SID, one can decode any dynamic videos to achieve real-time encoding and decoding of visual scenes by spikes. Altogether, our results shed new light on neuromorphic computing for artificial visual systems, such as event-based visual cameras and visual neuroprostheses.

q-bio.NC↗

arXiv2020

Towards the Next Generation of Retinal Neuroprosthesis: Visual Computation with Spikes

Neuroprosthesis, as one type of precision medicine device, is aiming for manipulating neuronal signals of the brain in a closed-loop fashion, together with receiving stimulus from the environment and controlling some part of our brain/body. In terms of vision, incoming information can be processed by the brain in millisecond interval. The retina computes visual scenes and then sends its output as neuronal spikes to the cortex for further computation. Therefore, the neuronal signal of interest for retinal neuroprosthesis is spike. Closed-loop computation in neuroprosthesis includes two stages: encoding stimulus to neuronal signal, and decoding it into stimulus. Here we review some of the recent progress about visual computation models that use spikes for analyzing natural scenes, including static images and dynamic movies. We hypothesize that for a better understanding of computational principles in the retina, one needs a hypercircuit view of the retina, in which different functional network motifs revealed in the cortex neuronal network should be taken into consideration for the retina. Different building blocks of the retina, including a diversity of cell types and synaptic connections, either chemical synapses or electrical synapses (gap junctions), make the retina an ideal neuronal network to adapt the computational techniques developed in artificial intelligence for modeling of encoding/decoding visual scenes. Altogether, one needs a systems approach of visual computation with spikes to advance the next generation of retinal neuroprosthesis as an artificial visual system.

q-bio.NC↗

arXiv2018

Winner-Take-All as Basic Probabilistic Inference Unit of Neuronal Circuits

Experimental observations of neuroscience suggest that the brain is working a probabilistic way when computing information with uncertainty. This processing could be modeled as Bayesian inference. However, it remains unclear how Bayesian inference could be implemented at the level of neuronal circuits of the brain. In this study, we propose a novel general-purpose neural implementation of probabilistic inference based on a ubiquitous network of cortical microcircuits, termed winner-take-all (WTA) circuit. We show that each WTA circuit could encode the distribution of states defined on a variable. By connecting multiple WTA circuits together, the joint distribution can be represented for arbitrary probabilistic graphical models. Moreover, we prove that the neural dynamics of WTA circuit is able to implement one of the most powerful inference methods in probabilistic graphical models, mean-field inference. We show that the synaptic drive of each spiking neuron in the WTA circuit encodes the marginal probability of the variable in each state, and the firing probability (or firing rate) of each neuron is proportional to the marginal probability. Theoretical analysis and experimental results demonstrate that the WTA circuits can get comparable inference result as mean-field approximation. Taken together, our results suggest that the WTA circuit could be seen as the minimal inference unit of neuronal circuits.

q-bio.NC↗

arXiv2018

CaMKII activation supports reward-based neural network optimization through Hamiltonian sampling

Synaptic plasticity is implemented and controlled through over thousand different types of molecules in the postsynaptic density and presynaptic boutons that assume a staggering array of different states through phosporylation and other mechanisms. One of the most prominent molecule in the postsynaptic density is CaMKII, that is described in molecular biology as a "memory molecule" that can integrate through auto-phosporylation Ca-influx signals on a relatively large time scale of dozens of seconds. The functional impact of this memory mechanism is largely unknown. We show that the experimental data on the specific role of CaMKII activation in dopamine-gated spine consolidation suggest a general functional role in speeding up reward-guided search for network configurations that maximize reward expectation. Our theoretical analysis shows that stochastic search could in principle even attain optimal network configurations by emulating one of the most well-known nonlinear optimization methods, simulated annealing. But this optimization is usually impeded by slowness of stochastic search at a given temperature. We propose that CaMKII contributes a momentum term that substantially speeds up this search. In particular, it allows the network to overcome saddle points of the fitness function. The resulting improved stochastic policy search can be understood on a more abstract level as Hamiltonian sampling, which is known to be one of the most efficient stochastic search methods.

cs.NE↗

arXiv2017

Revealing structure components of the retina by deep learning networks

Deep convolutional neural networks (CNNs) have demonstrated impressive performance on visual object classification tasks. In addition, it is a useful model for predication of neuronal responses recorded in visual system. However, there is still no clear understanding of what CNNs learn in terms of visual neuronal circuits. Visualizing CNN's features to obtain possible connections to neuronscience underpinnings is not easy due to highly complex circuits from the retina to higher visual cortex. Here we address this issue by focusing on single retinal ganglion cells with a simple model and electrophysiological recordings from salamanders. By training CNNs with white noise images to predicate neural responses, we found that convolutional filters learned in the end are resembling to biological components of the retinal circuit. Features represented by these filters tile the space of conventional receptive field of retinal ganglion cells. These results suggest that CNN could be used to reveal structure components of neuronal circuits.

stat.ML↗

arXiv2015

Sampling-based Causal Inference in Cue Combination and its Neural Implementation

Causal inference in cue combination is to decide whether the cues have a single cause or multiple causes. Although the Bayesian causal inference model explains the problem of causal inference in cue combination successfully, how causal inference in cue combination could be implemented by neural circuits, is unclear. The existing method based on calculating log posterior ratio with variable elimination has the problem of being unrealistic and task-specific. In this paper, we take advantages of the special structure of the Bayesian causal inference model and propose a hierarchical inference algorithm based on importance sampling. A simple neural circuit is designed to implement the proposed inference algorithm. Theoretical analyses and experimental results demonstrate that our algorithm converges to the accurate value as the sample size goes to infinite. Moreover, the neural circuit we design can be easily generalized to implement inference for other problems, such as the multi-stimuli cause inference and the same-different judgment.

cs.NE↗