1
|
Qin L, Wang Z, Yan R, Tang H. Attention-Based Deep Spiking Neural Networks for Temporal Credit Assignment Problems. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024; 35:10301-10311. [PMID: 37022405 DOI: 10.1109/tnnls.2023.3240176] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/19/2023]
Abstract
The temporal credit assignment (TCA) problem, which aims to detect predictive features hidden in distracting background streams, remains a core challenge in biological and machine learning. Aggregate-label (AL) learning is proposed by researchers to resolve this problem by matching spikes with delayed feedback. However, the existing AL learning algorithms only consider the information of a single timestep, which is inconsistent with the real situation. Meanwhile, there is no quantitative evaluation method for TCA problems. To address these limitations, we propose a novel attention-based TCA (ATCA) algorithm and a minimum editing distance (MED)-based quantitative evaluation method. Specifically, we define a loss function based on the attention mechanism to deal with the information contained within the spike clusters and use MED to evaluate the similarity between the spike train and the target clue flow. Experimental results on musical instrument recognition (MedleyDB), speech recognition (TIDIGITS), and gesture recognition (DVS128-Gesture) show that the ATCA algorithm can reach the state-of-the-art (SOTA) level compared with other AL learning algorithms.
Collapse
|
2
|
Zhang Y, Chen Y, Zhang J, Luo X, Zhang M, Qu H, Yi Z. Minicolumn-Based Episodic Memory Model With Spiking Neurons, Dendrites and Delays. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024; 35:7072-7086. [PMID: 36279337 DOI: 10.1109/tnnls.2022.3213688] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/16/2023]
Abstract
Episodic memory is fundamental to the brain's cognitive function, but how neuronal activity is temporally organized during its encoding and retrieval is still unknown. In this article, combining hippocampus structure with a spiking neural network (SNN), a new bionic spiking temporal memory (BSTM) model is proposed to explore the encoding, formation, and retrieval of episodic memory. For encoding episodic memory, the spike-timing-dependent-plasticity (STDP) learning algorithm and a proposed minicolumn selection algorithm are used to encode each input item into several active minicolumns. For the formation of episodic memory, a sequential memory algorithm is proposed to store the contexts between items. For retrieval of episodic memory, the local retrieval algorithm and the global retrieval algorithm are proposed to retrieve sequence information, achieving multisentence prediction and multitime step prediction. All functions of BSTM are based on bionic spiking neurons, which have biological characteristics including columnar and dendritic structures, firing and receiving spikes, and delaying transmission. To test the performance of the BSTM model, the Children's Book Test (CBT) data set was used to conduct a series of experiments under different settings, including changing the number of minicolumns, neurons and sequences, modifying sequence items, etc. Compared to other sequence memory algorithms, the experimental results show that the proposed BSTM achieves higher accuracy and better robustness.
Collapse
|
3
|
Wang J. Training multi-layer spiking neural networks with plastic synaptic weights and delays. Front Neurosci 2024; 17:1253830. [PMID: 38328553 PMCID: PMC10847234 DOI: 10.3389/fnins.2023.1253830] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/06/2023] [Accepted: 12/04/2023] [Indexed: 02/09/2024] Open
Abstract
Spiking neural networks are usually considered as the third generation of neural networks, which hold the potential of ultra-low power consumption on corresponding hardware platforms and are very suitable for temporal information processing. However, how to efficiently train the spiking neural networks remains an open question, and most existing learning methods only consider the plasticity of synaptic weights. In this paper, we proposed a new supervised learning algorithm for multiple-layer spiking neural networks based on the typical SpikeProp method. In the proposed method, both the synaptic weights and delays are considered as adjustable parameters to improve both the biological plausibility and the learning performance. In addition, the proposed method inherits the advantages of SpikeProp, which can make full use of the temporal information of spikes. Various experiments are conducted to verify the performance of the proposed method, and the results demonstrate that the proposed method achieves a competitive learning performance compared with the existing related works. Finally, the differences between the proposed method and the existing mainstream multi-layer training algorithms are discussed.
Collapse
Affiliation(s)
- Jing Wang
- School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China
| |
Collapse
|
4
|
Guo Y, Huang X, Ma Z. Direct learning-based deep spiking neural networks: a review. Front Neurosci 2023; 17:1209795. [PMID: 37397460 PMCID: PMC10313197 DOI: 10.3389/fnins.2023.1209795] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/21/2023] [Accepted: 06/01/2023] [Indexed: 07/04/2023] Open
Abstract
The spiking neural network (SNN), as a promising brain-inspired computational model with binary spike information transmission mechanism, rich spatially-temporal dynamics, and event-driven characteristics, has received extensive attention. However, its intricately discontinuous spike mechanism brings difficulty to the optimization of the deep SNN. Since the surrogate gradient method can greatly mitigate the optimization difficulty and shows great potential in directly training deep SNNs, a variety of direct learning-based deep SNN works have been proposed and achieved satisfying progress in recent years. In this paper, we present a comprehensive survey of these direct learning-based deep SNN works, mainly categorized into accuracy improvement methods, efficiency improvement methods, and temporal dynamics utilization methods. In addition, we also divide these categorizations into finer granularities further to better organize and introduce them. Finally, the challenges and trends that may be faced in future research are prospected.
Collapse
Affiliation(s)
- Yufei Guo
- Intelligent Science & Technology Academy of CASIC, Beijing, China
- Scientific Research Laboratory of Aerospace Intelligent Systems and Technology, Beijing, China
| | - Xuhui Huang
- Intelligent Science & Technology Academy of CASIC, Beijing, China
- Scientific Research Laboratory of Aerospace Intelligent Systems and Technology, Beijing, China
| | - Zhe Ma
- Intelligent Science & Technology Academy of CASIC, Beijing, China
- Scientific Research Laboratory of Aerospace Intelligent Systems and Technology, Beijing, China
| |
Collapse
|
5
|
Chen Y, Liu H, Shi K, Zhang M, Qu H. Spiking neural network with working memory can integrate and rectify spatiotemporal features. Front Neurosci 2023; 17:1167134. [PMID: 37389360 PMCID: PMC10300445 DOI: 10.3389/fnins.2023.1167134] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/16/2023] [Accepted: 05/26/2023] [Indexed: 07/01/2023] Open
Abstract
In the real world, information is often correlated with each other in the time domain. Whether it can effectively make a decision according to the global information is the key indicator of information processing ability. Due to the discrete characteristics of spike trains and unique temporal dynamics, spiking neural networks (SNNs) show great potential in applications in ultra-low-power platforms and various temporal-related real-life tasks. However, the current SNNs can only focus on the information a short time before the current moment, its sensitivity in the time domain is limited. This problem affects the processing ability of SNN in different kinds of data, including static data and time-variant data, and reduces the application scenarios and scalability of SNN. In this work, we analyze the impact of such information loss and then integrate SNN with working memory inspired by recent neuroscience research. Specifically, we propose Spiking Neural Networks with Working Memory (SNNWM) to handle input spike trains segment by segment. On the one hand, this model can effectively increase SNN's ability to obtain global information. On the other hand, it can effectively reduce the information redundancy between adjacent time steps. Then, we provide simple methods to implement the proposed network architecture from the perspectives of biological plausibility and neuromorphic hardware friendly. Finally, we test the proposed method on static and sequential data sets, and the experimental results show that the proposed model can better process the whole spike train, and achieve state-of-the-art results in short time steps. This work investigates the contribution of introducing biologically inspired mechanisms, e.g., working memory, and multiple delayed synapses to SNNs, and provides a new perspective to design future SNNs.
Collapse
|
6
|
Grimaldi A, Gruel A, Besnainou C, Jérémie JN, Martinet J, Perrinet LU. Precise Spiking Motifs in Neurobiological and Neuromorphic Data. Brain Sci 2022; 13:68. [PMID: 36672049 PMCID: PMC9856822 DOI: 10.3390/brainsci13010068] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/16/2022] [Revised: 12/20/2022] [Accepted: 12/23/2022] [Indexed: 12/31/2022] Open
Abstract
Why do neurons communicate through spikes? By definition, spikes are all-or-none neural events which occur at continuous times. In other words, spikes are on one side binary, existing or not without further details, and on the other, can occur at any asynchronous time, without the need for a centralized clock. This stands in stark contrast to the analog representation of values and the discretized timing classically used in digital processing and at the base of modern-day neural networks. As neural systems almost systematically use this so-called event-based representation in the living world, a better understanding of this phenomenon remains a fundamental challenge in neurobiology in order to better interpret the profusion of recorded data. With the growing need for intelligent embedded systems, it also emerges as a new computing paradigm to enable the efficient operation of a new class of sensors and event-based computers, called neuromorphic, which could enable significant gains in computation time and energy consumption-a major societal issue in the era of the digital economy and global warming. In this review paper, we provide evidence from biology, theory and engineering that the precise timing of spikes plays a crucial role in our understanding of the efficiency of neural networks.
Collapse
Affiliation(s)
- Antoine Grimaldi
- INT UMR 7289, Aix Marseille Univ, CNRS, 27 Bd Jean Moulin, 13005 Marseille, France
| | - Amélie Gruel
- SPARKS, Côte d’Azur, CNRS, I3S, 2000 Rte des Lucioles, 06900 Sophia-Antipolis, France
| | - Camille Besnainou
- INT UMR 7289, Aix Marseille Univ, CNRS, 27 Bd Jean Moulin, 13005 Marseille, France
| | - Jean-Nicolas Jérémie
- INT UMR 7289, Aix Marseille Univ, CNRS, 27 Bd Jean Moulin, 13005 Marseille, France
| | - Jean Martinet
- SPARKS, Côte d’Azur, CNRS, I3S, 2000 Rte des Lucioles, 06900 Sophia-Antipolis, France
| | - Laurent U. Perrinet
- INT UMR 7289, Aix Marseille Univ, CNRS, 27 Bd Jean Moulin, 13005 Marseille, France
| |
Collapse
|
7
|
Hu L, Liao X. Voltage slope guided learning in spiking neural networks. Front Neurosci 2022; 16:1012964. [PMID: 36440266 PMCID: PMC9685168 DOI: 10.3389/fnins.2022.1012964] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/06/2022] [Accepted: 10/25/2022] [Indexed: 04/19/2024] Open
Abstract
A thorny problem in machine learning is how to extract useful clues related to delayed feedback signals from the clutter of input activity, known as the temporal credit-assignment problem. The aggregate-label learning algorithms make an explicit representation of this problem by training spiking neurons to assign the aggregate feedback signal to potentially effective clues. However, earlier aggregate-label learning algorithms suffered from inefficiencies due to the large amount of computation, while recent algorithms that have solved this problem may fail to learn due to the inability to find adjustment points. Therefore, we propose a membrane voltage slope guided algorithm (VSG) to further cope with this limitation. Direct dependence on the membrane voltage when finding the key point of weight adjustment makes VSG avoid intensive calculation, but more importantly, the membrane voltage that always exists makes it impossible to lose the adjustment point. Experimental results show that the proposed algorithm can correlate delayed feedback signals with the effective clues embedded in background spiking activity, and also achieves excellent performance on real medical classification datasets and speech classification datasets. The superior performance makes it a meaningful reference for aggregate-label learning on spiking neural networks.
Collapse
Affiliation(s)
- Lvhui Hu
- School of Intelligent Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China
| | - Xin Liao
- Information Center, Hospital of Chengdu University of Traditional Chinese Medicine, Chengdu, China
| |
Collapse
|
8
|
Yang Z, Yang Y, Li C, Zhou Y, Zhang X, Yu Y, Liu D. Tasseled Crop Rows Detection Based on Micro-Region of Interest and Logarithmic Transformation. FRONTIERS IN PLANT SCIENCE 2022; 13:916474. [PMID: 35832229 PMCID: PMC9272774 DOI: 10.3389/fpls.2022.916474] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 04/09/2022] [Accepted: 05/06/2022] [Indexed: 06/15/2023]
Abstract
Machine vision-based navigation in the maize field is significant for intelligent agriculture. Therefore, precision detection of the tasseled crop rows for navigation of agricultural machinery with an accurate and fast method remains an open question. In this article, we propose a new crop rows detection method at the tasseling stage of maize fields for agrarian machinery navigation. The whole work is achieved mainly through image augment and feature point extraction by micro-region of interest (micro-ROI). In the proposed method, we first augment the distinction between the tassels and background by the logarithmic transformation in RGB color space, and then the image is transformed to hue-saturation-value (HSV) space to extract the tassels. Second, the ROI is approximately selected and updated using the bounding box until the multiple-region of interest (multi-ROI) is determined. We further propose a feature points extraction method based on micro-ROI and the feature points are used to calculate the crop rows detection lines. Finally, the bisector of the acute angle formed by the two detection lines is used as the field navigation line. The experimental results show that the algorithm proposed has good robustness and can accurately detect crop rows. Compared with other existing methods, our method's accuracy and real-time performance have improved by about 5 and 62.3%, respectively, which can meet the accuracy and real-time requirements of agricultural vehicles' navigation in maize fields.
Collapse
Affiliation(s)
- Zhenling Yang
- School of Engineering, Anhui Agricultural University, Hefei, China
| | - Yang Yang
- School of Engineering, Anhui Agricultural University, Hefei, China
- Institute of Artificial Intelligence, Hefei Comprehensive Nation Science Center, Hefei, China
| | - Chaorong Li
- Faculty of Artificial Intelligence and Big Data, Yibin University, Yibin, China
| | - Yang Zhou
- School of Engineering, Anhui Agricultural University, Hefei, China
| | - Xiaoshuang Zhang
- School of Engineering, Anhui Agricultural University, Hefei, China
| | - Yang Yu
- School of Engineering, Anhui Agricultural University, Hefei, China
| | - Dan Liu
- JD AI Research, Beijing, China
| |
Collapse
|