Constraints on Hebbian and STDP learned weights of a spiking neuron

Chu, Dominique; Nguyêݱn, Huy L.

doi:10.1016/j.neunet.2020.12.012

Cited by 10 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two learning processes also occur when acquiring new skills: a fast learning in a cortical structure is simultaneously slowly learned in a subcortical structure (habit learning for instance) [72][73][74][75] . Thus, according to previous experiences and the nature of inputs, the rate of learning may need to evolve so that neurons adapt more or less quickly to sensory inputs 69,72,76,77 . More precisely, adaptation in the brain is the result of multiple learning processes acting at distinct time scales allowing more or less rapid learning or relearning of new information without forgetting older memories 39,72 .…”

Section: /20mentioning

confidence: 99%

Inhibitory neurons control the consolidation of neural assemblies via adaptation to selective stimuli

Bergoin

Torcini

Deco

et al. 2023

Preprint

View full text Add to dashboard Cite

Brain circuits display modular architecture at different scales of organization. Such neural assemblies are typically associated to functional specialization but the mechanisms leading to their emergence and consolidation still remain elusive. In this paper we investigate the role of inhibition in structuring new neural assemblies driven by the entrainment to various inputs. In particular, we focus on the role of partially synchronized dynamics for the creation and maintenance of structural modules in neural circuits by considering a network of excitatory and inhibitory theta-neurons with plastic Hebbian synapses. The learning process consists of an entrainment to temporally alternating stimuli that are applied to separate regions of the network. This entrainment leads to the emergence of modular structures. Contrary to common practice in artificial neural networks - where the acquired weights are typically frozen after the learning session - we allow for synaptic adaptation even after the learning phase. We find that the presence of inhibitory neurons in the network is crucial for the emergence and the post-learning consolidation of the modular structures. Indeed networks made of purely excitatory neurons or of neurons not respecting Dale's principle are unable to form or maintain the modular architecture induced by the entrained stimuli. We also demonstrate that the number of inhibitory neurons in the network is directly related to the maximal number of neural assemblies that can be consolidated, supporting the idea that inhibition has a direct impact on the memory capacity of the neural network.

show abstract

Section: /20mentioning

confidence: 99%

Inhibitory neurons control the consolidation of neural assemblies via adaptation to selective stimuli

Bergoin

Torcini

Deco

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Metastable states are points in parameter space where the effective learning rate vanishes on average (Chu and Nguyen 2021). This means that the expected increase and decrease of weights at this point is zero for all weights.…”

Section: Chumentioning

confidence: 99%

“…It is possible to determine conditions that the weights have to fulfil when the neuron is in a metastable state. Following (Chu and Nguyen 2021), for the Hebbian neuron this condition is simply…”

Section: Chumentioning

confidence: 99%

“…Due to random fluctuations that occurred as a result of learning, the weights of a neuron will normally not be exactly in a metastable state. In order to know how far away the neuron is from a metastable state, we measure the distance from a metastable state or simply the ''distance,'' following (Chu and Nguyen 2021). In order to determine the distance in the case of the Hebbian neuron, we first estimated in simulation the probability PðijoÞ that input channel i triggered an output spike.…”

Section: Determining the Distance From A Metastable Statementioning

confidence: 99%

See 1 more Smart Citation

Information theoretical properties of a spiking neuron trained with Hebbian and STDP learning rules

Chu

2023

Nat Comput

Self Cite

View full text Add to dashboard Cite

Using formal methods complemented by large-scale simulations we investigate information theoretical properties of spiking neurons trained using Hebbian and STDP learning rules. It is shown that weight space contains meta-stable states, which are points where the average weight change under the learning rule vanishes. These points may capture the random walker transiently. The dwell time in the vicinity of the meta-stable state is either quasi-infinite or very short and depends on the level of noise in the system. Moreover, important information theoretic quantities, such as the amount of information the neuron transmits are determined by the meta-stable state. While the Hebbian learning rule reliably leads to meta-stable states, the STDP rule tends to be unstable in the sense that for most choices of hyper-parameters the weights are not captured by meta-stable states, except for a restricted set of choices. It emerges that stochastic fluctuations play an important role in determining which meta-stable state the neuron takes. To understand this, we model the trajectory of the neuron through weight space as an inhomogeneous Markovian random walk, where the transition probabilities between states are determined by the statistics of the input signal.

show abstract

“…At present, the research findings on memory generation generally agree that the connection between two neurons is usually built according to synaptic plasticity rules, such as the Hebbian rule and spike-timing-dependent plasticity (STDP). In NMIS, when a PN that an instance corresponds to, and an SN that stores this instance’s category pattern, are activated by the external input stimulation simultaneously, they tend to establish an inter-field connection, which represents the formation of memory [ 35 , 36 , 37 ]. We consider that the inter-field connections between PNs and SNs also follow the Hebbian rule.…”

Section: Introductionmentioning

confidence: 99%

Applying the Properties of Neurons in Machine Learning: A Brain-like Neural Model with Interactive Stimulation for Data Classification

Li²,

Huang³

2022

Brain Sciences

View full text Add to dashboard Cite

Some neural models achieve outstanding results in image recognition, semantic segmentation and natural language processing. However, their classification performance on structured and small-scale datasets that do not involve feature extraction is worse than that of traditional algorithms, although they require more time to train. In this paper, we propose a brain-like neural model with interactive stimulation (NMIS) that focuses on data classification. It consists of a primary neural field and a senior neural field that play different cognitive roles. The former is used to correspond to real instances in the feature space, and the latter stores the category pattern. Neurons in the primary field exchange information through interactive stimulation and their activation is transmitted to the senior field via inter-field interaction, simulating the mechanisms of neuronal interaction and synaptic plasticity, respectively. The proposed NMIS is biologically plausible and does not involve complex optimization processes. Therefore, it exhibits better learning ability on small-scale and structured datasets than traditional BP neural networks. For large-scale data classification, a nearest neighbor NMIS (NN_NMIS), an optimized version of NMIS, is proposed to improve computational efficiency. Numerical experiments performed on some UCI datasets show that the proposed NMIS and NN_NMIS are significantly superior to some classification algorithms that are widely used in machine learning.

show abstract

Constraints on Hebbian and STDP learned weights of a spiking neuron

Abstract: The version in the Kent Academic Repository may differ from the final published version. Users are advised to check http://kar.kent.ac.uk for the status of the paper. Users should always cite the published version of record.

Cited by 10 publications

References 28 publications

Inhibitory neurons control the consolidation of neural assemblies via adaptation to selective stimuli

Inhibitory neurons control the consolidation of neural assemblies via adaptation to selective stimuli

Information theoretical properties of a spiking neuron trained with Hebbian and STDP learning rules

Applying the Properties of Neurons in Machine Learning: A Brain-like Neural Model with Interactive Stimulation for Data Classification

Contact Info

Product

Resources

About