A Quaternionic Rate-Based Synaptic Learning Rule Derived from Spike-Timing Dependent Plasticity

2017 ◽  
pp. 457-465
Author(s):  
Guang Qiao ◽  
Hongyue Du ◽  
Yi Zeng
Keyword(s):  
Learning Rule ◽  
Spike Timing ◽  
Spike Timing Dependent Plasticity ◽  
Dependent Plasticity ◽  
Synaptic Learning

Supervised Spike-Timing-Dependent Plasticity: A Spatiotemporal Neuronal Learning Rule for Function Approximation and Decisions

2013 ◽  
Vol 25 (12) ◽  
pp. 3113-3130 ◽  
Author(s):  
Jan-Moritz P. Franosch ◽  
Sebastian Urban ◽  
J. Leo van Hemmen
Keyword(s):  
Function Approximation ◽  
Sensory Input ◽  
Learning Rule ◽  
Animal Learn ◽  
Spike Timing ◽  
Time Dependent ◽  
Stable Configuration ◽  
Spike Timing Dependent Plasticity ◽  
Dependent Plasticity ◽  
Tactile System

How can an animal learn from experience? How can it train sensors, such as the auditory or tactile system, based on other sensory input such as the visual system? Supervised spike-timing-dependent plasticity (supervised STDP) is a possible answer. Supervised STDP trains one modality using input from another one as “supervisor.” Quite complex time-dependent relationships between the senses can be learned. Here we prove that under very general conditions, supervised STDP converges to a stable configuration of synaptic weights leading to a reconstruction of primary sensory input.

scholarly journals Beyond spike-timing-dependent plasticity: a computational study of plasticity gradients across basal dendrites

2016 ◽  
Author(s):  
Jacopo Bono ◽  
Claudia Clopath
Keyword(s):  
Synaptic Plasticity ◽  
Computational Study ◽  
Learning Rule ◽  
Spike Timing ◽  
Memory Retention ◽  
Neuron Models ◽  
Spike Timing Dependent Plasticity ◽  
Principal Mechanism ◽  
Dependent Plasticity ◽  
Dendritic Spikes

AbstractSynaptic plasticity is thought to be the principal mechanism underlying learning in the brain. Models of plastic networks typically combine point neurons with spike-timing-dependent plasticity (STDP) as the learning rule. However, a point neuron does not capture the complexity of dendrites, which allow non-linear local processing of the synaptic inputs. Furthermore, experimental evidence suggests that STDP is not the only learning rule available to neurons. Implementing biophysically realistic neuron models, we studied how dendrites allow for multiple synaptic plasticity mechanisms to coexist in a single cell. In these models, we compared the conditions for STDP and for the synaptic strengthening by local dendritic spikes. We further explored how the connectivity between two cells is affected by these plasticity rules and the synaptic distributions. Finally, we show how memory retention in associative learning can be prolonged in networks of neurons with dendrites.

scholarly journals Environmental enrichment shapes striatal spike-timing-dependent plasticity in vivo

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Teresa Morera-Herreras ◽  
Yves Gioanni ◽  
Sylvie Perez ◽  
Gaetan Vignoud ◽  
Laurent Venance
Keyword(s):  
Failure Rate ◽  
Environmental Enrichment ◽  
Hebbian Learning ◽  
Learning Rule ◽  
Spike Timing ◽  
Long Term Effects ◽  
Spike Timing Dependent Plasticity ◽  
Temporal Window ◽  
Dependent Plasticity

AbstractBehavioural experience, such as environmental enrichment (EE), induces long-term effects on learning and memory. Learning can be assessed with the Hebbian paradigm, such as spike-timing-dependent plasticity (STDP), which relies on the timing of neuronal activity on either side of the synapse. Although EE is known to control neuronal excitability and consequently spike timing, whether EE shapes STDP remains unknown. Here, using in vivo long-duration intracellular recordings at the corticostriatal synapses we show that EE promotes asymmetric anti-Hebbian STDP, i.e. spike-timing-dependent-potentiation (tLTP) for post-pre pairings and spike-timing-dependent-depression (tLTD) for pre-post pairings, whereas animals grown in standard housing show mainly tLTD and a high failure rate of plasticity. Indeed, in adult rats grown in standard conditions, we observed unidirectional plasticity (mainly symmetric anti-Hebbian tLTD) within a large temporal window (~200 ms). However, rats grown for two months in EE displayed a bidirectional STDP (tLTP and tLTD depending on spike timing) in a more restricted temporal window (~100 ms) with low failure rate of plasticity. We also found that the effects of EE on STDP characteristics are influenced by the anaesthesia status: the deeper the anaesthesia, the higher the absence of plasticity. These findings establish a central role for EE and the anaesthetic regime in shaping in vivo, a synaptic Hebbian learning rule such as STDP.

scholarly journals Optimal Spike-Timing-Dependent Plasticity for Precise Action Potential Firing in Supervised Learning

2006 ◽  
Vol 18 (6) ◽  
pp. 1318-1348 ◽  
Author(s):  
Jean-Pascal Pfister ◽  
Taro Toyoizumi ◽  
David Barber ◽  
Wulfram Gerstner
Keyword(s):  
Supervised Learning ◽  
Time Course ◽  
Learning Rule ◽  
Spike Timing ◽  
Excitatory Postsynaptic Potential ◽  
Spike Timing Dependent Plasticity ◽  
Action Potential Firing ◽  
Dependent Plasticity ◽  
Potential Firing ◽  
Presynaptic Spike

In timing-based neural codes, neurons have to emit action potentials at precise moments in time. We use a supervised learning paradigm to derive a synaptic update rule that optimizes by gradient ascent the likelihood of postsynaptic firing at one or several desired firing times. We find that the optimal strategy of up- and downregulating synaptic efficacies depends on the relative timing between presynaptic spike arrival and desired postsynaptic firing. If the presynaptic spike arrives before the desired postsynaptic spike timing, our optimal learning rule predicts that the synapse should become potentiated. The dependence of the potentiation on spike timing directly reflects the time course of an excitatory postsynaptic potential. However, our approach gives no unique reason for synaptic depression under reversed spike timing. In fact, the presence and amplitude of depression of synaptic efficacies for reversed spike timing depend on how constraints are implemented in the optimization problem. Two different constraints, control of postsynaptic rates and control of temporal locality, are studied. The relation of our results to spike-timing-dependent plasticity and reinforcement learning is discussed.

scholarly journals Memory Retention and Spike-Timing-Dependent Plasticity

2009 ◽  
Vol 101 (6) ◽  
pp. 2775-2788 ◽  
Author(s):  
Guy Billings ◽  
Mark C. W. van Rossum
Keyword(s):  
Retention Time ◽  
Single Neuron ◽  
Receptive Fields ◽  
Learning Rule ◽  
Spike Timing ◽  
Memory Retention ◽  
Neuron Models ◽  
Spike Timing Dependent Plasticity ◽  
Dependent Plasticity ◽  
Dependent Learning

Memory systems should be plastic to allow for learning; however, they should also retain earlier memories. Here we explore how synaptic weights and memories are retained in models of single neurons and networks equipped with spike-timing-dependent plasticity. We show that for single neuron models, the precise learning rule has a strong effect on the memory retention time. In particular, a soft-bound, weight-dependent learning rule has a very short retention time as compared with a learning rule that is independent of the synaptic weights. Next, we explore how the retention time is reflected in receptive field stability in networks. As in the single neuron case, the weight-dependent learning rule yields less stable receptive fields than a weight-independent rule. However, receptive fields stabilize in the presence of sufficient lateral inhibition, demonstrating that plasticity in networks can be regulated by inhibition and suggesting a novel role for inhibition in neural circuits.

scholarly journals Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Yoshifumi Nishi ◽  
Kumiko Nomura ◽  
Takao Marukame ◽  
Koichi Mizushima
Keyword(s):  
Unsupervised Learning ◽  
Recognition Accuracy ◽  
Learning Rule ◽  
Spike Timing ◽  
Distribution Functions ◽  
Cumulative Distribution ◽  
Cumulative Probability ◽  
Spike Timing Dependent Plasticity ◽  
Dependent Plasticity ◽  
Neuromorphic Hardware

AbstractSpike timing-dependent plasticity (STDP), which is widely studied as a fundamental synaptic update rule for neuromorphic hardware, requires precise control of continuous weights. From the viewpoint of hardware implementation, a simplified update rule is desirable. Although simplified STDP with stochastic binary synapses was proposed previously, we find that it leads to degradation of memory maintenance during learning, which is unfavourable for unsupervised online learning. In this work, we propose a stochastic binary synaptic model where the cumulative probability of the weight change evolves in a sigmoidal fashion with potentiation or depression trials, which can be implemented using a pair of switching devices consisting of serially connected multiple binary memristors. As a benchmark test we perform simulations of unsupervised learning of MNIST images with a two-layer network and show that simplified STDP in combination with this model can outperform conventional rules with continuous weights not only in memory maintenance but also in recognition accuracy. Our method achieves 97.3% in recognition accuracy, which is higher than that reported with standard STDP in the same framework. We also show that the high performance of our learning rule is robust against device-to-device variability of the memristor's probabilistic behaviour.

scholarly journals Emergence of oscillations via spike timing dependent plasticity

2018 ◽  
Author(s):  
Sarit Soloduchin ◽  
Maoz Shamir
Keyword(s):  
Phase Diagram ◽  
Neuronal Network ◽  
Learning Process ◽  
Learning Rule ◽  
Spike Timing ◽  
Oscillatory Activity ◽  
Spike Timing Dependent Plasticity ◽  
Dependent Plasticity ◽  
Wide Range ◽  
Stdp Rule

AbstractNeuronal oscillatory activity has been reported in relation to a wide range of cognitive processes. In certain cases changes in oscillatory activity has been associated with pathological states. Although the specific role of these oscillations has yet to be determined, it is clear that neuronal oscillations are abundant in the central nervous system. These observations raise the question of the origin of these oscillations; and specifically whether the mechanisms responsible for the generation and stabilization of these oscillations are genetically hard-wired or whether they can be acquired via a learning process.Here we focus on spike timing dependent plasticity (STDP) to investigate whether oscillatory activity can emerge in a neuronal network via an unsupervised learning process of STDP dynamics, and if so, what features of the STDP learning rule govern and stabilize the resultant oscillatory activity?Here, the STDP dynamics of the effective coupling between two competing neuronal populations with reciprocal inhibitory connections was analyzed using the phase-diagram of the system that depicts the possible dynamical states of the network as a function of the effective inhibitory couplings. This phase diagram yields a rich repertoire of possible dynamical behaviors including regions of different fixed point solutions, bi-stability and a region in which the system exhibits oscillatory activity. STDP introduces dynamics for the inhibitory couplings themselves and hence induces a flow in the phase diagram. We investigate the conditions for the flow to converge to an oscillatory state of the neuronal network and then characterize how the features of the STDP rule govern and stabilize these oscillations.

Sign in / Sign up

Export Citation Format

Share Document