Dendritic Computing: Branching Deeper into Machine Learning

Acharya, Jyotibdha; Basu, Arindam; Legenstein, Robert; Limbacher, Thomas; Poirazi, Panayiota; Wu, Xundong

doi:10.1016/j.neuroscience.2021.10.001

Cited by 23 publications

(16 citation statements)

References 101 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These include improved associative learning 12 , 27 , better input discrimination (pattern separation 34 ), efficient binding/linking of information 12 , 35 , and increased memory storage and recall capacity 14 , 36 . Similar advantages were recently seen in the machine learning field: adding dendritic nodes in artificial neural networks (ANNs) reduced the number of trainable parameters required to achieve high-performance accuracy 37 (also see 38 , 39 ). Moreover, incorporating dendritic nodes in Self Organizing Map classifiers 40 and other neuro-inspired networks 41 improved their continuous learning ability.…”

Section: Introductionmentioning

confidence: 68%

Introducing the Dendrify framework for incorporating dendrites to spiking neural networks

2023

Self Cite

View full text Add to dashboard Cite

Computational modeling has been indispensable for understanding how subcellular neuronal features influence circuit processing. However, the role of dendritic computations in network-level operations remains largely unexplored. This is partly because existing tools do not allow the development of realistic and efficient network models that account for dendrites. Current spiking neural networks, although efficient, are usually quite simplistic, overlooking essential dendritic properties. Conversely, circuit models with morphologically detailed neuron models are computationally costly, thus impractical for large-network simulations. To bridge the gap between these two extremes and facilitate the adoption of dendritic features in spiking neural networks, we introduce Dendrify, an open-source Python package based on Brian 2. Dendrify, through simple commands, automatically generates reduced compartmental neuron models with simplified yet biologically relevant dendritic and synaptic integrative properties. Such models strike a good balance between flexibility, performance, and biological accuracy, allowing us to explore dendritic contributions to network-level functions while paving the way for developing more powerful neuromorphic systems.

show abstract

Section: Introductionmentioning

confidence: 68%

Introducing the Dendrify framework for incorporating dendrites to spiking neural networks

2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…These include improved associative learning 12,27 , better input discrimination (pattern separation 34 ), efficient binding/linking of information 12,35 , and increased memory storage and recall capacity 14,36 . Similar advantages were recently seen in the machine learning field: adding dendritic nodes in artificial neural networks (ANNs) reduced the number of trainable parameters required to achieve high-performance accuracy 37 (also see 38,39 ). Moreover, incorporating dendritic nodes in Self Organizing Map classifiers 40 and other neuro-inspired networks 41 improved their continuous learning ability.…”

Section: Introductionmentioning

confidence: 68%

Introducing the Dendrify framework for incorporating dendrites to spiking neural networks

Pagkalos

Chavlis

Poirazi

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Computational modeling has been indispensable for understanding how subcellular neuronal features influence circuit processing. However, the role of dendritic computations in network-level operations remains largely unexplored. This is partly because existing tools do not allow the development of realistic and efficient network models that account for dendrites. Current spiking neural networks, although efficient, are usually quite simplistic, overlooking essential dendritic properties. Conversely, circuit models with morphologically detailed neuron models are computationally costly, thus impractical for large-network simulations. To bridge the gap between these two extremes, we introduce Dendrify, an open-source Python package compatible with Brian2, designed to facilitate the development of bioinspired spiking neural networks. Dendrify, through simple commands, automatically generates reduced compartmental neuron models with simplified yet biologically relevant dendritic and synaptic integrative properties. Such models strike a good balance between flexibility, performance, and biological accuracy, allowing us to explore dendritic contributions to network-level functions while paving the way for developing more powerful neuromorphic systems.

show abstract

“…The majority of synapses are found on dendrites, branch-like extensions of a neuron that receive electrical stimulation from other neurons. Plasticity mechanisms, together with the processing capabilities of dendrites, are considered to play a crucial role in the emergence of human intelligence [3]. New insights into the functionalities of dendritic processing are driving the search for alternative neural network architectures in machine learning [4].…”

Section: Introductionmentioning

confidence: 99%

Implementation of input correlation learning with an optoelectronic dendritic unit

Ortín

Soriano²,

Tetzlaff³

et al. 2023

Front. Phys.

View full text Add to dashboard Cite

The implementation of machine learning concepts using optoelectronic and photonic components is rapidly advancing. Here, we use the recently introduced notion of optical dendritic structures, which aspires to transfer neurobiological principles to photonics computation. In real neurons, plasticity—the modification of the connectivity between neurons due to their activity—plays a fundamental role in learning. In the current work, we investigate theoretically and experimentally an artificial dendritic structure that implements a modified Hebbian learning model, called input correlation (ICO) learning. The presented optical fiber-based dendritic structure employs the summation of the different optical intensities propagating along the optical dendritic branches and uses Gigahertz-bandwidth modulation via semiconductor optical amplifiers to apply the necessary plasticity rules. In its full deployment, this optoelectronic ICO learning analog can be an efficient hardware platform for ultra-fast control.

show abstract

Dendritic Computing: Branching Deeper into Machine Learning

Cited by 23 publications

References 101 publications

Introducing the Dendrify framework for incorporating dendrites to spiking neural networks

Introducing the Dendrify framework for incorporating dendrites to spiking neural networks

Introducing the Dendrify framework for incorporating dendrites to spiking neural networks

Implementation of input correlation learning with an optoelectronic dendritic unit

Contact Info

Product

Resources

About