Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Neural Circuits01:25

Neural Circuits

Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neural Regulation01:37

Neural Regulation

Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
Overview of Synapses01:25

Overview of Synapses

A synapse is a specialized structure where two neurons connect, allowing them to pass an electrical or chemical signal to another neuron. It is the point of communication between neurons. The term "synapse" is derived from the Greek word "synapsis," which means "conjunction." The entire process of neural communication revolves around the synapse. When activated, a neuron releases chemicals known as neurotransmitters into the synapse. These neurotransmitters cross the synapse and bind to...
Neurons: The Axon01:21

Neurons: The Axon

Axons are long, cytoplasmic processes of nerve cells capable of propagating electrical impulses known as action potentials. The cytoplasm or axoplasm of an axon contains neurofibrils, neurotubules, small vesicles, lysosomes, mitochondria, and various enzymes, all encased within the axolemma, the plasma membrane of the axon.
The axon attaches to the cell body at a cone-shaped elevation called the axon hillock. The initial part of the axon, closest to the hillock, is known as the initial segment.
Electrical Synapses01:28

Electrical Synapses

Electrical synapses found in all nervous systems play important and unique roles. In these synapses, the presynaptic and postsynaptic membranes are very close together (3.5 nm) and are actually physically connected by channel proteins forming gap junctions.
Gap junctions allow the current to pass directly from one cell to the next. In contrast, in the chemical synapse, the neurotransmitters carry the information through the synaptic cleft from one neuron to the next. They consist of two...
Neuronal Communication01:28

Neuronal Communication

Neurons, the fundamental units of the brain and nervous system, communicate through complex electrochemical signals that underpin all cognitive and bodily functions. This communication is primarily facilitated by a process involving the generation and propagation of an action potential along the axon of the neuron. When the internal electrical charge of a neuron surpasses a certain threshold, an action potential is triggered. This rapid change in voltage travels swiftly along the axon to the...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Characterizing open-ended evolution through undecidability mechanisms in random Boolean networks.

NPJ systems biology and applications·2026
Same author

Closing the loop: how semantic closure enables open-ended evolution?

Journal of the Royal Society, Interface·2026
Same author

Correction: Sustainable visions: Unsupervised machine learning insights on global development goals.

PloS one·2025
Same author

Sustainable visions: unsupervised machine learning insights on global development goals.

PloS one·2025
Same author

Complexity, Artificial Life, and Artificial Intelligence.

Artificial life·2024
Same author

Language Statistics at Different Spatial, Temporal, and Grammatical Scales.

Entropy (Basel, Switzerland)·2024
Same journal

An efficient hybrid CNN-transformer framework for real-time weapon detection and face recognition.

Frontiers in artificial intelligence·2026
Same journal

Ontology-based annotation and fuzzy recommendation for community formation in smart city knowledge platforms.

Frontiers in artificial intelligence·2026
Same journal

A generalized logistic-logit function and its application to multi-layer perceptron and neuron segmentation.

Frontiers in artificial intelligence·2026
Same journal

A multimodal, risk-stratified framework for AI-driven early risk prediction and personalised prevention in obesity.

Frontiers in artificial intelligence·2026
Same journal

The quantified immune-aging dysregulation index: a large-language model-powered method for annotating and quantifying systems-level dysregulation.

Frontiers in artificial intelligence·2026
Same journal

CA<sup>2</sup>PNet: a context-aware multi-scale architecture with adaptive attention and progressive dilated convolutions for biomedical image segmentation.

Frontiers in artificial intelligence·2026
See all related articles
  1. Home
  3. Spark: Modular Spiking Neural Networks.
  1. Home
  3. Spark: Modular Spiking Neural Networks.

Related Experiment Video

Two-Photon Polymerization 3D-Printing of Micro-scale Neuronal Cell Culture Devices
07:38

Two-Photon Polymerization 3D-Printing of Micro-scale Neuronal Cell Culture Devices

Published on: June 7, 2024

Spark: modular spiking neural networks.

Mario Franco1, Carlos Gershenson1

  • 1School of Systems Science and Industrial Enginnering, Binghamton University, Binghamton, NY, United States.

Frontiers in Artificial Intelligence
|May 25, 2026

View abstract on PubMed

Summary
This summary is machine-generated.

This article introduces Spark, a modular software framework designed to simplify the creation and training of spiking neural networks. By focusing on energy efficiency and biological-like learning, the authors provide tools to help researchers build models that learn continuously, similar to how animals process information.

Keywords:
bio-inspired AIcartpolepython packagereinforcement learningspike-timing dependent plasticityspiking neural networksartificial intelligencesynaptic plasticitymachine learningevent-based computing

Frequently Asked Questions

More Related Videos

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
11:18

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments
05:19

Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments

Published on: November 12, 2019

Related Experiment Videos

Two-Photon Polymerization 3D-Printing of Micro-scale Neuronal Cell Culture Devices
07:38

Two-Photon Polymerization 3D-Printing of Micro-scale Neuronal Cell Culture Devices

Published on: June 7, 2024

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
11:18

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments
05:19

Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments

Published on: November 12, 2019

Area of Science:

  • Computational neuroscience and spiking neural networks research
  • Artificial intelligence systems within machine learning

Background:

Current artificial intelligence models often suffer from high energy consumption and excessive data requirements during training phases. This limitation hinders the development of sustainable computing architectures for complex tasks. Spiking neural networks offer a promising alternative due to their potential for hardware-level efficiency. Yet, researchers struggle to define robust learning rules that function effectively within these temporal architectures. Prior work suggests that synaptic plasticity might resolve these efficiency bottlenecks. That uncertainty drove the development of new computational paradigms for event-based processing. No prior work had resolved the integration of modularity with these specific biological learning mechanisms. This gap motivated the creation of a flexible software environment for testing such theories.

Purpose Of The Study:

The aim of this framework is to provide an efficient and streamlined pipeline for spiking neural networks. Researchers sought to address the inherent inefficiencies found in current artificial intelligence models regarding energy and data usage. This project specifically targets the difficulty of defining effective learning algorithms for temporal spiking architectures. The team intended to prove that modularity could simplify the construction of complex models from basic components. They also wanted to demonstrate that synaptic plasticity could improve learning performance in sparse-reward environments. By creating an open-source tool, the authors hoped to foster wider experimentation within the scientific community. This effort addresses the need for architectures that support continuous, unbatched learning processes. The study ultimately seeks to bridge the gap between artificial systems and the adaptive capabilities observed in biological organisms.

The researchers propose that synaptic plasticity mechanisms enable the network to solve the sparse-reward cartpole problem. This approach allows the system to adjust connections based on temporal spikes rather than traditional backpropagation, improving data efficiency compared to standard deep learning models.

Spark serves as the modular software framework. It allows users to assemble spiking neural networks from simple, interchangeable components, contrasting with monolithic architectures that are difficult to modify or scale for specific hardware implementations.

The authors emphasize that modularity is necessary to bridge the gap between simple spiking components and entire functional models. This structural organization allows for testing diverse plasticity rules without rebuilding the underlying computational pipeline from scratch.

The framework uses unbatched, continuous data streams to mimic biological learning. This contrasts with traditional machine learning pipelines that rely on large, static datasets processed in fixed batches to update network weights.

The researchers measure performance by solving the sparse-reward cartpole problem. They observe that the network successfully learns to balance the pole using simple plasticity, demonstrating efficiency that exceeds non-spiking models in energy-constrained environments.

The authors propose that their framework will accelerate research into continuous, unbatched learning. They suggest this will help bridge the divide between artificial intelligence and the adaptive learning capabilities exhibited by animals.

Main Methods:

The review approach involves evaluating a novel software architecture designed for event-based neural computation. Researchers constructed a library that allows for the hierarchical assembly of network components. This design strategy prioritizes ease of integration with existing machine learning workflows. The team implemented specific plasticity rules to test the adaptability of the spiking nodes. They utilized the cartpole environment as a benchmark for assessing task performance. The evaluation focused on the ability of the system to handle sparse rewards effectively. Developers ensured the code remains accessible via an open-source repository for community verification. This systematic approach highlights the utility of modularity in building complex, energy-efficient models.

Main Results:

Key findings from the literature indicate that the framework successfully solves the sparse-reward cartpole challenge. The implementation of simple plasticity mechanisms allows for effective weight updates without traditional gradient-based methods. This result confirms that spiking architectures can achieve functional goals using biologically inspired learning rules. The software demonstrates high compatibility with standard machine learning pipelines, facilitating broader adoption. Observations show that the modular design significantly reduces the complexity of building large-scale spiking models. The system maintains efficiency during continuous, unbatched learning scenarios. These outcomes suggest that the proposed tools effectively address the data efficiency problems identified in earlier models. The data confirms that modular spiking networks represent a viable path toward sustainable artificial intelligence.

Conclusions:

The authors demonstrate that their modular framework successfully addresses current challenges in spiking network design. This software enables researchers to construct complex models from basic, reusable building blocks. Synaptic plasticity mechanisms integrated into the system improve performance on sparse-reward tasks. The results suggest that such architectures facilitate continuous learning patterns observed in biological organisms. Compatibility with standard machine learning pipelines remains a priority for future adoption. The researchers propose that this tool will accelerate progress in energy-efficient artificial intelligence. Synthesis of these findings indicates that modularity is a viable strategy for advancing event-based computing. The study provides a foundation for further exploration into unbatched, real-time learning systems.