一个单一的快速的Hebbian-like过程,使得在深度神经网络中能够一次性添加类,而无需进行骨干修改
Kazufumi Hosoda1,2, Keigo Nishida3, Shigeto Seno4
1Center for Information and Neural Networks, Advanced ICT Research Institute, National Institute of Information and Communications Technology, Suita, Japan.
Frontiers in neuroscience
|June 27, 2024
概括
本研究介绍了一个简单的,类似于Hebbian的过程,用于深度学习图像分类器中的一次性学习. 这种方法可以通过减少干扰从单个图像中学习新概念,为AI和神经科学提供了洞察力.
科学领域:
- 人工智能的人工智能
- 神经科学是一个神经科学.
- 机器学习 机器学习
背景情况:
- 一次性学习,模仿人类大脑功能,对于AI发展至关重要.
- 当前的方法,如重量印记,用于深度学习中的一次性学习,具有局限性,包括神经科学相关性的模糊性和对原始分类的干扰.
- 人工神经网络的进步为模拟大脑功能和开发新的AI能力提供了机会.
研究的目的:
- 在深度学习图像分类器中提出一种新,简单和实用的方法,用于一次性类添加.
- 将重量印记过程与赫比规则保持一致,提供基于神经科学的方法.
- 为了减少对原始分类的干扰,同时可以从单个实例中学习.
主要方法:
- 一个单一的Hebbian-like过程被采用,使预先训练的深度学习图像分类器能够一次性添加类.
- 非参数规范化用于模仿大脑快速的赫比安可塑性.
- 拟议的方法在没有修改原始分类器的骨干的情况下运行.
主要成果:
- 类似于Hebbian的过程成功地实现了一次性类添加,而不会改变分类器的核心架构.
- 使用非参数规范化显著降低了以前在类似方法中看到的干扰问题.
- 这种方法证明了对一次性学习任务的高实用性和简单性.
结论:
- 这项研究提出了一个简单而有效的Hebbian-like机制,用于深度学习中的一次性类添加.
- 该方法为神经科学提供了一个功能有效的假设,将AI机制与大脑功能联系起来.
- 这些发现有助于开发更高效,更易于解释的AI系统,其灵感来源于生物学习过程.
更多相关视频
09:34A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
Published on: September 25, 2021
3.9K
05:41A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
Published on: February 6, 2020
9.4K
相关概念视频
01:25Long-term Potentiation
2.8K
Long-term potentiation, or LTP, is one of the ways by which synaptic plasticity—changes in the strength of chemical synapses—can occur in the brain. LTP is the process of synaptic strengthening that occurs over time between pre and postsynaptic neuronal connections. The synaptic strengthening of LTP works in opposition to the synaptic weakening of long-term depression (LTD) and together are the main mechanisms that underlie learning and memory.
Hebbian LTP
LTP can occur when...
Hebbian LTP
LTP can occur when...
2.8K
01:25Neural Circuits
1.1K
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...
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...
1.1K
01:19The Role of Ion Channels in Neuronal Computation
3.2K
A postsynaptic neuron usually receives numerous impulses from several other presynaptic neurons. The axon hillock of the postsynaptic neuron integrates all these signals and determines the likelihood of firing an action potential.
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
3.2K
01:01Neuroplasticity
322
Neuroplasticity reflects the brain's remarkable capacity to adapt and evolve, responding dynamically to learning, experiences, or injury by reorganizing its neural circuitry. This reorganization involves creating new neural connections and refining old ones through a series of biological processes that contribute to the brain's lifelong development and adaptability.
322
01:21Neurons: The Axon
3.4K
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....
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....
3.4K
01:28Integration of Synaptic Events
1.5K
Synaptic integration mainly includes the summation of graded potentials. Graded potentials, regardless of their type, cause subtle alterations in membrane voltage, resulting in either depolarization or hyperpolarization. These incremental changes, when combined or summed, can propel the neuron toward its threshold. Consider, for example, a membrane experiencing a +15 mV shift, causing it to depolarize from -70 mV to -55 mV. In this scenario, graded potentials govern the membrane's ability to...
1.5K