.png){kind=icon}
返回
Learning hierarchically-structured concepts
.png){kind=icon}
OA
AI
DOI：10.1016/j.neunet.2021.07.033.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
We use a recently developed synchronous Spiking Neural Network (SNN) model to study the problem of learning hierarchically-structured concepts. We introduce an abstract data model that describes simple hierarchical concepts. We define a feed-forward layered SNN model, with learning modeled using Oja's local learning rule, a well known biologically-plausible rule for adjusting synapse weights. We define what it means for such a network to recognize hierarchical concepts; our notion of recognition is robust, in that it tolerates a bounded amount of noise. Then, we present a learning algorithm by which a layered network may learn to recognize hierarchical concepts according to our robust definition. We analyze correctness and performance rigorously; the amount of time required to learn each concept, after learning all of the sub-concepts, is approximately O (1/eta k (l(max) log(k) + 1/epsilon) + b log(k)), where k is the number of sub-concepts per concept, l(max) is the maximum hierarchical depth, eta is the learning rate, epsilon describes the amount of uncertainty allowed in robust recognition, and b describes the amount of weight decrease for irrelevant'' edges. An interesting feature of this algorithm is that it allows the network to learn sub-concepts in a highly interleaved manner. This algorithm assumes that the concepts are presented in a noise-free way; we also extend these results to accommodate noise in the learning process. Finally, we give a simple lower bound saying that, in order to recognize concepts with hierarchical depth two with noise-tolerance, a neural network should have at least two layers. The results in this paper represent first steps in the theoretical study of hierarchical concepts using SNNs. The cases studied here are basic, but they suggest many directions for extensions to more elaborate and realistic cases. (C) 2021 Elsevier Ltd. All rights reserved.
Keyword:
Hierarchical concepts
Representing hierarchical concepts
Recognizing hierarchical concepts
Learning hierarchical concepts
Spiking Neural Networks
Brain-inspired algorithms
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6.3
论文数:
8.2K
被引数:
3.0W
机构
引用论文
Eligibility Traces and Plasticity on Behavioral Time Scales: Experimental Support of NeoHebbian Three-Factor Learning Rules行为时间尺度上的资格痕迹和可塑性: NeoHebbian三因素学习规则的实验支持OAAI
OAAI分享收藏Networks of spiking neurons: The third generation of neural network models尖峰神经元网络: 第三代神经网络模型.png)
NEURAL NETWORKS
NEURAL NETWORKSIF6.3
PREAI
分享收藏Long-term outcome of endoscopic shortening and stiffening of the vocal folds to raise the pitch内窥镜缩短和加强声带以提高音调的长期结果
PREAI
分享收藏LONG-TERM DEPRESSION OF EXCITATORY SYNAPTIC TRANSMISSION AND ITS RELATIONSHIP TO LONG-TERM POTENTIATIONTRENDS IN NEUROSCIENCES
TRENDS IN NEUROSCIENCESIF15.1
PREAI
分享收藏
查看更多内容