Artificial neural networks
❏ A Study on Developmental Artificial Neural Networks by Integrating Lower-order Functions
Overview
In the conventional deep learning, the topology of the networks is determined by learning the network parameters. This leads to a lot of useless learning. On the other hand, animals develop innate fundamental ability, called instinct, from their experiences in the environment, and acquire behavioral knowledge with higher-order functions. Such intellectual development can be captured as ”development” rather than ”learning”. In this study, we adopt Weight Agnostic Neural Networks (WANN), a model that captures the topology of networks through their development, as a fundamental technology. We aim at constructing Developmental Artificial Neural Networks (DANNs) that acquire hierarchical relationships of functions, in which higher-order functions expressed from pre-existing lower- order ones. We took up a higher order human behavior, jump forward, and tried to acquire the behavior from two pre-existing lower-order behaviors, walk and jump, applying WANN framework to developing the network including the neural networks of those two behaviors. As a result, we understood that it is necessary to more elaborate our ideas for DANN.
Slides
iwai-01
iwai-02
iwai-03
iwai-04
iwai-05
iwai-06
iwai-07
iwai-08
iwai-09
iwai-10
iwai-11
iwai-12
iwai-13
iwai-14
iwai-15
iwai-16
iwai-17
iwai-18
iwai-19
iwai-20
iwai-21
iwai-22
iwai-23
iwai-24
iwai-25
iwai-26
iwai-27
iwai-28
iwai-29
iwai-30
iwai-31
iwai-32
iwai-33
iwai-34
iwai-35
iwai-36
previous arrow
next arrow
Haruka Iwai
岩井 遥,小林 一郎「低次機能を融合する発達型人工神経回路網構築への取り組み」人工知能学会全国大会(第36回),国立京都国際会館,京都,2022年6月.(in Japanese)