相关实验视频

Updated: Jun 8, 2025

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

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

Published on: March 2, 2015

你的艺术有多深:在单任务,单模式神经网络中对艺术理解的极限进行实验研究

Mahan Agha Zahedi¹, Niloofar Gholamrezaei², Alex Doboli¹

¹Department of Electrical and Computer Engineering, Stony Brook University, Stony Brook, New York, United States of America.

|November 6, 2024

概括

深度卷积神经网络 (DCNNs) 可以根据视觉特征对概念艺术进行分类. 然而,这些机器学习模型忽略了关键的背景,如艺术家的意图,突出了当前AI在艺术解释方面的局限性.

科学领域:

计算机视觉计算机视觉
人工智能的人工智能
艺术史艺术史艺术史

背景情况:

艺术解释是主观和复杂的,这给计算分析带来了挑战.
深度卷积神经网络 (DCNNs) 是用于图像识别的先进机器学习模型.

研究的目的:

调查DCNN是否可以将概念艺术作品分类为策划画廊.
测试DCNN是否依赖视觉 (展示) 或上下文 (非展示) 属性进行分类.

主要方法:

一个VGG-11 DCNN,在ImageNet上进行预训练,进行了微调.
该模型是在概念摄影画廊手工制作的数据集上训练的.
关于财产使用的两个假设被实验验证.

主要成果:

DCNN的分类与策展画廊的任务保持一致.
结果支持了DCNN仅使用视觉属性 (形状,颜色) 的假设.
DCNN模型忽略了非展示属性,如历史背景和艺术家的意图.

结论:

目前的DCNN仅限于根据视觉属性对艺术进行分类.
未来的深度神经网络 (DNN) 模型需要将更广泛的上下文理解纳入艺术分析.
这项研究揭示了人工智能目前无法掌握概念艺术的全部含义.

更多相关视频

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

相关概念视频

Neuroplasticity

Neuroplasticity

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.