CyberFish – 正面系列

CyberFish - Front Series（CyberFish - 正面系列）CyberFish是始于2021年的机器学习系列作品，作者罗霄将实验艺术和计算编程两套思维模式，以数据集和算法的形式交给神经网络进行学习训练，所产出的AI模型承载了作者本人以及AI自我的双重意识和技术特性。“CyberFish–Front Series” 更新于2024年，在初代基础上进化而来，使用生成对抗网络（StyleGAN）进行模型训练和视觉生成，并加入了着色器像素排序算法。这一代CyberFish模型在训练时，使用了扩散模型（Diffusion Model）创建数据集，这与初代采集自人工插画的数据集完全不同。是在反思CyberFish 向完全智能体进化过程中的信息来源问题，也是与两代经典神经网络（GAN、Diffusion）的对话。

CyberFish - Front Series (CyberFish - Front Series) CyberFish is a machine learning series work that began in 2021. The author, Luo Xiao, combines experimental art and computational programming, presenting these two thinking modes in the form of datasets and algorithms to the neural network for learning and training. The AI model produced by this process carries the dual consciousness and technical characteristics of both the author and the AI itself. "CyberFish - Front Series" was updated in 2024, evolving from the original version. It uses the Generative Adversarial Network (StyleGAN) for model training and visual generation, and incorporates the shader pixel sorting algorithm. In this generation of CyberFish models, during training, a diffusion model (Diffusion Model) was used to create the dataset, which is completely different from the dataset collected from hand-drawn illustrations in the original version. This is a reflection on the issue of information sources in the evolution process of CyberFish towards a fully intelligent entity, and it is also a dialogue with two generations of classic neural networks (GAN, Diffusion).