评论
打印
Visualizing our dreams
未来AI或能帮人们记录梦境
After waking up, you may feel frustrated that you cannot recall the dreams you had last night.
如果醒来后记不得昨晚做的梦，你可能会觉得沮丧，
Artificial intelligence (AI) may be able to help.
而人工智能（AI）或许可以帮上忙。
Previously, there have been AI models that can turn text into images.
此前，研究人员已创建出能够将文本转换为图像的AI模型，
They can do this by learning from a large amount of data from both tests and images.
AI可以学习来自测试和图像中的大量数据，从而实现转换。
This time, researchers from Osaka University in Japan have trained an AI system called Stable Diffusion
据《科学》杂志报道，这一次，日本大阪大学的研究团队训练了一个名为“稳定扩散”的AI模型，
to re-create images based on people’s brain scans, reported Science magazine.
基于人们的大脑扫描再生成图像。
The researchers used an online data set provided by the University of Minnesota, US,
研究人员使用了美国明尼苏达大学提供的在线数据集，
which consisted of brain scans from four participants as they each viewed a set of 10,000 photos.
其中包括4名参与者观看10000张照片时的脑部扫描结果，
The scans were recorded by functional magnetic resonance imaging (fMRI).
该结果由功能性磁共振成像（fMRI）记录。
The AI then learned about the brain activities by analyzing changes in blood flow shown by the fMRI data –
随后，AI通过分析fMRI数据显示的血流变化了解大脑活动——
when a part of the brain is activated, more blood will flow to it.
大脑的一部分激活时，会有更多的血液流向该区域，
It then matched the brain activities with the photos.
而后便能将大脑活动与照片相匹配。
Through this method, AI learned how human brains would react when seeing different photos.
通过这种方法，AI了解到人类大脑观看不同照片时的反应。
Finally, the researchers tested the AI on additional brain scans from the same participants when they viewed photos of a toy bear,airplane, clock and train.
最后，研究人员对其进行了额外的脑部扫描，在同一批参与者观看玩具熊、飞机、时钟和火车的照片时。
If the person looked at an airplane, for example, the AI would use the brain scan data to create an image of a very blurry airplane.
例如，一个人看着一架飞机时，AI会使用大脑扫描数据生成一个非常模糊的飞机图像。
Then, it would turn on the previous “text-to-image” model and improve the quality of the image by feeding itself the keyword “airplane”.
然后，AI会启动之前的“文本转图像”模型，并给自己“喂”下“飞机”这一关键词，以提高图像的质量。
The final images were “convincing” with about 80 percent of accuracy, according to the researchers.
据研究人员称，最终得到的图像“几可乱真”，且准确率高达约80%。
The new study created a novel approach that incorporates texts and images to “decipher the brain”,
这项新研究创造了一种结合文本和图像来“破译”大脑的新方法，
Ariel Goldstein from Princeton University, US, told Science magazine.
美国普林斯顿大学的科学家Ariel Goldstein在接受《科学》杂志采访时表示。
In the future, scientists hope that the technology can be used to record imagined thoughts and dreams
未来，科研人员们希望这项技术可以用来记录幻想和梦境，
or allow people to understand how differently other animals perceive reality.
或是让人们了解其他动物对现实的感知有何不同。
以上文章内容选自《21世纪学生英文报》高三808期
