TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

What will Machine Learning look like in future?

3 点作者 vkkhare超过 3 年前
Let&#x27;s take an example of recommendation engines. I think a truly decentralised content curation will be based on edge computing.<p>Twitter, FB (Meta), Youtube everyone harvests user data and train their recommendation engines which are then monetised by them (often unfairly).<p>In the future, the data stays on the users&#x27; devices and anyone can train their models by asking the user for the consent. The data never leaves the device and ML models get trained on user device itself. The users get to choose from a host of recommendation models (from twitter, WSJ, &lt;some_random_company&gt;) and can ask for payment in return for using their data. So no one party can build a monopoly over the platform.

暂无评论

暂无评论