|
所在平台: Coursera |
课程主页: https://www.coursera.org/learn/ai-data-bias
课程评论：没有评论
课程名称：人工智能数据公平性与偏见 概述：本课程将探讨机器学习中公平性和偏见的基本问题。随着预测模型在重要决策（如大学录取和贷款决策）中发挥作用，避免模型做出不公正预测变得至关重要。从人类偏见到数据集意识，我们将探讨构建更具伦理模型的诸多方面。 课程大纲： 1. **机器学习中的公平性与保护** 描述：欢迎来到课程！在第一周，我们将讨论在机器学习背景下公平性意味着什么，以及在不同场景下真实平等的含义。 2. **构建公平模型：理论与实践** 描述：本周我们将采取措施对抗不公平性。在我们了解公平性问题后，如何构建不会违反这些原则的模型？ 3. **人类因素：尽量减少数据中的偏见** 描述：这一周，我们将解决进入数据收集和属性选择过程的人类偏见。目标是在模型构建之前消除偏见。
Name:Fairness and protections in machine learning
Description:Welcome to the course! In week one, we will be discussing what fairness means in the context of machine learning and what true parity means in different scenarios
Name:Building fair models: theory and practice
Description:This week we will take action against unfairness. Now that we have an understanding of fairness issues, how do we build models that do not violate them?
Name:Human factors: minimizing bias in data
Description:This week, we will tackle the human biases that enter the data collection and attribute selection processes. The goal? Removing bias before the model is built
In this course, we will explore fundamental issues of fairness and bias in machine learning. As predictive models begin making important decisions, from college admission to loan decisions, it becomes paramount to keep models from making unfair predictions. From human bias to dataset awareness, we will explore many aspects of building more ethical models.
