科技新闻_每日一闻：Microsoft reveals DoWhy library for causal inference

文章出处：https://sdtimes.com/softwaredev/sd-times-news-digest-googles-vr-labs-armorys-series-a-funding-and-microsofts-dowhy-library/

DoWhy is a Python library that makes it easy to estimate causal effects. DoWhy is based on a unified language for causal inference, combining causal graphical models and potential outcomes frameworks. http://causalinference.gitlab.io/dowhy

DoWhy是一个Python库，可以很容易地估计因果效应。 DoWhy基于统一的因果推理语言，结合因果图形模型和潜在的结果框架。

https://github.com/Microsoft/dowhy

Microsoft reveals DoWhy library for causal inference

Microsoft has announced a new library called DoWhy. DoWhy provides a programmatic interface for causal inference. It was designed to highlight the often neglected assumptions that underlie causal inference analyses.

它旨在强调构成因果推断分析的经常被忽视的假设。

The motivation for creating the library is that in causal inference studies, Microsoft found themselves repeating common steps of finding the right identification strategy, devising a suitable estimator, and conduction robustness checks, the company explained.

该公司解释说，创建该库的动机是，在因果推理研究中，微软发现自己重复寻找正确识别策略，设计合适的估算器以及传导稳健性检查的常用步骤。

devising ：发明  devise：vt. 设计；想出；发明；图谋；遗赠给；conduction：n.传导

Microsoft designed the library following two principles: making causal assumptions explicit and testing the robustness of estimates when those assumptions were violated.

violated：v. 违反（violate的过去分词）

Microsoft根据两个原则设计了该库：明确了因果假设，并在违反这些假设时测试估计的稳健性。

2.Hortonworks adds news tools and updates to existing platforms

Hortonworks为现有平台添加了新闻工具和更新。

Hortonworks has announces innovations that will enable customers to get insights on data generated at the edge.

Hortonworks宣布了一系列创新，使客户能够深入了解边缘生成的数据。

It has announced Streams Messaging Manager, which is an open-source operations monitoring tool that offers end-to-end visibility in enterprise Kafka environments.

它已经发布了Streams Messaging Manager，这是一个开源操作监控工具，可在企业Kafka环境中提供端到端的可见性。

The company also released a new version of Hortonworks DataFlow, 3.2. The new version makes operations simpler, provides stronger integration and interoperability between Hortonworks DataFlow and Hortonworks Data Platform 3.0, and improves performance for data-in-motion in hybrid environments, it explained. Key features include enhanced platform resiliency, more granular control, increased performance, and consistent management for Hortonworks DataFlow and Hortonworks Data Platform.

该公司还发布了新版本的Hortonworks DataFlow，3.2。它解释说，新版本使操作更简单，在Hortonworks DataFlow和Hortonworks Data Platform 3.0之间提供更强的集成和互操作性，并提高混合环境中的动态数据性能。主要功能包括增强的平台弹性，更精细的控制，更高的性能以及Hortonworks DataFlow和Hortonworks数据平台的一致管理。

interoperability ：n. [计] 互操作性；互用性

resiliency：/rɪ'zɪlɪənsɪ/ n. 弹性；跳回

granular：adj. 颗粒的；粒状的

https://github.com/Microsoft/dowhy

1.1DoWhy | Making causal inference easy

As computing systems are more frequently and more actively intervening in societally critical domains such as healthcare, education and governance, it is critical to correctly predict and understand the causal effects of these interventions. Without an A/B test, conventional machine learning methods, built on pattern recognition and correlational analyses, are insufficient for causal reasoning。

随着计算系统更频繁和更积极地干预社会关键领域，如医疗保健，教育和治理，正确预测和理解这些干预措施的因果效应至关重要。在没有A / B测试的情况下，基于模式识别和相关分析的传统机器学习方法不足以进行因果推理。

Much like machine learning libraries have done for prediction, "DoWhy" is a Python library that aims to spark causal thinking and analysis. DoWhy provides a unified interface for causal inference methods and automatically tests many assumptions, thus making inference accessible to non-experts.

就像机器学习库为预测所做的那样，“DoWhy”是一个旨在激发因果思考和分析的Python库。 DoWhy为因果推理方法提供了统一的界面，并自动测试了许多假设，从而使非专家可以进行推理。

spark ：n. vi. vt.  n. 火花；朝气；闪光；vi. 闪烁；发火花；求婚；vt. 发动；鼓舞；求婚

For a quick introduction to causal inference, check out this tutorial. Documentation for DoWhy is available at causalinference.gitlab.io/dowhy.

Contents

• The need for causal inference
• Sample causal inference analysis in DoWhy
• Installation
• Graphical Models and Potential Outcomes: Best of both worlds
• A unifying language for causal inference
• Model a causal problem
• Identify a target estimand under the model
• Estimate causal effect based on the identified estimand
• Refute the obtained estimate
• Contributing

The need for causal inference

Predictive models uncover patterns that connect the inputs and outcome in observed data. To intervene, however, we need to estimate the effect of changing an input from its current value, for which no data exists. Such questions, involving estimating a counterfactual, are common in decision-making scenarios.

预测模型揭示了连接观察数据中的输入和结果的模式。但是，为了进行干预，我们需要估计从当前值更改输入的效果，因为没有数据存在。涉及估计反事实的此类问题在决策方案中很常见。

uncover :vt. 发现；揭开；揭露;vi. 发现；揭示；揭去盖子

intervene:vi. 干涉；调停；插入

counterfactual: /,kaʊntə'fæktʃʊəl; -tjʊəl/   adj. 反事实的

• Will it work?

  • Does a proposed change to a system improve people's outcomes?

• Why did it work?

  • What led to a change in a system's outcome?

• What should we do?

  • What changes to a system are likely to improve outcomes for people?

• What are the overall effects?

  • How does the system interact with human behavior?
  • What is the effect of a system's recommendations on people's activity?

  • 系统建议对人们活动的影响是什么？

Answering these questions requires causal reasoning. While many methods exist for causal inference, it is hard to compare their assumptions and robustness of results. DoWhy makes three contributions,

1. Provides a principled way of modeling a given problem as a causal graph so that all assumptions explicit.
2. Provides a unified interface for many popular causal inference methods, combining the two major frameworks of graphical models and potential outcomes.
3. Automatically tests for the validity of assumptions if possible and assesses robustness of the estimate to violations.

Microsoft reveals DoWhy library for causal inference

微软揭示了DoWhy库的因果推断

Python