Toward the future of causal ML

In this section, we’ll briefly explore some possible future directions for causality from business, application, and research point of views. As always when talking about the future, this is somewhat of a gamble, especially in the second part of this section where we will discuss more advanced ideas.

Let’s start our journey into the future from where we’re currently standing.

Where are we now and where are we heading?

With an average of 3.2 new papers published on arXiv every day in 2022, causal inference has exploded in popularity, attracting a large amount of talent and interest from top researchers and institutions, including industry giants such as Amazon or Microsoft.

At the same time, for many organizations, causal methods are much less accessible than traditional statistical and machine learning techniques. This state of affairs is likely driven by a strong focus of educational system on associational methods when...

Learning causality

In this section, we’ll point to the resources to learn more about causality after finishing this book.

For many people starting with causality, their learning path begins with excitement. The promise of causality is attractive and powerful. After learning about the basics and realizing the challenges that any student of causality has to face, many of us lose hope in the early stages of our journeys.

Some of us regain it, learning that solutions do exist, although not necessarily where we initially expected to find them.

After overcoming the first challenges and going deeper into the topic, many of us realize that there are more difficulties to come. Learning from earlier experiences, it’s easier at this stage to realize that (many of) these difficulties can be tackled using a creative and systematic approach.

I like the way the Swiss educator and researcher Quentin Gallea presented the journey into learning causality in a graphical form...

Let’s stay in touch

Community is a catalyst for growth. Let’s connect on LinkedIn and Twitter so that we can learn from each other:

• LinkedIn: https://www.linkedin.com/in/aleksandermolak/
• Twitter: @AleksanderMolak

If you want to consult a project or run a workshop on causality for your team, drop me a line at [e-mail].

For comments and questions regarding this book, email me at [e-mail].

Wrapping it up

It’s time to conclude our journey.

In this chapter, we summarized what we’ve learned in this book, discussed five steps to make the best out of our causal projects, took a look at the intersection of causality and business, and sneaked into the (potential) future of causal research and applications. Finally, we listed a number of resources that you mind find useful in the next stages of your causal journey.

I hope finishing this book won't be the end for you, but rather the beginning of a new causal chapter!

I hope to see you again!

References

Bareinboim, E., & Pearl, J. (2016). Causal inference and the data-fusion problem. Proceedings of the National Academy of Sciences of the United States of America, 113(27), 7345–7352.

Berrevoets, J., Kacprzyk, K., Qian, Z., & van der Schaar, M. (2023). Causal Deep Learning. arXiv.

Chau, S. L., Ton, J.-F., González, J., Teh, Y., & Sejdinovic, D. (2021). BayesIMP: Uncertainty Quantification for Causal Data Fusion.

In M. Ranzato, A. Beygelzimer, Y. Dauphin, P. S. Liang, & J. W. Vaughan (Eds.), Advances in Neural Information Processing Systems, 34, 3466–3477. Curran Associates, Inc.

Curth, A., Svensson, D., Weatherall, J., & van der Schaar, M. (2021). Really Doing Great at Estimating CATE? A Critical Look at ML Benchmarking Practices in Treatment Effect Estimation. Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks.

Deng, Z., Zheng, X., Tian, H., & Zeng, D. D. (2022). Deep Causal...

Learning causality

In this section, we’ll point to the resources to learn more about causality after finishing this book.

For many people starting with causality, their learning path begins with excitement. The promise of causality is attractive and powerful. After learning about the basics and realizing the challenges that any student of causality has to face, many of us lose hope in the early stages of our journeys.

Some of us regain it, learning that solutions do exist, although not necessarily where we initially expected to find them.

After overcoming the first challenges and going deeper into the topic, many of us realize that there are more difficulties to come. Learning from earlier experiences, it’s easier at this stage to realize that (many of) these difficulties can be tackled using a creative and systematic approach.

I like the way the Swiss educator and researcher Quentin Gallea presented the journey into learning causality in a graphical form...

Let’s stay in touch

Community is a catalyst for growth. Let’s connect on LinkedIn and Twitter so that we can learn from each other:

• LinkedIn: https://www.linkedin.com/in/aleksandermolak/
• Twitter: @AleksanderMolak

If you want to consult a project or run a workshop on causality for your team, drop me a line at [e-mail].

For comments and questions regarding this book, email me at [e-mail].

Wrapping it up

It’s time to conclude our journey.

In this chapter, we summarized what we’ve learned in this book, discussed five steps to make the best out of our causal projects, took a look at the intersection of causality and business, and sneaked into the (potential) future of causal research and applications. Finally, we listed a number of resources that you mind find useful in the next stages of your causal journey.

I hope finishing this book won't be the end for you, but rather the beginning of a new causal chapter!

I hope to see you again!

References

Bareinboim, E., & Pearl, J. (2016). Causal inference and the data-fusion problem. Proceedings of the National Academy of Sciences of the United States of America, 113(27), 7345–7352.

Berrevoets, J., Kacprzyk, K., Qian, Z., & van der Schaar, M. (2023). Causal Deep Learning. arXiv.

Chau, S. L., Ton, J.-F., González, J., Teh, Y., & Sejdinovic, D. (2021). BayesIMP: Uncertainty Quantification for Causal Data Fusion.

In M. Ranzato, A. Beygelzimer, Y. Dauphin, P. S. Liang, & J. W. Vaughan (Eds.), Advances in Neural Information Processing Systems, 34, 3466–3477. Curran Associates, Inc.

Curth, A., Svensson, D., Weatherall, J., & van der Schaar, M. (2021). Really Doing Great at Estimating CATE? A Critical Look at ML Benchmarking Practices in Treatment Effect Estimation. Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks.

Deng, Z., Zheng, X., Tian, H., & Zeng, D. D. (2022). Deep Causal...

Join our book's Discord space

Join our Discord community to meet like-minded people and learn alongside more than 2000 members at: https://packt.link/infer