Timezone: »

From 'What If?' To 'What Next?' : Causal Inference and Machine Learning for Intelligent Decision Making
Ricardo Silva · Panagiotis Toulis · John Shawe-Taylor · Alexander Volfovsky · Thorsten Joachims · Lihong Li · Nathan Kallus · Adith Swaminathan

Fri Dec 08 08:30 AM -- 06:30 PM (PST) @ Hall C
Event URL: https://sites.google.com/view/causalnips2017 »

In recent years machine learning and causal inference have both seen important advances, especially through a dramatic expansion of their theoretical and practical domains. Machine learning has focused on ultra high-dimensional models and scalable stochastic algorithms, whereas causal inference has been guiding policy in complex domains involving economics, social and health sciences, and business. Through such advances a powerful cross-pollination has emerged as a new set of methodologies promising to deliver robust data analysis than each field could individually -- some examples include concepts such as doubly-robust methods, targeted learning, double machine learning, causal trees, all of which have recently been introduced.

This workshop is aimed at facilitating more interactions between researchers in machine learning and causal inference. In particular, it is an opportunity to bring together highly technical individuals who are strongly motivated by the practical importance and real-world impact of their work. Cultivating such interactions will lead to the development of theory, methodology, and - most importantly - practical tools, that better target causal questions across different domains.

In particular, we will highlight theory, algorithms and applications on automatic decision making systems, such as recommendation engines, medical decision systems and self-driving cars, as both producers and users of data. The challenge here is the feedback between learning from data and then taking actions that may affect what data will be made available for future learning. Learning algorithms have to reason about how changes to the system will affect future data, giving rise to challenging counterfactual and causal reasoning issues that the learning algorithm has to account for. Modern and scalable policy learning algorithms also require operating with non-experimental data, such as logged user interaction data where users click ads suggested by recommender systems trained on historical user clicks.

To further bring the community together around the use of such interaction data, this workshop will host a Kaggle challenge problem based on the first real-world dataset of logged contextual bandit feedback with non-uniform action-selection propensities. The dataset consists of several gigabytes of data from an ad placement system, which we have processed into multiple well-defined learning problems of increasing complexity, feedback signal, and context. Participants in the challenge problem will be able to discuss their results at the workshop.

Fri 8:30 a.m. - 8:45 a.m. [iCal]
Introductions (Panel)
Panos Toulis, Alexander Volfovsky
Fri 8:45 a.m. - 9:20 a.m. [iCal]

We know how to spot object in images, but we must learn on more images than a human can see in a lifetime. We know how to translate text (somehow), but we must learn it on more text than a human can read in a lifetime. We know how to learn playing Atari games, but we must learn it by playing more games than any teenager can endure. The list is long. We can of course try to pin this inefficiently to some properties of our algorithms. However, we can also take the point of view that there is possibly a lot of signal in natural data that we simply do not exploit. I will report on two works in this direction. The first one establishes that something as simple as a collection of static images contains nontrivial information about the causal relations between the objects they represent. The second one, time permitting, shows how an attempt to discover such a structure in observational data led to a clear improvement of Generative Adversarial Networks.

Leon Bottou
Fri 9:20 a.m. - 9:55 a.m. [iCal]
Invited Talk
Emma Brunskill
Fri 10:00 a.m. - 10:15 a.m. [iCal]
Contributed Talk 1 (Contributed Talks)
David Heckerman
Fri 10:15 a.m. - 10:30 a.m. [iCal]
Contributed Talk 2 (Contributed Talks)
Jovana Mitrovic
Fri 11:00 a.m. - 11:35 a.m. [iCal]
Invited Talk 3 (Invited Talk)
Richard Hahn
Fri 11:35 a.m. - 12:10 p.m. [iCal]
Invited Talk 4 (Invited Talk)
David Sontag
Fri 12:10 p.m. - 1:35 p.m. [iCal]
Poster session
Abbas Zaidi, Christoph Kurz, David Heckerman, YiJyun Lin, Stefan Riezler, Ilya Shpitser, Songbai Yan, Olivier Goudet, Yash Deshpande, Judea Pearl, Jovana Mitrovic, Brian Vegetabile, Tae Hwy Lee, Karen Sachs, Karthika Mohan, Reagan Rose, Julius Ramakers, Negar Hassanpour, Pierre Baldi, Razieh Nabi, Noah Hammarlund, Eli Sherman, Carolin Lawrence, Fattaneh Jabbari, Vira Semenova, Maria Dimakopoulou, Pratik Gajane, Russell Greiner, Ilias Zadik, Alex Blocker, Hao Xu, Tal EL HAY, Tony Jebara, Benoit Rostykus
Fri 1:35 p.m. - 1:50 p.m. [iCal]
Contributed Talk 3 (Contributed Talks)
Ilya Shpitser
Fri 1:50 p.m. - 2:05 p.m. [iCal]
Contributed Talk 4 (Contributed Talks)
Judea Pearl
Fri 2:05 p.m. - 2:40 p.m. [iCal]
Invited Talk 5 (Invited Talk)
guido imbens
Fri 3:30 p.m. - 4:05 p.m. [iCal]
Invited Talk
Bin Yu
Fri 4:05 p.m. - 5:00 p.m. [iCal]
Causal inference with machine learning (Discussion Panel)
Fri 5:00 p.m. - 6:00 p.m. [iCal]
Causality and Machine Learning Challenge: Criteo Ad Placement Challenge (Posters, discussion and talks)

Author Information

Ricardo Silva (University College London)
Panos Toulis (University of Chicago)
John Shawe-Taylor (UCL)

John Shawe-Taylor has contributed to fields ranging from graph theory through cryptography to statistical learning theory and its applications. However, his main contributions have been in the development of the analysis and subsequent algorithmic definition of principled machine learning algorithms founded in statistical learning theory. This work has helped to drive a fundamental rebirth in the field of machine learning with the introduction of kernel methods and support vector machines, driving the mapping of these approaches onto novel domains including work in computer vision, document classification, and applications in biology and medicine focussed on brain scan, immunity and proteome analysis. He has published over 300 papers and two books that have together attracted over 60000 citations. He has also been instrumental in assembling a series of influential European Networks of Excellence. The scientific coordination of these projects has influenced a generation of researchers and promoted the widespread uptake of machine learning in both science and industry that we are currently witnessing.

Alexander Volfovsky (Duke University)
Thorsten Joachims (Cornell)
Lihong Li (Google Brain)
Nathan Kallus (Cornell University)
Adith Swaminathan (Microsoft Research)

More from the Same Authors