I am a senior scientist in the Department of Computer Science at ETH Zurich, Switzerland. I love to learn and explain things, and strive to understand the universe better.
Convinced that Einstein was right all along about quantum theory not being a complete description of reality, I am spending a lot of my time on proving him right, with a deterministic model that brings game theory into physics, and that turns quantum experiments into (put simply) a huge game played between experimenters and the universe that can be solved with novel, non-Nashian solution concepts that circumvent impossibility theorems (see published papers and preprints in the publication section).
Topics of interest
|Fall 2022||Big Data|
|Fall 2022||Information Systems for Engineers|
|Spring 2022||Big Data for Engineers|
|Spring 2022||Information Retrieval|
|Fall 2021||Big Data|
|Fall 2021||Information Systems for Engineers|
|Spring 2021||Big Data for Engineers|
|Spring 2021||Information Retrieval|
|Fall 2020||Big Data|
|Fall 2020||Information Systems for Engineers|
|Spring 2020||Big Data for Engineers|
|Spring 2020||Information Systems for Engineers|
|Fall 2019||Big Data|
|Fall 2019||Information Retrieval|
|Spring 2019||Big Data for Engineers|
|Spring 2019||Information Retrieval|
|Fall 2018||Big Data|
|Fall 2018||Information Systems for Engineers|
|Spring 2018||Big Data for Engineers|
|Spring 2018||Information Retrieval|
|Fall 2017||Big Data|
|Fall 2017||Information Systems for Engineers|
|Fall 2016||Big Data|
Non-nashian game theory
We live in a world where AI/ML technology is getting better every day at predicting the action of human beings, leading to always more interdependent decisions. Yet, the discipline that is dedicated to the modelling of what rational agents should or would do, game theory, is commonly based on a strong assumption of free choice in which agents are by essence unpredictable: in the Nash paradigm, economic agents optimize their decisions by freezing what everybody else does and unilaterally considering their possible decisions. The domain of validity of Nashian game theory is thus increasingly challenged by the progress made in data science.
I am investigating what happens if, instead, we assume that rational agents are perfectly predictable, an idea that Prof. J.-P. Dupuy (Stanford) coined in the 1990s. Free choice can be reconciled with predictability with a fixpoint framework in which a decision is entangled with its always correct prediction, and this decision must be immune to its prediction in terms of causality. In other words, the correct prediction of a decision must trigger a chain of events that causes this very decision to be taken (self-fulfilling prophecy). This kind of thinking is similar to what the Director of a central bank tries to avoid on a macroscopic level: saying too boldly that the economy is doing well may lead to a market crash, contradicting their prediction, since interest rate hikes would then be expected by the public.
I co-designed the corresponding non-Nashian equilibrium concepts and algorithms for games in both extensive and normal form. As predicted in a conjecture by Prof. Dupuy, notably, Pareto optimal outcomes are always reached. Concretely, this means that, unlike in the Nash paradigm, it is rational to be honest and not betray other agents. Philosophically, the underlying concept is that of a Kantian categorical imperative urging to avoid Grandfather's paradoxes.
Future directions of research include investigating the feasibility of programming robots and artificial intelligences with non-Nashian algorithms, in order to produce a benevolent AI system.
Deterministic quantum theory
Another direction of research is the link with quantum theory: indeed, the core assumption behind the inherent non-determinism of quantum theory is the same one as that of the Nash paradigm, namely, the free choice of the physicist who unpredictably picks a measurement axis. It is used, for example, in combination with the Kochen-Specker impossibility theorem.
However, this strong free choice assumption seems to be in contradiction with some findings of the neuroscience community, indicating that a human may only have a veto right on their decisions.
I am thus building a model that weakens the free choice assumption in quantum physics, along the same lines as in non-Nashian game theory, namely, that the correct prediction of an agent's action could be entangled with their action, and nevertheless causing it to happen. This decision model is based firstly on a generalization of game theory to special relativity to account for spacelike and timelike separated choices of measurement axes and of measurement outcomes -- in a structure compatible with and familiar to the quantum foundations community. Secondly, this decision model can be seen as a big game between experimenters and the universe, who both maximize their utility (e.g., principle of least action), leading to a deterministic resolution predicting an outcome for each experiment, without probabilities. We have established that the equilibrium cannot be a Nash equilibrium, as it contradicts contextuality, and built non-Nashian equilibria and algorithms to use instead. From an abstract perspective, in this model, our actual world can be seen as the unique solution to a fixpoint equation.
The next phase is to instantiate these models into theories by playing with the parameters.
Query languages for heterogeneous and denormalized data
The latest technological developments in the area of Big Data have come at the cost of more complexity, making efficient querying costly and, in spite of the field of data science's being by essence interdisciplinary, not easily accessible to non-IT-savvy-users. While the database community invests a lot of efforts into the performance aspects of such systems, many end users care about productivity as well.
I believe that we need to come back to the fundamental ideas laid out by Edgar Codd in 1970, namely data independence, in his founding paper on relational databases: use natural, simple querying languages that hide away all the complexity and storage details of a system. Such languages must be functional, declarative and tailored to the data shapes at hand: tables (SQL), trees, cubes or graphs.
With this line of reasoning, I co-designed the functional JSONiq language during my doctoral studies to query arborescent documents, which correspond to heterogeneous, denormalized datasets commonly found on the Web as JSON or XML. This language is stable and is now supported by several engines including IBM WebSphere. In order to build on the shoulders of giants and learn from past experience, 95% of the language is directly inherited from the XQuery standard, itself designed at the World Wide Web Consortium by some of the same people who designed the SQL language, and involving major industry players.
I am now working on the implementation of JSONiq on top of large-scale clusters such as Spark, in a project called RumbleDB. The JSONiq expressions are seamlessly mapped to Spark constructs, hiding the complexity of Spark to the end user and making it accessible to non-Computer Scientists.
Further investigation includes making JSONiq accessible to the Machine Learning community with the same goal in mind.
|Fall 2022||Master's thesis||Remo Röthlisberger||Making JSONiq on Snowflake more complete and faster|
|Fall 2022||Master's thesis||Dominik Bruggisser||Can we find Higgs Bosons faster? Optimizing High-Energy-Physics Queries with JSONiq|
|Spring 2022||Semester project||Olivier Goerens||Introducing Snowflake support in RumbleDB|
|Spring 2022||Master's thesis||Thomas Zhou||Compiling JSONiq to SQL|
|Fall 2021||Master's thesis||Zirun Wang||JSONiq and RumbleDB on Snowflake|
|Spring 2021||Semester project||Elwin Stephan||Exploring an alternative approach for grouping and ordering numbers in Rumble|
|Fall 2020||Master's thesis||Stevan Mihajlovic||A Test Suite for Rumble|
|Fall 2020||Master's thesis||Mario Arduini||Type-based optimizations in Rumble|
|Fall 2020||Bachelor's thesis||Martin Fiebig||Building a JSONiq Query Optimizer using MLIR|
|Spring 2020||Bachelor's thesis||Eric Enzler||Reading ROOT Files as Apache Arrow Tables|
|Spring 2020||Bachelor's thesis||Manuel Reber||Optimizing JSONiq execution in Rumble using MLIR|
|Spring 2020||Bachelor's thesis||Gianluca Moro||Storage Format for Almost-Homogeneous Data Sets|
|Spring 2020||Semester project||Emilien Pilloud||Exploring the limitations of the Bayesian approach|
|Spring 2020||Semester project||Ioana Stefan||More features for Rumble|
|Spring 2020||Semester project||Falko Noé||More features for Rumble|
|Fall 2019||Bachelor's thesis||Ramon Gomm||Building a Perfect Prediction Engine|
|Fall 2019||Bachelor's thesis||Diana Steffen||Building a TYSON processor|
|Fall 2019||Master's thesis||Andrea Rinaldi||Querying heterogeneous, typed data at large scales|
|Fall 2019||Master's thesis||Can Berker Cikis||Machine learning with JSONiq|
|Spring 2019||Master's thesis||Christopher Gerber||Composey: Web Application that Composes Messages from Movies|
|Fall 2018||Semester project||Christopher Gerber||Composing messages from movie subtitles|
|Fall 2018||Master's thesis||Felipe Sulser||A Data-Driven Exploration of Non-Nashian Game Theory|
|Fall 2017||Master's thesis||Daniel Yu||Efficient Processing of Almost-Homogeneous Semi-Structured Data|
|Spring 2017||Master's thesis||Stefan Irimescu||JSONiq on Spark|
Click on the titles of the papers to access them.
The Big Data Textbook - teaching large-scale databases in universities.
May 2021 (ongoing). Freely available.
The XBRL Book..
January 2022. Fifth edition. Self-published on Amazon KDP.
|M. Baczyk, G. Fourny.
Nashian game theory is incompatible with quantum contextuality..
December 2021. ArXiv preprint, to be submitted for publication.
|G. Fourny, D. Dao, C. B. Cikis, C. Zhang, G. Alonso.
RumbleML: program the lakehouse with JSONiq.
December 2021. ArXiv preprint.
|D. Graur, I. Müller, M. Proffitt, G. Fourny, G. T. Watts, G. Alonso.|
Evaluating query languages and systems for high-energy physics data.
October 2021. Proceedings of the VLDB Endowment, Volume 15, Issue 2, pp 154–168./td>
The Future of Quantum Theory: A Way Out of the Impasse.
September 2021. ArXiv preprint.
|G. Fourny, F. Sulser.
Data on the existence ratio and social utility of Nash Equilibria and of the Perfectly Transparent Equilibrium.
February 2021. Data in Brief, Volume 34, 106623.
|I. Müller, G. Fourny, S. Irimescu, C. B. Cikis, G. Alonso.
Rumble: Data Independence for Large Messy Data Sets.
February 2021. Proceedings of the VLDB Endowment, Volume 14, Issue 4, December 2020, pp 498–506.
|G. Fourny, S. Reiche, J.-P. Dupuy.
Perfect Prediction Equilibrium.
June 2020. The individual and the other in economic thought: an introduction. Routledge, pp. 209-257.
Perfect Prediction in Normal Form: Superrational Thinking Extended to Non-Symmetric Games.
June 2020. Journal of Mathematical Psychology, Volume 96, 102332.
Games in Minkowski Spacetime.
April 2020. Preprint on arXiv.org, currently under review.
Contingent Free Choice: On Extending Quantum Theory to a Contextual, Deterministic Theory With Improved Predictive Power.
July 2019. Vision paper. ArXiv preprint.
Perfect Prediction in Minkowski Spacetime: Perfectly Transparent Equilibrium for Dynamic Games with Imperfect Information.
May 2019. ArXiv preprint.
Kripke semantics of the Perfectly Transparent Equilibrium.
July 2018. ArXiv preprint.
Common Counterfactual Belief of Rationality Subsumes Superrationality On Symmetric Games.
Preprint now merged into a published paper. arXiv.org.
On the Importance of Correlations in Rational Choice: A Case for Non-Nashian Game Theory.
decimalInfinite: All Decimals In Bits. No Loss. Same Order. Simple.
arXiv preprint arXiv:1506.01598.
arXiv preprint arXiv:1410.0600.
|D. Florescu, G. Fourny.
JSONiq: The history of a query language.
IEEE Internet Computing, 86-90.
JSONiq: the SQL of NoSQL.
Amazon. CreateSpace Independent Publishing Platform.
|J. Robie, M. Brantner, D. Florescu, G. Fourny, T. Westmann.
JSONiq: XQuery for JSON, JSON for XQuery.
XML Prague conference.
|T. Etter. D. Florescu. P. Fisher. G. Fourny. D. Kossmann.
XQuery in the Browser reloaded.
XML Prague conference.
Flexible models for programming the Web.
PhD Thesis. ETH Zurich.
|G. Fourny, D. Florescu, D. Kossmann, M. Zaharioudakis, D. Kossmann.
A time machine for XML.
Technical report. ETH Zurich.
|G. Fourny, D. Florescu, D. Kossmann, M. Zacharioudakis.
A time machine for XML: PUL composition.
XML Prague conference.
|G. Fourny, M. Pilman, D. Florescu, D. Kossmann, T. Kraska, D. McBeath.
XQuery in the Browser.
Proceedings of the 18th international conference on World wide web, 1011-1020.
QuantumML: Modelling Hidden Worlds of Information.
2007. Master’s Thesis report.
Simplified Fault-Tolerant Algorithms for Embedded and Real-Time Application in Unmanned Aerial Vehicles.
2006. Semester’s Thesis report.
Jive/JML Generating Proof Obligations From JML Specifications.
2005. Semester’s Thesis report.
Games in Extensive Form with Essential Prediction, without Indifference and without Chance Moves: Projected Equilibrium.
Stanford, 2004 (Visiting fellow). Report.
|Getting started with RumbleDB. Tutorial. Declarative Amsterdam, November 7, 2022.|
|RumbleDB: Data independence for large, messy datasets. Declarative Amsterdam, November 5, 2021.|
|RumbleDB: Data independence for large, messy datasets. VLDB 2021, August 2021.|
|RumbleDB: Data independence for large, messy datasets. National University of Singapore, Computer Science Seminar, June 29, 2021.|
|And if Einstein was right? The paradigm shift we need. D-INFK Faculty Seminar, ETH Zurich. March 23, 2021.|
|RumbleDB: Data independence for large, messy datasets. Vanderbilt University, March 2021.|
|Non-Nashian Game Theory: the invisible hand that leads to Pareto optima. DeepMind Reading Group. July 2020.|
|Quantum theory and determinism: why there is really no contradiction. Systems Group Lunch Seminar, ETH Zurich. May 2020.|
|Non-Nashian Game Theory: learning being an open book. Chair of Learning Sciences and Higher Education, ETH Zurich, November 2019.|
|Big Data. Edmond de Rothchild. October 3, 2019.|
|JSONiq on Spark. Systems Group Lunch Seminar, ETH Zurich. April 2019.|
|Non-Nashian Game Theory: towards friendly and trustworthy robots? Coffee talk, Institut für Automatik (D-ITET), January 17, 2019.|
|Non-Nashian Game Theory and what it matters to Data Science. Lunch seminar, Systems Group (D-INFK), November 17, 2017.|
|Perfect Prediction Equilibrium, Annual Summer Interdisciplinary Conference, Interlaken, July 17, 2017.|
|Big Data, CFA Society Continuing Education Seminar, Zurich, May 31, 2017.|
|Perfect Prediction Equilibrium, Guest talk, General assembly IAETH, Zurich, March 14, 2016.|
|Perfect Prediction Equilibrium, Institute of Neuroinformatics, UZH/ETH, November 19, 2015.|
|JSONiq: the SQL of NoSQL, Jazoon conference, Technopark, Zurich, October 23, 2015.|
|Cell stores, Lunch seminar, Systems group, November 2014.|
|Perfect Prediction Equilibrium, 2nd international conference on Economic Philosophy, October 10, 2014.|
|The JSONiq language, Lunch seminar, Systems group, October 2012.|
|Programming the cloud, ETH/industry event, Biel, November 2011.|
|Introduction to relativity theory, Lunch seminar, Systems group, October 2011.|
|A time machine for XML, XQuery Meetup, Dozentenfoyer, ETH, July 2011.|
|Perfect prediction: an intriguing paradox, a prediction model, and its application to game theory, Lunch seminar, Systems group, June 2011.|
|XQuery in the browser, Mozilla, California, March 2011.|
|XQuery Scripting proposal, Face-to-face meeting of the W3C XML Query Working Group, Marklogic, California, February 2011.|
|A 20-minute XML standard primer, Lunch seminar, Systems Group, October 2010.|
|Perfect prediction: an intriguing paradox, a prediction model, and its application to game theory, Lunch seminar, Institute for Theoretical Physics, ETH, August 28, 2009.|