The 17th International Symposium on Algorithmic Game Theory (SAGT)

The program of SAGT 2024 will include a tutorial day, invited lectures and presentations of peer-reviewed submissions. 

Foundational work is solicited on topics including but not limited to:

• Solution Concepts in Game Theory
• Efficiency of Equilibria and Price of Anarchy
• Computational Aspects of Equilibria
• Learning and Dynamics in Games
• Game-Theoretic Aspects of Networks
• Auction Design and Analysis
• Algorithmic Contract Design
• Mechanism Design and Pricing
• Internet Economics and Computational Advertising
• Reputation, Recommendation and Trust Systems
• Economic Aspects of Distributed Computing
• Blockchain and Cryptocurrencies
• Decision Theory and Information Design
• Computational Social Choice and Fair Division
• Market Design and Matching Markets
• Cooperative Game Theory

Industrial application works and position papers presenting novel ideas, issues, challenges and directions are also welcome.

The symposium proceedings will be published by Springer as a Lecture Notes in Computer Science (LNCS) proceedings volume in the ARCoSS subline. Accepted papers will be allocated at most 18 pages in the proceedings. To accommodate the publishing traditions of different fields, authors of accepted papers can choose to publish a one-page abstract in the proceedings. Please see the submission instructions below for more details.

There will be a SAGT 2024 Best Paper Award, accompanied by a prize of 1000 Euro offered by Springer.

SAGT 2024 will extend invitations to a selection of accepted papers for publication in a dedicated special issue of the ACM Transactions on Economics and Computation (TEAC) (details will be communicated in due course). 

Please feel free to contact us if you have any questions about the symposium: sagt2024@easychair.org.

NOTE: The 7th International Workshop on Matching Under Preferences (MATCHUP) takes place right after SAGT at the University of Oxford, United Kingdom, September 9–11, 2024.

Submission deadline: extended (was May 21, 2024, 23:59 AoE)

Abstract submission deadline: May 22, 2024, 23:59 AoE

Full paper submission deadline: May 26, 2024, 23:59 AoE

(see below for submission instructions)

Notification: July 5, 2024

Camera ready deadline: July 15, 2024 (note: 10 days after acceptance!)

Conference dates: September 3–6, 2024

Authors are invited to submit original research for possible presentation at the conference. Each paper will be evaluated on significance, originality, technical quality, and exposition. It should clearly establish the research contribution, its relevance, and its relation to prior research.

Submission Server and Deadlines

Paper submissions will be handled through EasyChair. Please use the following link to submit your paper:

https://easychair.org/conferences/?conf=sagt2024

The submission deadline has been extended:

Abstract submission deadline: May 22, 2024, 23:59 AoE

Full paper submission deadline: May 26, 2024, 23:59 AoE

NOTE: An empty paper is sufficient to register the paper before the abstract submission deadline. However, a full abstract is required and only minor changes will be allowed to that before the full paper deadline. Only papers that have been registered before the abstract submission deadline will be considered.

The expectation is that at least one of the authors of each accepted paper will attend SAGT and give a presentation for the work. However, we are open to consider the possibility of remote talks in exceptional circumstances (due to, e.g., visa/travel issues) provided that at least one author has registered to the conference.

Submission Format

Submissions may be up to 18 pages long (excluding title page and references) in single-column format, using at least 11-point fonts, single-spacing between lines, and at least 1-inch margins all around. In addition, an appendix may be included at the end of the paper and will be read at the discretion of the reviewers. Submissions deviating significantly from these guidelines may be rejected without review. Please note that, as for all previous editions, SAGT 2024 implements a single-blind peer-review process.

Authors are strongly encouraged to structure their paper in a way that includes a clear presentation of the merits of the paper and a discussion of the importance of the results, as well as an exposition of the key conceptual and technical ideas.

The symposium proceedings will be published by Springer as a Lecture Notes in Computer Science (LNCS) proceedings volume in the ARCoSS subline. Accepted papers will be allocated at most 18 pages (including title page and references) in LNCS format in the proceedings. Please note that the LNCS format has much wider page margins. We recommend that authors use the LNCS format (provided as part of Springer's LaTeX2e package) to prepare their submission, but this is not a requirement for the submission. Please refer to Springer's Information for Authors for detailed guidelines on how to prepare the final manuscript.

To accommodate the publishing traditions of different fields, authors of accepted papers can choose to publish a one-page abstract of their paper in the proceedings. The paper must then provide a URL referring to the full version of the paper; authors should guarantee the link to be reliable for at least two years. Such papers must be formatted and submitted just like regular papers (as described above). 

Results previously published or presented at another archival conference prior to SAGT, or published (or accepted for publication) at a journal prior to the submission deadline, will not be considered for publication as regular papers. Simultaneous submission of regular papers to another conference with published proceedings is not allowed. Simultaneous submission of results to a journal is allowed only if the authors intend to publish the paper as a one-page abstract in SAGT 2024.

Best Paper Award

There will be a SAGT 2024 Best Paper Award, accompanied by a prize of 1000 Euro offered by Springer.

Paul Duetting, Google

Ambiguous Contracts

Contract theory captures situations where two parties—a principal and an agent— can benefit from mutual cooperation. The prototypical situation is one in which the principal seeks to delegate the execution of a job to an agent. The agent can take different costly actions, and his choice of action entails a stochastic outcome (with attached reward) for the principal. The principal cannot directly observe the agent’s choice of action but can influence the agent’s decision through a contract that specifies outcome-contingent payments. Given a contract, the agent aims to maximize his expected payment minus cost. The goal of the principal is to maximize her expected utility given by expected reward minus payment, under the action chosen by the agent.
One feature of real-life contracts that is not captured (or explained) by this classic model is that practical contracts are often ambiguous. For example, the promotion guidelines of a university may require a candidate to demonstrate “research productivity and excellence.” Similarly, a professional services contract may require that a provider exert “due diligence.” This ambiguity may be due to an inability of the two parties to provide more precise specification. In contrast, we explore the deliberate infusion of ambiguity as a tool to enhance the principal’s contracting power over the agent.
Towards this goal we propose an extension of the classic (hidden action) principal-agent model. In this model, an ambiguous contract consists of a set of classic contracts, which the agent evaluates by considering the minimum utility a given action yields against any contract in the support of the ambiguous contract. At the same time, we require that the principal’s utility—under the action chosen by the agent—is the same for all contracts in the support of the ambiguous contract. We show that this expands the set of actions that the principal can implement, and that the principal’s gain from using an ambiguous contract can be arbitrarily large. We further characterize the structure of optimal ambiguous contracts, showing that ambiguity drives optimal contracts towards simplicity. We also provide a characterization of ambiguity-proof classes of contracts, where the principal cannot gain by infusing ambiguity. Finally, we show that when the agent can engage in mixed actions, the advantages of ambiguous contracts disappear.

Vasilis Gkatzelis, Drexel University

Learning-Augmented Mechanism Design

For more than half a century, the dominant approach for the mathematical analysis of algorithms in computer science has been worst-case analysis. While worst-case analysis provides a useful signal regarding the robustness of an algorithm, it can be overly pessimistic, and it often leads to uninformative bounds or impossibility results that may not reflect real-world obstacles. Meanwhile, advances in machine learning have led to very practical algorithms, most of which do not provide any non-trivial worst-case performance guarantees. Motivated by the tension between worst-case analysis and machine learning, a surge of recent work focuses on the design algorithms that are guided by machine-learned predictions, aiming to perform better in practice, while maintaining their robustness. Specifically, the goal of this literature on “learning-augmented algorithms” is to design algorithms that simultaneously provide two types of guarantees: “robustness” (which corresponds to the classic worst-case guarantees, even if the predictions that the algorithm is provided with are arbitrarily bad) and “consistency” (i.e., stronger performance guarantees when the predictions are accurate). This “learning-augmented framework” has been used successfully in a variety of settings, e.g., toward a refined analysis of competitive ratios in online algorithms and running times in traditional algorithms.

A recent line of work on “learning-augmented mechanism design” has deployed this learning augmented framework in settings involving strategic agents. In such settings, the designer often faces additional obstacles which further limit their ability to reach desired outcomes. For example, some of the input that the designer needs may be private information held by the participating agents, and the agents could strategically misreport this information, aiming to maximize their own utility. In other settings, the agents may even have direct control over some aspects of the outcome. The long literature on mechanism design has proposed a variety of solutions for these types of problems, aiming to align the incentives of the agents with those of the designer, but the worst-case guarantees of these solutions are often underwhelming from a practical perspective. This talk will introduce the “learning-augmented mechanism design” model and provide an overview of some of the results in this line of work.

Katrina Ligett, Hebrew University of Jerusalem

Actually, Data is a Rival Good

There is a tendency in many fields, including computer science, economics, and industry, to model data as a non-rival good, meaning that one entity using a particular piece of data doesn’t impinge on its use by others. Food is a classic rival good (if I eat the apple, you cannot); digital music is a classic non-rival good (my listening to the song has no effect on your listening experience). Data might, at first blush, seem more like digital music than like an apple. In this talk, I will give arguments from three fields—economics, privacy, and statistics—for why modeling data as non-rival is problematic, and will argue that we need a new paradigm.
The core of the economic argument is that generative AI has transformed the market for data, making competition (and rivalrousness) for and around data newly central. The privacy and statistical validity arguments rely on mathematical frameworks that help us understand how repeated uses of a dataset accumulate and interact. All of these arguments suggest new models and metaphors for data, and directions for further work.

The tutorials on September 3 will be given by:

Jan Maly, Vienna University of Economics and Business & TU Wien

Proportionality in Multi-Winner Voting: Axioms, Voting Rules and Equilibria

Voting is one of the oldest and most studied forms of collective decision making. In this tutorial, I will focus on a specific form of voting, namely multi-winner voting, where a set of voters must elect a committee of k candidates out of a larger set of available alternatives. One of the central requirements in multi-winner voting is that the chosen committee represents the electorate’s preferences in a proportional manner. While proportionality is well understood in traditional parliamentary elections, it is not immediately clear what it means without a rigid party structure. However, recent advances in computational social choice theory have allowed us to develop a new and deeper understanding of fairness and proportionality in multi-winner voting, even without explicitly defined parties.

In the first part of the tutorial I will give an introduction to this recent literature on proportionality in multi-winner voting, presenting some axioms that have developed to formally capture the intuitive idea of proportionality, examine some voting rules designed to provide proportional committees, and look at some computational aspects of finding fair committees. In the second part of the tutorial, I will link multi-winner voting with game theory by showing that many of the axioms and rules touched on in the first part can be understood in terms of equilibria in normal-form games or in markets for public goods. To conclude, I will discuss important open questions in multi-winner voting and how methods from (algorithmic) game theory might be used to answer them.

Rebecca Reiffenhäuser, University of Amsterdam

Prophet Inequalities with Limited Information

Many applications require assigning resources in an online fashion: decisions on incoming bids have to be made immediately, and before seeing all n bids in the sequence.
A fundamental hardness inherent to this setting limits online strategies to perform no better than the one that simply picks a random winner for each good, so additional assumptions are needed. Famously, in prophet inequalities, one assumes access to all n (independent) distributions that the participant’s bids will be drawn from, achieving exp. competitive guarantees of up to half the expected offline optimum.

Recently, considerable interest has been drawn to the fact that full prior distributional knowledge is in general not actually necessary to circumvent the impossibility.
Instead, one can often achieve similar (or even equally good) approximations with access to just a few samples from each distribution. This data-driven approach poses a clear advantage in practical settings, where full distributions are rarely available (but one might, for example, know the bids the same participant placed in a few previous auctions).

The tutorial gives an overview of such results, with a focus on the edge case of ‘minimal’ prior knowledge, i.e. the online algorithm only has access to one single sample from each distribution. We first derive how and why constant ratios can be obtained for simple settings, and then show how the same principle surprisingly generalizes up to very recent results on XOS combinatorial assignments. Then, we consider the design of truthful mechanisms, a central goal whenever agents might bid strategically, and discuss the existing results as well as known obstacles to extending them.

Program Committee

• Georgios Amanatidis, University of Essex
• Vittorio Bilò, University of Salento
• Ioannis Caragiannis, Aarhus University
• Riccardo Colini Baldeschi, Meta
• Andrés Cristi, EPFL
• Argyrios Deligkas, Royal Holloway University of London
• Tomer Ezra, Harvard University
• Diodato Ferraioli, Università di Salerno
• Federico Fusco, Sapienza University of Rome
• Martin Gairing, University of Liverpool
• Thekla Hamm, Utrecht University
• Kristoffer Arnsfelt Hansen, Aarhus University
• Tobias Harks, Passau University
• Martin Hoefer, RWTH Aachen University
• Alexandros Hollender, University of Oxford
• Simon Jantschgi, University of Oxford
• Panagiotis Kanellopoulos, University of Essex
• Bart de Keijzer, King's College London
• Pieter Kleer, Tilburg University
• Max Klimm, Technische Universität Berlin
• Piotr Krysta, Augusta University & University of Liverpool
• Pascal Lenzner, Hasso-Plattner-Institute Potsdam
• Bo Li, The Hong Kong Polytechnic University
• Minming Li, City University of Hong Kong
• David Manlove, University of Glasgow
• Vangelis Markakis, Athens University of Economics and Business
• Gianpiero Monaco, University of Chieti-Pescara
• Paolo Penna, IOG
• Guido Schäfer, CWI & University of Amsterdam (co-chair)
• Daniel Schoepflin, Rutgers University
• Alkmini Sgouritsa, Athens University of Economics and Business & Archimedes/Athena RC
• Tami Tamir, Reichman University
• Yixin Tao, Shanghai University of Finance and Economics
• Zoi Terzopoulou, Saint-Etienne School of Economics
• Carmine Ventre, King's College London (co-chair)
• Adrian Vetta, McGill University
• Anaëlle Wilczynski, Université Paris-Saclay
  

Local Organization

• Ulle Endriss, University of Amsterdam
• Sophie Klumper, CWI & University of Amsterdam
• Guido Schäfer, CWI & University of Amsterdam
• Artem Tsikiridis, CWI
• Susanne van Dam, CWI

The list of papers accepted at SAGT 2024 is given below (in no specific order).

• Haris Aziz, Peter Biro, Gergely Csáji and Ali Pourmiri.
  Ex-post Stability under Two-Sided Matching: Complexity and Characterization
• Edith Elkind, Ayumi Igarashi and Nicholas Teh.
  Fair Division of Chores with Budget Constraints
• Zihan Li, Pasin Manurangsi, Jonathan Scarlett and Warut Suksompong.
  Complexity of Round-Robin Allocation with Potentially Noisy Queries
• Yotam Gafni, Ronen Gradwohl and Moshe Tennenholtz.
  Prediction-Sharing During Training and Inference
• Haris Aziz, Jiarui Gan, Grzegorz Lisowski and Ali Pourmiri.
  The Team Order Problem: Maximizing the Probability of Matching Being Large Enough
• Haris Aziz, Venkateswara Kagita, Baharak Rastegari and Mashbat Suzuki.
  Approval-Based Committee Voting under Uncertainty
• Jon Kleinberg, Emily Ryu and Eva Tardos.
  Calibrated Recommendations for Users with Decaying Attention
• Farid Arthaud, Edan Orzech and Martin Rinard.
  Edge-dominance games on graphs
• Georgios Birmpas, Tomer Ezra, Stefano Leonardi and Matteo Russo.
  Fair Division with Interdependent Values
• Frederik Glitzner and David Manlove.
  Structural and algorithmic results for stable cycles and partitions in the Roommates problem
• Ioannis Caragiannis and Sebastian Homrighausen.
  Estimating the Expected Social Welfare and Cost of Random Serial Dictatorship
• Rachit Agarwal, Giannis Fikioris and Eva Tardos.
  Incentives in Dominant Resource Fair Allocation under Dynamic Demands
• Farid Arthaud.
  Playing repeated games with sublinear randomness
• Argyrios Deligkas, Mohammad Lotfi and Alexandros Voudouris.
  Agent-Constrained Truthful Facility Location Games
• Ahuva Mualem and Juan Carlos Carbajal.
  Mind the Revenue Gap: On the Performance of Approximation Mechanisms under Budget Constraints
• Bainian Hao and Carla Michini.
  Price of Anarchy in Paving Matroid Congestion Games
• Agnes Totschnig, Rohit Vasishta and Adrian Vetta.
  Matrix Rationalization via Partial Orders
• Andrzej Turko and Jarosław Byrka.
  Sublogarithmic Approximation for Tollbooth Pricing on a Cactus
• Rashida Hakim, Jason Milionis, Christos Papadimitriou and Georgios Piliouras.
  Swim till you sink: Computing the limit of a game
• Vijay Vazirani.
  The Investment Management Game: Extending the Scope of the Notion of Core
• Shuchi Chawla, Kira Goldner, Anna Karlin and Benjamin Miller.
  Non-Adaptive Matroid Prophet Inequalities
• Meryem Essaidi, Kira Goldner and S. Matthew Weinberg.
  To Regulate or Not to Regulate: Using Revenue Maximization Tools to Maximize Consumer Utility
• Wouter Fokkema, Ruben Hoeksma and Marc Uetz.
  Price of Anarchy for Graphic Matroid Congestion Games
• Dinesh Kumar Baghel, Alex Ravsky and Erel Segal-Halevi.
  k-times bin-packing and its application to fair electricity distribution
• Stephane Airiau, Nicholas Kees Dupuis and Davide Grossi.
  Condorcet Markets
• Aggelos Kiayias, Elias Koutsoupias, Francisco Marmolejo-Cossío and Akaterini-Panagiota Stouka.
  Balancing Participation and Decentralization in Proof-of-Stake Cryptocurrencies
• Gennaro Auricchio and Jie Zhang.
  The k-Facility Location Problem Via Optimal Transport: A Bayesian Study of the Percentile Mechanisms
• Samuel Bismuth, Ivan Bliznets and Erel Segal-Halevi.
  Fair Division with Bounded Sharing
• Yiannis Giannakopoulos and Johannes Hahn.
  Discrete Single-Parameter Optimal Auction Design
• Edwin Lock, Zephyr Qiu and Alexander Teytelboym.
  The Computational Complexity of the Housing Market
• Milena Mihail and Thorben Tröbst.
  Online Matching with High Probability
• Ian DeHaan and Kanstantsin Pashkovich.
  Matroid Bayesian Online Selection

The SAGT 2024 booklet containing the complete scientific program together with more detailed information about the planned events is available as PDF: here.

Tuesday, September 3, 2024: Tutorial Day

09:00 - 10:00 Registration - Coffee

10:00 - 11:00 Tutorial 1A: Proportionality in Multi-Winner Voting: Axioms, Voting Rules and Equilibria - Jan Maly. (Chair: Guido Schäfer)

11:00 - 11:30 Coffee

11:30 - 12:30 Tutorial 1B: Proportionality in Multi-Winner Voting: Axioms, Voting Rules and Equilibria - Jan Maly. (Chair: Guido Schäfer)

12:30 - 13:30 Lunch

13:30 - 14:30 Tutorial 2A: Prophet Inequalities with Limited Information - Rebecca Reiffenhäuser. (Chair: Carmine Ventre)

14:30-15:00 Coffee

15:00-16:00 Tutorial 2B: Prophet Inequalities with Limited Information - Rebecca Reiffenhäuser. (Chair: Carmine Ventre)

16:00 - 16:30 Coffee

16:30 - 18:00 LightNLing Talks (Chair: Guido Schäfer)

1. An Improved Bound for the Price of Anarchy for Related Machine Scheduling
   Arman Rouhani
2. Fair Division with Minimal Withheld Information in Social Networks
   Ivan Bliznets
3. Distributionally Robust Monopoly Pricing: Switching from Low to High Prices in Volatile Markets
   Tim van Eck
4. The Secretary Problem with Independent Sampling
   Tim Oosterwijk

18:00 - 19:00 Reception

Wednesday, September 4, 2024: Conference Day 1

09:00 - 10:00 Registration - Coffee

10:00 - 10:10 Opening

10:10 - 11:10 Invited Talk: Actually, Data is a Rival Good - Katrina Ligett (Chair: Carmine Ventre)

11:10 - 11:40 Coffee break

11:40 - 13:00 Session 1: Matching. (Chair: Rebecca Reiffenhäuser)

1. Ex-post Stability under Two-Sided Matching: Complexity and Characterization
   Haris Aziz, Peter Biro, Gergely Csáji and Ali Pourmiri
2. Structural and algorithmic results for stable cycles and partitions in the Roommates problem
   Frederik Glitzner and David Manlove
3. The Team Order Problem: Maximizing the Probability of Matching Being Large Enough
   Haris Aziz, Jiarui Gan, Grzegorz Lisowski and Ali Pourmiri
4. *Online Matching with High Probability
   Milena Mihail and Thorben Tröbst

13:00 - 14:00 Lunch

14:00 - 15:20 Session 2: Fair Division and Resource Allocation. (Chair: Ulle Endriss)

1. Fair Division of Chores with Budget Constraints
   Edith Elkind, Ayumi Igarashi and Nicholas Teh
2. Fair Division with Interdependent Values
   Georgios Birmpas, Tomer Ezra, Stefano Leonardi and Matteo Russo
3. Fair Division with Bounded Sharing
   Samuel Bismuth, Ivan Bliznets and Erel Segal-Halevi
4. Incentives in Dominant Resource Fair Allocation under Dynamic Demands
   Rachit Agarwal, Giannis Fikioris and Eva Tardos

15:20 - 15:40 Coffee break

15:40 - 17:00 Session 3: Mechanism Design. (Chair: Kira Goldner)

1. Agent-Constrained Truthful Facility Location Games
   Argyrios Deligkas, Mohammad Lotfi and Alexandros Voudouris
2. The k-Facility Location Problem Via Optimal Transport: A Bayesian Study of the Percentile Mechanisms
   Gennaro Auricchio and Jie Zhang
3. Discrete Single-Parameter Optimal Auction Design
   Yiannis Giannakopoulos and Johannes Hahn
4. Estimating the Expected Social Welfare and Cost of Random Serial Dictatorship
   Ioannis Caragiannis and Sebastian Homrighausen

17:00 - 17:30 Business Meeting

17:30 - 19:00 Conference Reception

Thursday, September 5, 2024: Conference Day 2

09:30 - 10:00 Registration - Coffee

10:00 - 11:00 Invited Talk: Learning-Augmented Mechanism Design - Vasilis Gkatzelis (Chair: Guido Schäfer)

11:00 - 11:30 Coffee break

11:30 - 12:50 Session 4: Game Theory and Repeated Games (Chairs: Guido Schäfer & Carmine Ventre)

1. Best Paper: Swim Till You Sink: Computing the Limit of a Game
   Rashida Hakim, Jason Milionis, Christos Papadimitriou and Georgios Piliouras
2. Best Student Paper: Playing Repeated Games with Sublinear Randomness
   Farid Arthaud
3. Edge-Dominance Games on Graphs
   Farid Arthaud, Edan Orzech and Martin Rinard
4. *The Investment Management Game: Extending the Scope of the Notion of Core
   Vijay Vazirani

12:50 - 13:50 Lunch

13:50 - 15:10 Session 5: Pricing, Revenue, and Regulation (Chair: Marc Uetz)

1. Mind the Revenue Gap: On the Performance of Approximation Mechanisms under Budget Constraints
   Ahuva Mualem and Juan Carlos Carbajal
2. Sublogarithmic Approximation for Tollbooth Pricing on a Cactus
   Andrzej Turko and Jarosław Byrka
3. To Regulate or Not to Regulate: Using Revenue Maximization Tools to Maximize Consumer Utility
   Meryem Essaidi, Kira Goldner and S. Matthew Weinberg
4. Balancing Participation and Decentralization in Proof-of-Stake Cryptocurrencies
   Aggelos Kiayias, Elias Koutsoupias, Francisco Marmolejo-Cossío and Akaterini-Panagiota Stouka

15:10 - 15:30 Coffee break

15:30 - 16:50 Session 6: Matroid Theory in Game Theory (Chair: Pieter Kleer)

1. Price of Anarchy in Paving Matroid Congestion Games
   Bainian Hao and Carla Michini
2. Price of Anarchy for Graphic Matroid Congestion Games
   Wouter Fokkema, Ruben Hoeksma and Marc Uetz
3. Non-Adaptive Matroid Prophet Inequalities
   Shuchi Chawla, Kira Goldner, Anna Karlin and Benjamin Miller
4. Matroid Bayesian Online Selection
   Ian DeHaan and Kanstantsin Pashkovich

17:45 Boat Departure Amsterdam Centraal

19:30 - 22:30 Conference Dinner at Restaurant Baut Oost

Friday, September 6, 2024: Conference Day 3

09:30 - 10:00 Registration - Coffee

10:00 - 11:00 Invited Talk: Ambiguous Contracts - Paul Duetting (Chair: Carmine Ventre)

11:00 - 11:30 Coffee break

11:30 - 12:50 Session 7: Information Sharing and Decision Making (Chair: Nicole Immorlica)

1. Calibrated Recommendations for Users with Decaying Attention
   Jon Kleinberg, Emily Ryu and Eva Tardos
2. Matrix Rationalization via Partial Orders
   Agnes Totschnig, Rohit Vasishta and Adrian Vetta
3. Approval-Based Committee Voting under Uncertainty
   Haris Aziz, Venkateswara Kagita, Baharak Rastegari and Mashbat Suzuki
4. *Prediction-Sharing During Training and Inference
   Yotam Gafni, Ronen Gradwohl and Moshe Tennenholtz

12:50 - 13:50 Lunch

13:50 - 15:10 Session 8: Computational Complexity and Resource Allocation (Chair: Vangelis Markakis)

1. k-Times Bin-Packing and its Application to Fair Electricity Distribution
   Dinesh Kumar Baghel, Alex Ravsky and Erel Segal-Halevi
2. The Computational Complexity of the Housing Market
   Edwin Lock, Zephyr Qiu and Alexander Teytelboym
3. Condorcet Markets
   Stephane Airiau, Nicholas Kees Dupuis and Davide Grossi
4. Complexity of Round-Robin Allocation with Potentially Noisy Queries
   Zihan Li, Pasin Manurangsi, Jonathan Scarlett and Warut Suksompong

15:10 Appeltaart & Slagroom

The SAGT 2024 symposium proceedings will be published by Springer as a Lecture Notes in Computer Science (LNCS) proceedings volume in the ARCoSS subline.

Conference participants have free access to the online conference proceedings of SAGT via the following link (note: free access will be granted until 15 October 2024).

SAGT 2024 will extend invitations to a selection of accepted papers for publication in a dedicated special issue of the ACM Transactions on Economics and Computation (TEAC) (details will be communicated in due course). 

The SAGT proceedings of previous years are available on SpingerLink.

Registration fees are as follows: 

• student: 220 EUR (early) / 285 EUR (late)
• regular: 285 EUR (early) / 350 EUR (late)

Early registration deadline: August 16, 2024 AoE

Register here.

The registration fee includes participation in the tutorial day and conference, as well as the reception on Tuesday and the conference dinner on Thursday.

You can also purchase extra tickets (50 Euro) for the conference dinner on Thursday. The conference dinner will be at Restaurant Baut Oost.

For visa-related enquiries, see here.

We are pleased to announce that, due to generous financial support from NWO, we can offer waived registration fees for a limited number of junior researchers working in the Netherlands to attend SAGT 2024 and present a short talk on their research.

Who can apply?
Junior researchers (PhD students, postdocs, assistant professors) employed at Dutch universities/research institutes.

What is covered?
Waived registration fee for full attendance at SAGT 2024.

What is required?
Present a short talk (about 10-15 minutes, TBD) on your research within the scope of SAGT during the Tutorial Day (September 3, afternoon).

How to apply?
Please send your full name, position, affiliation, and the title of your talk (including a PDF or link to the respective paper, if available) via email to sagt2024@easychair.org with subject “SAGT 2024 waived registration”.

Note: Only a limited number of slots are available. Act quickly to secure your place!

UPDATE 31-07: Please do not submit any new applications. The generous travel support budget provided by G-Research has now been fully allocated.

We are pleased to announce a new travel grant opportunity generously funded by G-Research. These grants are designed to support students who wish to attend SAGT 2024 but do not have enough funds for that. If you are interested in this opportunity, please send a CV, proof that you are a student, and a detailed budget of the required financial support via email to sagt2024@easychair.org with subject “SAGT travel grant”. G-Research will review applications on an ongoing basis and directly contact successful applicants. The company may also request additional information as part of their selection process.

• SAGT 2023, London
• SAGT 2022, Colchester
• SAGT 2021, Aarhus
• SAGT 2020, virtual event
• SAGT 2019, Athens
• SAGT 2018, Beijing
• SAGT 2017, L'Aquila
• SAGT 2016, Liverpool
• SAGT 2015, Saarbrücken
• SAGT 2014, Patras
• SAGT 2013, Aachen
• SAGT 2012, Barcelona
• SAGT 2011, Salerno
• SAGT 2010, Athens
• SAGT 2009, Paphos
• SAGT 2008, Paderborn

• Elias Koutsoupias, University of Oxford
• Marios Mavronicolas, University of Cyprus
• Dov Monderer, Technion
• Burkhard Monien, University of Paderborn
• Christos Papadimitriou, Columbia University
• Giuseppe Persiano, University of Salerno
• Paul Spirakis, University of Liverpool (chair)

Venue

The conference will be held in the Turing room (TBC) at the Congress Centre of Amsterdam Science Park, next to Centrum Wiskunde & Informatica (CWI).

Address: Science Park 125, 1098 XG Amsterdam

See here for location in Google Maps.

Travel / Accommodation

Please be aware that hotel prices in Amsterdam can be quite steep. We strongly recommend all participants to secure their hotel reservations as early as possible!

Hotel Recommendations

• Hotel Casa
• MEININGER Hotel (Amsterdam Amstel)
• The Manor Amsterdam
• The Lancaster Hotel Amsterdam

From these hotels, the venue can be reached in 15-30 minutes with public transport. In all public transportation, you can check in and out with a Mastercard or Visa contactless credit card and also with Apple Pay and Google Wallet.

Conference Dinner

The conference dinner will be at Restaurant Baut Oost. Information on how to get there will follow in due time. Extra tickets (50 Euro) for the conference dinner can be purchased here.

If you need an official invitation letter to apply for a visa to attend the conference, please contact Susanne van Dam.

Please make sure to provide the following details: 

Full Name
Date of Birth
Nationality
Passport Number
Affiliation
Address
Phone Number
Email Address
Title of Accepted Paper (if any) 
Role at the Conference (presenter, participant) 

In an effort to combat bullying, discrimination, and harassment, SAGT 2024 endorses the code of conduct outlined in appendix D of the Report from the Ad hoc committee to Combat Harassment and Discrimination in the Theory of Computing Community.