Showing 1–24 of 24 results for author: Haupt, A

AISPA: User-Centric System Prompt Auditing for Large Language Model Applications

Authors: Xiangning Lin, Shenzhe Zhu, Shu Yang, Zhenyu Zhang, Haoqian Zhang, Yipeng Zhao, Chengxuan Qian, Tianwei Wang, Ziheng Zhang, Zhenlong Yuan, Dingcheng Wang, Juncheng Wu, Yuan Si, Jiaxin Liu, Baolong Bi, Robert Mahari, Tobin South, Dazza Greenwood, Zexue He, Rishi Bommasani, Sophia Kazinnik, Andreas Haupt, Samuele Marro, Erik Brynjolfsson, Alex Pentland , et al. (1 additional authors not shown)

Abstract: System prompts are instructions configured by developers to govern the behaviors of foundation models in AI applications. They are used throughout commercial AI products, but are rarely disclosed to the public or regulators, creating a serious trust and accountability gap in the wide deployment of AI systems. In this paper, we introduce Artificial Intelligence System Prompt Assurance (AISPA), a us… ▽ More System prompts are instructions configured by developers to govern the behaviors of foundation models in AI applications. They are used throughout commercial AI products, but are rarely disclosed to the public or regulators, creating a serious trust and accountability gap in the wide deployment of AI systems. In this paper, we introduce Artificial Intelligence System Prompt Assurance (AISPA), a user-centric framework for systematically auditing system prompts in AI systems. AISPA examines specific parts of a system prompt and evaluates them along eight dimensions that matter to users. We then use this framework to review 3,249 instructions from system prompts in 88 commercial AI products, classifying each instruction as either protective (of users) or problematic. Our audit surfaces four core findings. First, system prompt design varies substantially across products and developers, with some organizations averaging over 60 protective instructions per product while others average fewer than 5. Second, protective instructions are widely adopted but shallow in scope: 98.9% of products contain at least one, yet only 24% cover all eight dimensions of the AISPA taxonomy. Third, system prompts have grown steadily longer and more protective of users, suggesting that user protection is becoming a more visible concern in commercial prompt design. Fourth, despite this progress, problematic instructions remain pervasive: roughly 40% of products contain at least one instruction that works against user interests, and protective and problematic instructions frequently coexist within the same prompt. Our findings highlight the need for greater transparency, standardization, and independent oversight for system prompts in commercial AI products. △ Less

Submitted 6 August, 2026; v1 submitted 30 July, 2026; originally announced July 2026.

Welfare, Improvability, and Variance: A Principal-Agent Approach to Optimal Benchmark Item Aggregation

Authors: Andreas Haupt, Justin Hartenstein, Anka Reuel, Mykel Kochenderfer, Sanmi Koyejo

Abstract: AI benchmarks have well-documented limitations, with prior work examining contamination, saturation, and construct underspecification. Aggregation has received far less attention: benchmarks are typically summarized by uniformly averaging item-level scores, implicitly treating every test item as equally valuable. We model benchmarking as a multitask principal-agent game and show that the welfare l… ▽ More AI benchmarks have well-documented limitations, with prior work examining contamination, saturation, and construct underspecification. Aggregation has received far less attention: benchmarks are typically summarized by uniformly averaging item-level scores, implicitly treating every test item as equally valuable. We model benchmarking as a multitask principal-agent game and show that the welfare loss from a benchmark is determined jointly by three item-level primitives: alignment with normative welfare priorities, marginal improvability, and performance variance. We translate the theory into an audit framework that ranks items along each of these three axes, and apply it to OLMES items using WORKBank for welfare, the EvoLM 4B suite for improvability, and the PolyPythias 410M panel for variance. The framework surfaces items that are Pareto-inferior within OLMES subject to a pro-worker welfare operationalization. All code is available at https://github.com/stair-lab/principal-agent-benchmarks. △ Less

Submitted 29 May, 2026; originally announced May 2026.

Reward Bias Substitution: Single-Axis Bias Mitigations Redirect Optimization Pressure

Authors: Max Lamparth, Daniel Fein, Andreas Haupt, Marcel Hussing, Mykel J. Kochenderfer

Abstract: Single-axis mitigations of reward-model biases (e.g., reducing proxy reliance on length, sycophancy, or style) can rotate optimization pressure onto correlated proxies rather than eliminate it, a failure mode we call reward bias substitution. The failure is enabled by a measurement-versus-optimization gap between audit and policy-induced distributions during mitigation evaluation and policy traini… ▽ More Single-axis mitigations of reward-model biases (e.g., reducing proxy reliance on length, sycophancy, or style) can rotate optimization pressure onto correlated proxies rather than eliminate it, a failure mode we call reward bias substitution. The failure is enabled by a measurement-versus-optimization gap between audit and policy-induced distributions during mitigation evaluation and policy training. We formalize mitigation outcomes into a regime taxonomy and prove that successful mitigation, bias substitution, and overcorrection produce identical observables under any audit-distribution scoring, including ranking accuracy and win-rate, even when granted oracle access to the true reward. Across published preference-learning mitigation work, no method we survey reports the evidence needed to certify successful mitigation. Augmenting evaluation with policy-induced distributions while tracking multiple biases provably closes the gap, and we translate this into actionable prescriptions for mitigation methods and benchmarks. We demonstrate bias substitution in language model RLHF, where a length penalty during GRPO training compresses responses as intended yet redirects optimization pressure onto confidence calibration, driving the policy into overconfidence while factual free-form accuracy falls. We also show a published length-debiasing operator that zeroes reward-length correlation on the audit distribution but reintroduces bias under best-of-N selection on three of four SOTA reward models, and a length-sycophancy coupling whose direction reverses under human-LLM judge disagreement. △ Less

Submitted 28 May, 2026; v1 submitted 27 May, 2026; originally announced May 2026.

Comments: Improved readability (mostly appendix D)

General Preference Reinforcement Learning

Authors: Muhammad Umer, Muhammad Ahmed Mohsin, Ahsan Bilal, Arslan Chaudhry, Andreas Haupt, Sanmi Koyejo, Emily Fox, John M. Cioffi

Abstract: Post-training has split large language model (LLM) alignment into two largely disconnected tracks. Online reinforcement learning (RL) with verifiable rewards drives emergent reasoning on math and code but depends on a programmatic verifier that cannot reach open-ended tasks, while preference optimization handles open-ended generation yet forgoes the continuous exploration that powers online RL. Cl… ▽ More Post-training has split large language model (LLM) alignment into two largely disconnected tracks. Online reinforcement learning (RL) with verifiable rewards drives emergent reasoning on math and code but depends on a programmatic verifier that cannot reach open-ended tasks, while preference optimization handles open-ended generation yet forgoes the continuous exploration that powers online RL. Closing this gap requires a verifier for open-ended quality, but a scalar reward model is the wrong shape for the job. Quality is multi-dimensional, and any scalar score is an incomplete proxy that lets online RL collapse onto whichever axis the score is most sensitive to. We turn instead to the General Preference Model (GPM), which embeds responses into $k$ skew-symmetric subspaces and represents preference as a structured, intransitivity-aware comparison. Building on this, we propose General Preference Reinforcement Learning (GPRL), which carries the $k$-way structure through to the policy update. GPRL computes per-dimension group-relative advantages, normalizes each on its own scale so no axis can dominate, and aggregates them with context-dependent eigenvalues. The same structure powers a closed-loop drift monitor that detects single-axis exploitation and corrects it on the fly by reweighting dimensions and tightening the trust region. Starting from $\texttt{Llama-3-8B-Instruct}$, GPRL reaches a length-controlled win rate of $56.51\%$ on AlpacaEval~2.0 while also outperforming SimPO and SPPO on Arena-Hard, MT-Bench, and WildBench by resisting reward hacking across extended training runs. △ Less

Submitted 21 May, 2026; v1 submitted 18 May, 2026; originally announced May 2026.

The Collapse of Heterogeneity in Silicon Philosophers

Authors: Yuanming Shi, Andreas Haupt

Abstract: Silicon samples are increasingly used as a low-cost substitute for human panels and have been shown to reproduce aggregate human opinion with high fidelity. We show that, in the alignment-relevant domain of philosophy, silicon samples systematically collapse heterogeneity. Using data from $N = {277}$ professional philosophers drawn from PhilPeople profiles, we evaluate seven proprietary and open-s… ▽ More Silicon samples are increasingly used as a low-cost substitute for human panels and have been shown to reproduce aggregate human opinion with high fidelity. We show that, in the alignment-relevant domain of philosophy, silicon samples systematically collapse heterogeneity. Using data from $N = {277}$ professional philosophers drawn from PhilPeople profiles, we evaluate seven proprietary and open-source large language models on their ability to replicate individual philosophical positions and to preserve cross-question correlation structures across philosophical domains. We find that language models substantially over-correlate philosophical judgments, producing artificial consensus across domains. This collapse is associated in part with specialist effects, whereby models implicitly assume that domain specialists hold highly similar philosophical views. We assess the robustness of these findings by studying the impact of DPO fine-tuning and by validating results against the full PhilPapers 2020 Survey ($N = {1785}$). We conclude by discussing implications for alignment, evaluation, and the use of silicon samples as substitutes for human judgment. The code of this project can be found at https://github.com/stanford-del/silicon-philosophers. △ Less

Submitted 29 April, 2026; v1 submitted 26 April, 2026; originally announced April 2026.

Latent Adversarial Regularization for Offline Preference Optimization

Authors: Enyi Jiang, Yibo Jacky Zhang, Yinglun Xu, Andreas Haupt, Nancy Amato, Sanmi Koyejo

Abstract: Learning from human feedback typically relies on preference optimization that constrains policy updates through token-level regularization. However, preference optimization for language models is particularly challenging because token-space similarity does not imply semantic or behavioral similarity. To address this challenge, we leverage latent-space regularization for language model preference o… ▽ More Learning from human feedback typically relies on preference optimization that constrains policy updates through token-level regularization. However, preference optimization for language models is particularly challenging because token-space similarity does not imply semantic or behavioral similarity. To address this challenge, we leverage latent-space regularization for language model preference optimization. We introduce GANPO, which achieves latent-space regularization by penalizing divergence between the internal representations of a policy model and a reference model. Given that latent representations are not associated with explicit probability densities, we adopt an adversarial approach inspired by GANs to minimize latent-space divergence. We integrate GANPO as a regularizer into existing offline preference optimization objectives. Experiments across multiple model architectures and tasks show consistent improvements from latent-space regularization. Further, by comparing GANPO-induced inferential biases with those from token-level regularization, we find that GANPO provides more robust structural feedback under distributional shift and noise while maintaining comparable downstream performance with minor computational overhead. △ Less

Submitted 2 February, 2026; v1 submitted 29 January, 2026; originally announced January 2026.

Full-Stack Alignment: Co-Aligning AI and Institutions with Thick Models of Value

Authors: Joe Edelman, Tan Zhi-Xuan, Ryan Lowe, Oliver Klingefjord, Vincent Wang-Mascianica, Matija Franklin, Ryan Othniel Kearns, Ellie Hain, Atrisha Sarkar, Michiel Bakker, Fazl Barez, David Duvenaud, Jakob Foerster, Iason Gabriel, Joseph Gubbels, Bryce Goodman, Andreas Haupt, Jobst Heitzig, Julian Jara-Ettinger, Atoosa Kasirzadeh, James Ravi Kirkpatrick, Andrew Koh, W. Bradley Knox, Philipp Koralus, Joel Lehman , et al. (8 additional authors not shown)

Abstract: Beneficial societal outcomes cannot be guaranteed by aligning individual AI systems with the intentions of their operators or users. Even an AI system that is perfectly aligned to the intentions of its operating organization can lead to bad outcomes if the goals of that organization are misaligned with those of other institutions and individuals. For this reason, we need full-stack alignment, the… ▽ More Beneficial societal outcomes cannot be guaranteed by aligning individual AI systems with the intentions of their operators or users. Even an AI system that is perfectly aligned to the intentions of its operating organization can lead to bad outcomes if the goals of that organization are misaligned with those of other institutions and individuals. For this reason, we need full-stack alignment, the concurrent alignment of AI systems and the institutions that shape them with what people value. This can be done without imposing a particular vision of individual or collective flourishing. We argue that current approaches for representing values, such as utility functions, preference orderings, or unstructured text, struggle to address these and other issues effectively. They struggle to distinguish values from other signals, to support principled normative reasoning, and to model collective goods. We propose thick models of value will be needed. These structure the way values and norms are represented, enabling systems to distinguish enduring values from fleeting preferences, to model the social embedding of individual choices, and to reason normatively, applying values in new domains. We demonstrate this approach in five areas: AI value stewardship, normatively competent agents, win-win negotiation systems, meaning-preserving economic mechanisms, and democratic regulatory institutions. △ Less

Submitted 2 December, 2025; originally announced December 2025.

Preference Measurement Error, Concentration in Recommendation Systems, and Persuasion

Authors: Andreas Haupt

Abstract: Algorithmic recommendation based on noisy preference measurement is prevalent in recommendation systems. This paper discusses the consequences of such recommendation on market concentration and inequality. Binary types denoting a statistical majority and minority are noisily revealed through a statistical experiment. The achievable utilities and recommendation shares for the two groups can be anal… ▽ More Algorithmic recommendation based on noisy preference measurement is prevalent in recommendation systems. This paper discusses the consequences of such recommendation on market concentration and inequality. Binary types denoting a statistical majority and minority are noisily revealed through a statistical experiment. The achievable utilities and recommendation shares for the two groups can be analyzed as a Bayesian Persuasion problem. While under arbitrary noise structures, effects on concentration compared to a full-information market are ambiguous, under symmetric noise, concentration increases and consumer welfare becomes more unequal. We define symmetric statistical experiments and analyze persuasion under a restriction to such experiments, which may be of independent interest. △ Less

Submitted 19 October, 2025; originally announced October 2025.

Comments: 12 pages, 3 figures

Don't Walk the Line: Boundary Guidance for Filtered Generation

Authors: Sarah Ball, Andreas Haupt

Abstract: Generative models are increasingly paired with safety classifiers that filter harmful or undesirable outputs. A common strategy is to fine-tune the generator to reduce the probability of being filtered, but this can be suboptimal: it often pushes the model toward producing samples near the classifier's decision boundary, increasing both false positives and false negatives. We propose Boundary Guid… ▽ More Generative models are increasingly paired with safety classifiers that filter harmful or undesirable outputs. A common strategy is to fine-tune the generator to reduce the probability of being filtered, but this can be suboptimal: it often pushes the model toward producing samples near the classifier's decision boundary, increasing both false positives and false negatives. We propose Boundary Guidance, a reinforcement learning fine-tuning method that explicitly steers generation away from the classifier's margin. On a benchmark of jailbreak, ambiguous, and longcontext prompts, Boundary Guidance improves both the safety and the utility of outputs, as judged by LLM-as-a-Judge evaluations. Comprehensive ablations across model scales and reward designs demonstrate the robustness of our approach. △ Less

Submitted 4 August, 2026; v1 submitted 13 October, 2025; originally announced October 2025.

Comments: Accepted at ICML 2026

Scaling Human Judgment in Community Notes with LLMs

Authors: Haiwen Li, Soham De, Manon Revel, Andreas Haupt, Brad Miller, Keith Coleman, Jay Baxter, Martin Saveski, Michiel A. Bakker

Abstract: This paper argues for a new paradigm for Community Notes in the LLM era: an open ecosystem where both humans and LLMs can write notes, and the decision of which notes are helpful enough to show remains in the hands of humans. This approach can accelerate the delivery of notes, while maintaining trust and legitimacy through Community Notes' foundational principle: A community of diverse human rater… ▽ More This paper argues for a new paradigm for Community Notes in the LLM era: an open ecosystem where both humans and LLMs can write notes, and the decision of which notes are helpful enough to show remains in the hands of humans. This approach can accelerate the delivery of notes, while maintaining trust and legitimacy through Community Notes' foundational principle: A community of diverse human raters collectively serve as the ultimate evaluator and arbiter of what is helpful. Further, the feedback from this diverse community can be used to improve LLMs' ability to produce accurate, unbiased, broadly helpful notes--what we term Reinforcement Learning from Community Feedback (RLCF). This becomes a two-way street: LLMs serve as an asset to humans--helping deliver context quickly and with minimal effort--while human feedback, in turn, enhances the performance of LLMs. This paper describes how such a system can work, its benefits, key new risks and challenges it introduces, and a research agenda to solve those challenges and realize the potential of this approach. △ Less

Submitted 30 June, 2025; originally announced June 2025.

Position: Machine Learning Conferences Should Establish a "Refutations and Critiques" Track

Authors: Rylan Schaeffer, Joshua Kazdan, Yegor Denisov-Blanch, Brando Miranda, Matthias Gerstgrasser, Susan Zhang, Andreas Haupt, Isha Gupta, Elyas Obbad, Jesse Dodge, Jessica Zosa Forde, Francesco Orabona, Sanmi Koyejo, David Donoho

Abstract: Science progresses by iteratively advancing and correcting humanity's understanding of the world. In machine learning (ML) research, rapid advancements have led to an explosion of publications, but have also led to misleading, incorrect, flawed or perhaps even fraudulent studies being accepted and sometimes highlighted at ML conferences due to the fallibility of peer review. While such mistakes ar… ▽ More Science progresses by iteratively advancing and correcting humanity's understanding of the world. In machine learning (ML) research, rapid advancements have led to an explosion of publications, but have also led to misleading, incorrect, flawed or perhaps even fraudulent studies being accepted and sometimes highlighted at ML conferences due to the fallibility of peer review. While such mistakes are understandable, ML conferences do not offer robust processes to help the field systematically correct when such errors are made. This position paper argues that ML conferences should establish a dedicated "Refutations and Critiques" (R&C) Track. This R&C Track would provide a high-profile, reputable platform to support vital research that critically challenges prior research, thereby fostering a dynamic self-correcting research ecosystem. We discuss key considerations including track design, review principles, potential pitfalls, and provide an illustrative example submission concerning a recent ICLR 2025 Oral. We conclude that ML conferences should create official, reputable mechanisms to help ML research self-correct. △ Less

Submitted 6 July, 2025; v1 submitted 23 June, 2025; originally announced June 2025.

MassSpecGym: A benchmark for the discovery and identification of molecules

Authors: Roman Bushuiev, Anton Bushuiev, Niek F. de Jonge, Adamo Young, Fleming Kretschmer, Raman Samusevich, Janne Heirman, Fei Wang, Luke Zhang, Kai Dührkop, Marcus Ludwig, Nils A. Haupt, Apurva Kalia, Corinna Brungs, Robin Schmid, Russell Greiner, Bo Wang, David S. Wishart, Li-Ping Liu, Juho Rousu, Wout Bittremieux, Hannes Rost, Tytus D. Mak, Soha Hassoun, Florian Huber , et al. (5 additional authors not shown)

Abstract: The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a resu… ▽ More The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a result, the vast majority of acquired MS/MS spectra remain uninterpreted, thereby limiting our understanding of the underlying (bio)chemical processes. Despite decades of progress in machine learning applications for predicting molecular structures from MS/MS spectra, the development of new methods is severely hindered by the lack of standard datasets and evaluation protocols. To address this problem, we propose MassSpecGym -- the first comprehensive benchmark for the discovery and identification of molecules from MS/MS data. Our benchmark comprises the largest publicly available collection of high-quality labeled MS/MS spectra and defines three MS/MS annotation challenges: de novo molecular structure generation, molecule retrieval, and spectrum simulation. It includes new evaluation metrics and a generalization-demanding data split, therefore standardizing the MS/MS annotation tasks and rendering the problem accessible to the broad machine learning community. MassSpecGym is publicly available at https://github.com/pluskal-lab/MassSpecGym. △ Less

Submitted 14 February, 2025; v1 submitted 30 October, 2024; originally announced October 2024.

Convex Markov Games: A New Frontier for Multi-Agent Reinforcement Learning

Authors: Ian Gemp, Andreas Haupt, Luke Marris, Siqi Liu, Georgios Piliouras

Abstract: Behavioral diversity, expert imitation, fairness, safety goals and others give rise to preferences in sequential decision making domains that do not decompose additively across time. We introduce the class of convex Markov games that allow general convex preferences over occupancy measures. Despite infinite time horizon and strictly higher generality than Markov games, pure strategy Nash equilibri… ▽ More Behavioral diversity, expert imitation, fairness, safety goals and others give rise to preferences in sequential decision making domains that do not decompose additively across time. We introduce the class of convex Markov games that allow general convex preferences over occupancy measures. Despite infinite time horizon and strictly higher generality than Markov games, pure strategy Nash equilibria exist. Furthermore, equilibria can be approximated empirically by performing gradient descent on an upper bound of exploitability. Our experiments reveal novel solutions to classic repeated normal-form games, find fair solutions in a repeated asymmetric coordination game, and prioritize safe long-term behavior in a robot warehouse environment. In the prisoner's dilemma, our algorithm leverages transient imitation to find a policy profile that deviates from observed human play only slightly, yet achieves higher per-player utility while also being three orders of magnitude less exploitable. △ Less

Submitted 16 June, 2025; v1 submitted 21 October, 2024; originally announced October 2024.

Comments: Published at ICML 2025

Black-Box Access is Insufficient for Rigorous AI Audits

Authors: Stephen Casper, Carson Ezell, Charlotte Siegmann, Noam Kolt, Taylor Lynn Curtis, Benjamin Bucknall, Andreas Haupt, Kevin Wei, Jérémy Scheurer, Marius Hobbhahn, Lee Sharkey, Satyapriya Krishna, Marvin Von Hagen, Silas Alberti, Alan Chan, Qinyi Sun, Michael Gerovitch, David Bau, Max Tegmark, David Krueger, Dylan Hadfield-Menell

Abstract: External audits of AI systems are increasingly recognized as a key mechanism for AI governance. The effectiveness of an audit, however, depends on the degree of access granted to auditors. Recent audits of state-of-the-art AI systems have primarily relied on black-box access, in which auditors can only query the system and observe its outputs. However, white-box access to the system's inner workin… ▽ More External audits of AI systems are increasingly recognized as a key mechanism for AI governance. The effectiveness of an audit, however, depends on the degree of access granted to auditors. Recent audits of state-of-the-art AI systems have primarily relied on black-box access, in which auditors can only query the system and observe its outputs. However, white-box access to the system's inner workings (e.g., weights, activations, gradients) allows an auditor to perform stronger attacks, more thoroughly interpret models, and conduct fine-tuning. Meanwhile, outside-the-box access to training and deployment information (e.g., methodology, code, documentation, data, deployment details, findings from internal evaluations) allows auditors to scrutinize the development process and design more targeted evaluations. In this paper, we examine the limitations of black-box audits and the advantages of white- and outside-the-box audits. We also discuss technical, physical, and legal safeguards for performing these audits with minimal security risks. Given that different forms of access can lead to very different levels of evaluation, we conclude that (1) transparency regarding the access and methods used by auditors is necessary to properly interpret audit results, and (2) white- and outside-the-box access allow for substantially more scrutiny than black-box access alone. △ Less

Submitted 29 May, 2024; v1 submitted 25 January, 2024; originally announced January 2024.

Comments: FAccT 2024

Journal ref: The 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT '24), June 3-6, 2024, Rio de Janeiro, Brazil

Steering No-Regret Learners to a Desired Equilibrium

Authors: Brian Hu Zhang, Gabriele Farina, Ioannis Anagnostides, Federico Cacciamani, Stephen Marcus McAleer, Andreas Alexander Haupt, Andrea Celli, Nicola Gatti, Vincent Conitzer, Tuomas Sandholm

Abstract: A mediator observes no-regret learners playing an extensive-form game repeatedly across $T$ rounds. The mediator attempts to steer players toward some desirable predetermined equilibrium by giving (nonnegative) payments to players. We call this the steering problem. The steering problem captures problems several problems of interest, among them equilibrium selection and information design (persuas… ▽ More A mediator observes no-regret learners playing an extensive-form game repeatedly across $T$ rounds. The mediator attempts to steer players toward some desirable predetermined equilibrium by giving (nonnegative) payments to players. We call this the steering problem. The steering problem captures problems several problems of interest, among them equilibrium selection and information design (persuasion). If the mediator's budget is unbounded, steering is trivial because the mediator can simply pay the players to play desirable actions. We study two bounds on the mediator's payments: a total budget and a per-round budget. If the mediator's total budget does not grow with $T$, we show that steering is impossible. However, we show that it is enough for the total budget to grow sublinearly with $T$, that is, for the average payment to vanish. When players' full strategies are observed at each round, we show that constant per-round budgets permit steering. In the more challenging setting where only trajectories through the game tree are observable, we show that steering is impossible with constant per-round budgets in general extensive-form games, but possible in normal-form games or if the per-round budget may itself depend on $T$. We also show how our results can be generalized to the case when the equilibrium is being computed online while steering is happening. We supplement our theoretical positive results with experiments highlighting the efficacy of steering in large games. △ Less

Submitted 17 March, 2026; v1 submitted 8 June, 2023; originally announced June 2023.

Computing Optimal Equilibria and Mechanisms via Learning in Zero-Sum Extensive-Form Games

Authors: Brian Hu Zhang, Gabriele Farina, Ioannis Anagnostides, Federico Cacciamani, Stephen Marcus McAleer, Andreas Alexander Haupt, Andrea Celli, Nicola Gatti, Vincent Conitzer, Tuomas Sandholm

Abstract: We introduce a new approach for computing optimal equilibria via learning in games. It applies to extensive-form settings with any number of players, including mechanism design, information design, and solution concepts such as correlated, communication, and certification equilibria. We observe that optimal equilibria are minimax equilibrium strategies of a player in an extensive-form zero-sum gam… ▽ More We introduce a new approach for computing optimal equilibria via learning in games. It applies to extensive-form settings with any number of players, including mechanism design, information design, and solution concepts such as correlated, communication, and certification equilibria. We observe that optimal equilibria are minimax equilibrium strategies of a player in an extensive-form zero-sum game. This reformulation allows to apply techniques for learning in zero-sum games, yielding the first learning dynamics that converge to optimal equilibria, not only in empirical averages, but also in iterates. We demonstrate the practical scalability and flexibility of our approach by attaining state-of-the-art performance in benchmark tabular games, and by computing an optimal mechanism for a sequential auction design problem using deep reinforcement learning. △ Less

Submitted 23 May, 2024; v1 submitted 8 June, 2023; originally announced June 2023.

Recommending to Strategic Users

Authors: Andreas Haupt, Dylan Hadfield-Menell, Chara Podimata

Abstract: Recommendation systems are pervasive in the digital economy. An important assumption in many deployed systems is that user consumption reflects user preferences in a static sense: users consume the content they like with no other considerations in mind. However, as we document in a large-scale online survey, users do choose content strategically to influence the types of content they get recommend… ▽ More Recommendation systems are pervasive in the digital economy. An important assumption in many deployed systems is that user consumption reflects user preferences in a static sense: users consume the content they like with no other considerations in mind. However, as we document in a large-scale online survey, users do choose content strategically to influence the types of content they get recommended in the future. We model this user behavior as a two-stage noisy signalling game between the recommendation system and users: the recommendation system initially commits to a recommendation policy, presents content to the users during a cold start phase which the users choose to strategically consume in order to affect the types of content they will be recommended in a recommendation phase. We show that in equilibrium, users engage in behaviors that accentuate their differences to users of different preference profiles. In addition, (statistical) minorities out of fear of losing their minority content exposition may not consume content that is liked by mainstream users. We next propose three interventions that may improve recommendation quality (both on average and for minorities) when taking into account strategic consumption: (1) Adopting a recommendation system policy that uses preferences from a prior, (2) Communicating to users that universally liked ("mainstream") content will not be used as basis of recommendation, and (3) Serving content that is personalized-enough yet expected to be liked in the beginning. Finally, we describe a methodology to inform applied theory modeling with survey results. △ Less

Submitted 13 February, 2023; originally announced February 2023.

Comments: 35 pages

Certification Design for a Competitive Market

Authors: Andreas A. Haupt, Nicole Immorlica, Brendan Lucier

Abstract: Motivated by applications such as voluntary carbon markets and educational testing, we consider a market for goods with varying but hidden levels of quality in the presence of a third-party certifier. The certifier can provide informative signals about the quality of products, and can charge for this service. Sellers choose both the quality of the product they produce and a certification. Prices a… ▽ More Motivated by applications such as voluntary carbon markets and educational testing, we consider a market for goods with varying but hidden levels of quality in the presence of a third-party certifier. The certifier can provide informative signals about the quality of products, and can charge for this service. Sellers choose both the quality of the product they produce and a certification. Prices are then determined in a competitive market. Under a single-crossing condition, we show that the levels of certification chosen by producers are uniquely determined at equilibrium. We then show how to reduce a revenue-maximizing certifier's problem to a monopolistic pricing problem with non-linear valuations, and design an FPTAS for computing the optimal slate of certificates and their prices. In general, both the welfare-optimal and revenue-optimal slate of certificates can be arbitrarily large. △ Less

Submitted 31 January, 2023; originally announced January 2023.

Comments: 22 pages, 1 figure

Formal Contracts Mitigate Social Dilemmas in Multi-Agent RL

Authors: Andreas A. Haupt, Phillip J. K. Christoffersen, Mehul Damani, Dylan Hadfield-Menell

Abstract: Multi-agent Reinforcement Learning (MARL) is a powerful tool for training autonomous agents acting independently in a common environment. However, it can lead to sub-optimal behavior when individual incentives and group incentives diverge. Humans are remarkably capable at solving these social dilemmas. It is an open problem in MARL to replicate such cooperative behaviors in selfish agents. In this… ▽ More Multi-agent Reinforcement Learning (MARL) is a powerful tool for training autonomous agents acting independently in a common environment. However, it can lead to sub-optimal behavior when individual incentives and group incentives diverge. Humans are remarkably capable at solving these social dilemmas. It is an open problem in MARL to replicate such cooperative behaviors in selfish agents. In this work, we draw upon the idea of formal contracting from economics to overcome diverging incentives between agents in MARL. We propose an augmentation to a Markov game where agents voluntarily agree to binding transfers of reward, under pre-specified conditions. Our contributions are theoretical and empirical. First, we show that this augmentation makes all subgame-perfect equilibria of all Fully Observable Markov Games exhibit socially optimal behavior, given a sufficiently rich space of contracts. Next, we show that for general contract spaces, and even under partial observability, richer contract spaces lead to higher welfare. Hence, contract space design solves an exploration-exploitation tradeoff, sidestepping incentive issues. We complement our theoretical analysis with experiments. Issues of exploration in the contracting augmentation are mitigated using a training methodology inspired by multi-objective reinforcement learning: Multi-Objective Contract Augmentation Learning (MOCA). We test our methodology in static, single-move games, as well as dynamic domains that simulate traffic, pollution management and common pool resource management. △ Less

Submitted 29 January, 2024; v1 submitted 22 August, 2022; originally announced August 2022.

Towards Psychologically-Grounded Dynamic Preference Models

Authors: Mihaela Curmei, Andreas Haupt, Dylan Hadfield-Menell, Benjamin Recht

Abstract: Designing recommendation systems that serve content aligned with time varying preferences requires proper accounting of the feedback effects of recommendations on human behavior and psychological condition. We argue that modeling the influence of recommendations on people's preferences must be grounded in psychologically plausible models. We contribute a methodology for developing grounded dynamic… ▽ More Designing recommendation systems that serve content aligned with time varying preferences requires proper accounting of the feedback effects of recommendations on human behavior and psychological condition. We argue that modeling the influence of recommendations on people's preferences must be grounded in psychologically plausible models. We contribute a methodology for developing grounded dynamic preference models. We demonstrate this method with models that capture three classic effects from the psychology literature: Mere-Exposure, Operant Conditioning, and Hedonic Adaptation. We conduct simulation-based studies to show that the psychological models manifest distinct behaviors that can inform system design. Our study has two direct implications for dynamic user modeling in recommendation systems. First, the methodology we outline is broadly applicable for psychologically grounding dynamic preference models. It allows us to critique recent contributions based on their limited discussion of psychological foundation and their implausible predictions. Second, we discuss implications of dynamic preference models for recommendation systems evaluation and design. In an example, we show that engagement and diversity metrics may be unable to capture desirable recommendation system performance. △ Less

Submitted 6 August, 2022; v1 submitted 1 August, 2022; originally announced August 2022.

Comments: In Sixteenth ACM Conference on Recommender Systems, September 18-23, 2022, Seattle, WA, USA, 14 pages

Risk Preferences of Learning Algorithms

Authors: Andreas Haupt, Aroon Narayanan

Abstract: Agents' learning from feedback shapes economic outcomes, and many economic decision-makers today employ learning algorithms to make consequential choices. This note shows that a widely used learning algorithm, $\varepsilon$-Greedy, exhibits emergent risk aversion: it prefers actions with lower variance. When presented with actions of the same expectation, under a wide range of conditions,… ▽ More Agents' learning from feedback shapes economic outcomes, and many economic decision-makers today employ learning algorithms to make consequential choices. This note shows that a widely used learning algorithm, $\varepsilon$-Greedy, exhibits emergent risk aversion: it prefers actions with lower variance. When presented with actions of the same expectation, under a wide range of conditions, $\varepsilon$-Greedy chooses the lower-variance action with probability approaching one. This emergent preference can have wide-ranging consequences, ranging from concerns about fairness to homogenization, and holds transiently even when the riskier action has a strictly higher expected payoff. We discuss two methods to correct this bias. The first method requires the algorithm to reweight data as a function of how likely the actions were to be chosen. The second requires the algorithm to have optimistic estimates of actions for which it has not collected much data. We show that risk-neutrality is restored with these corrections. △ Less

Submitted 12 December, 2023; v1 submitted 9 May, 2022; originally announced May 2022.

Comments: 11 pages, 6 figures

The Optimality of Upgrade Pricing

Authors: Dirk Bergemann, Alessandro Bonatti, Andreas Haupt, Alex Smolin

Abstract: We consider a multiproduct monopoly pricing model. We provide sufficient conditions under which the optimal mechanism can be implemented via upgrade pricing -- a menu of product bundles that are nested in the strong set order. Our approach exploits duality methods to identify conditions on the distribution of consumer types under which (a) each product is purchased by the same set of buyers as und… ▽ More We consider a multiproduct monopoly pricing model. We provide sufficient conditions under which the optimal mechanism can be implemented via upgrade pricing -- a menu of product bundles that are nested in the strong set order. Our approach exploits duality methods to identify conditions on the distribution of consumer types under which (a) each product is purchased by the same set of buyers as under separate monopoly pricing (though the transfers can be different), and (b) these sets are nested. We exhibit two distinct sets of sufficient conditions. The first set of conditions is given by a weak version of monotonicity of types and virtual values, while maintaining a regularity assumption, i.e., that the product-by-product revenue curves are single-peaked. The second set of conditions establishes the optimality of upgrade pricing for type spaces with monotone marginal rates of substitution (MRS) -- the relative preference ratios for any two products are monotone across types. The monotone MRS condition allows us to relax the earlier regularity assumption. Under both sets of conditions, we fully characterize the product bundles and prices that form the optimal upgrade pricing menu. Finally, we show that, if the consumer's types are monotone, the seller can equivalently post a vector of single-item prices: upgrade pricing and separate pricing are equivalent. △ Less

Submitted 2 December, 2021; v1 submitted 21 July, 2021; originally announced July 2021.

Comments: 22 pages, 4 figures

Report number: Web and Internet Economics 2021

Prior-Independent Auctions for the Demand Side of Federated Learning

Authors: Andreas Haupt, Vaikkunth Mugunthan

Abstract: Federated learning (FL) is a paradigm that allows distributed clients to learn a shared machine learning model without sharing their sensitive training data. While largely decentralized, FL requires resources to fund a central orchestrator or to reimburse contributors of datasets to incentivize participation. Inspired by insights from prior-independent auction design, we propose a mechanism, FIPIA… ▽ More Federated learning (FL) is a paradigm that allows distributed clients to learn a shared machine learning model without sharing their sensitive training data. While largely decentralized, FL requires resources to fund a central orchestrator or to reimburse contributors of datasets to incentivize participation. Inspired by insights from prior-independent auction design, we propose a mechanism, FIPIA (Federated Incentive Payments via Prior-Independent Auctions), to collect monetary contributions from self-interested clients. The mechanism operates in the semi-honest trust model and works even if clients have a heterogeneous interest in receiving high-quality models, and the server does not know the clients' level of interest. We run experiments on the MNIST, FashionMNIST, and CIFAR-10 datasets to test clients' model quality under FIPIA and FIPIA's incentive properties. △ Less

Submitted 13 April, 2021; v1 submitted 26 March, 2021; originally announced March 2021.

Classification on Large Networks: A Quantitative Bound via Motifs and Graphons

Authors: Andreas Haupt, Mohammad Khatami, Thomas Schultz, Ngoc Mai Tran

Abstract: When each data point is a large graph, graph statistics such as densities of certain subgraphs (motifs) can be used as feature vectors for machine learning. While intuitive, motif counts are expensive to compute and difficult to work with theoretically. Via graphon theory, we give an explicit quantitative bound for the ability of motif homomorphisms to distinguish large networks under both generat… ▽ More When each data point is a large graph, graph statistics such as densities of certain subgraphs (motifs) can be used as feature vectors for machine learning. While intuitive, motif counts are expensive to compute and difficult to work with theoretically. Via graphon theory, we give an explicit quantitative bound for the ability of motif homomorphisms to distinguish large networks under both generative and sampling noise. Furthermore, we give similar bounds for the graph spectrum and connect it to homomorphism densities of cycles. This results in an easily computable classifier on graph data with theoretical performance guarantee. Our method yields competitive results on classification tasks for the autoimmune disease Lupus Erythematosus. △ Less

Submitted 24 October, 2017; originally announced October 2017.

Comments: 17 pages, 2 figures, 1 table

MSC Class: 68T05; 05C80; 62G99