Classical treatments of machine learning rely on the assumption that the data, after deployment, resembles the data the model was trained on. However, as machine learning models are increasingly used to make consequential decisions about people, individuals often react strategically to the deployed model. These strategic behaviors---which effectively invalidate the predictive models---have opened up new avenues of research and added new challenges to the deployment of machine learning algorithms in the real world.
Different aspects of strategic behavior have been studied by several communities both within and outside of machine learning. For example, the growing literature on strategic classification studies algorithms for finding strategy-robust decision rules, as well as the properties of such rules. Behavioral economics aims to understand and model people’s strategic responses. Recent works on learning in games study optimization algorithms for finding meaningful equilibria and solution concepts in competitive environments.
This workshop aims to create a dialogue between these different communities, all studying aspects of decision-making and learning with strategic feedback. The goal is to identify common points of interest and open problems in the different subareas, as well as to encourage cross-disciplinary collaboration.
| Opening remarks (Introduction) | |
| Analysis and interventions in large network games: graphon games and graphon contagion (Invited Talk) | |
| Closing the loop in Machine Learning: Learning to optimize with decision dependent data (Invited Talk) | |
| Strategic Classification and the Quest for the Holy Grail (Invited Talk) | |
| Panel 1 (Discussion Panel) | |
| The Platform Design Problem (Contributed talk) | |
| Learning Losses for Strategic Classification (Contributed talk) | |
| Unfairness Despite Awareness: Group-Fair Classification with Strategic Agents (Contributed talk) | |
| Test-optional Policies: Overcoming Strategic Behavior and Informational Gaps (Contributed talk) | |
| Poster Session (Poster Session, social & break) | |
| Microfoundations of Algorithmic decisions (Invited Talk) | |
| Improving Information from Manipulable Data (Invited Talk) | |
| Revisiting Dynamics in Strategic ML (Invited Talk) | |
| Panel 2 (Discussion Panel) | |
| Algorithmic Monoculture and Social Welfare (Invited Talk) | |
| Leveraging strategic interactions for causal discovery (Invited Talk) | |
| Online intermediation in legacy industries: The effect of reservation platforms on restaurants’ prices and survival (Invited Talk) | |
| Panel 3 (Discussion Panel) | |
| Social | |
| Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environments (Poster) | |
| Alternative Microfoundations for Strategic Classification (Poster) | |
| Nash Convergence of Mean-Based Learning Algorithms in First Price Auctions (Poster) | |
| Normative disagreement as a challenge for Cooperative AI (Poster) | |
| Promoting Resilience of Multi-Agent Reinforcement Learning via Confusion-Based Communication (Poster) | |
| Bounded Rationality for Multi-Agent Motion Planning and Behavior Learning (Poster) | |
| Test-optional Policies: Overcoming Strategic Behavior and Informational Gaps (Poster) | |
| Learning Losses for Strategic Classification (Poster) | |
| Interactive Robust Policy Optimization for Multi-Agent Reinforcement Learning (Poster) | |
| Price Discovery and Efficiency in Waiting Lists: A Connection to Stochastic Gradient Descent (Poster) | |
| Gaming Helps! Learning from Strategic Interactions in Natural Dynamics (Poster) | |
| The Strategic Perceptron (Poster) | |
| Unfairness Despite Awareness: Group-Fair Classification with Strategic Agents (Poster) | |
| Improving Fairness in Credit Lending Models using Subgroup Threshold Optimization (Poster) | |
| Regret, stability, and fairness in matching markets with bandit learners (Poster) | |
| Models of fairness in federated learning (Poster) | |
| Learning through Recourse under Censoring (Poster) | |
| The Price of Incentivizing Exploration: A Characterization via Thompson Sampling and Sample Complexity (Poster) | |
| Learning in Matrix Games can be Arbitrarily Complex (Poster) | |
| The Platform Design Problem (Poster) | |
| Bayesian Persuasion for Algorithmic Recourse (Poster) | |
| Timing is Money: The Impact of Arrival Order in Beta-Bernoulli Prediction Markets (Poster) | |
| Scoring Rules for Performative Binary Prediction (Poster) | |
| Information Discrepancy in Strategic Learning (Poster) | |
| Exploration-Exploitation in Multi-Agent Competition: Convergence with Bounded Rationality (Poster) | |
| Estimation of Standard Asymmetric Auction Models (Poster) | |
| Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games (Poster) | |
| Pseudo-Competitive Games and Algorithmic Pricing (Poster) | |
| Pessimistic Offline Reinforcement Learning with Multiple Agents (Poster) | |
| Improving Robustness of Malware Classifiers using Adversarial Strings Generated from Perturbed Latent Representations (Poster) | |
| Near-Optimal No-Regret Learning in General Games (Poster) | |
| Efficient Competitions and Online Learning with Strategic Forecasters (Poster) | |
| Negotiating networks in oligopoly markets for price sensitive products (Poster) | |
| One for One, or All for All: Equilibria and Optimality of Collaboration in Federated Learning (Poster) | |
| On classification of strategic agents who can both game and improve (Poster) | |
| Coopetition Against an Amazon (Poster) | |
| When to Call Your Neighbor? Strategic Communication in Cooperative Stochastic Bandits (Poster) | |
| Strategic Classification Made Practical (Poster) | |
| Strategic clustering (Poster) | |
| Reward-Free Attacks in Multi-Agent Reinforcement Learning (Poster) | |
| Game Redesign in No-regret Game Playing (Poster) | |
| Strategic Classification in the Dark (Poster) | |
| Approximating Bayes Nash Equilibria in Auction Games via Gradient Dynamics (Poster) | |
| Exploration and Incentives in Reinforcement Learning (Poster) | |