Workshop - RL & GT

Reinforcement Learning and Game Theory sit at the core of decision-making in interactive, uncertain, multi-agent environments, also characterized by diverse human feedback. As AI systems increasingly operate in shared infrastructures, marketplaces, autonomous platforms, and socio-technical ecosystems, there is a growing need for methods that can reason not only about optimal individual behavior, but also about strategic interaction, incentives, equilibrium, robustness, and adaptation. This workshop is motivated by the convergence of these two areas and by the need to bridge foundational theory with deployable AI systems.

The workshop is strongly aligned with the scope of AIMLSystems 2026, which emphasizes the interplay between AI/ML and systems, including scalable and efficient AI/ML, decentralized and distributed learning, and the emergence of new socio-techno-economic systems shaped by AI. The conference explicitly highlights how AI/ML both depends on advances in computational systems and creates new system-level requirements, making RL and game-theoretic perspectives especially relevant for multi-agent, resource-aware, and system-constrained settings.

This workshop goes beyond the main conference tracks by creating a focused venue for problems that cut across learning, incentives, mechanism design, strategic behavior, and system optimization. Many such topics do not fit neatly into standard ML or systems categories, despite being central to modern AI deployment.

Impact: Its novelty lies in bringing together researchers from reinforcement learning, multi-agent systems, algorithmic game theory, and AI systems to discuss both theory and practice. The expected impact is to stimulate new cross-disciplinary collaborations, identify open research directions, and accelerate principled methods for building reliable, strategic, and efficient AI systems in real-world environments.