Introduces a structure-aware solver synthesis method with a learned game representation that organizes solvability into a continuous geometry aligned with solver dynamics.
arXiv preprint arXiv:2011.00583 , year=
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
verdicts
UNVERDICTED 9roles
background 2polarities
background 2representative citing papers
New benchmark ICRL4AHT reveals that history-conditioned ICRL methods fail to show robust adaptation in multi-agent Overcooked-V2, underperforming random baselines on unseen teammates and layouts.
Derives ODE limits of Adam-DA showing that first- and second-order momentum parameters reverse their convergence roles in zero-sum games compared to minimization, validated on GAN experiments.
A projected gradient descent algorithm for noisy inductive matrix completion achieves linear convergence and stable recovery at sample complexity governed by side-information dimension, extending to inexact side-information with optimal error degradation.
The paper establishes equilibrium existence and uniqueness for nonlinear utility consumer networks under contraction conditions and proposes a shape-constrained isotonic regression approach with strict no-regret convergence for learning utilities in targeted monopoly pricing.
Proposes HAD-MFC framework that decouples upper-level vulnerable agent selection from lower-level adversarial policy learning in large-scale MARL using Fenchel-Rockafellar transform and MDP reformulation with provable optimality preservation.
Almost Greedy Fictitious Play achieves an instance-dependent O(1/T) convergence rate to Nash equilibrium in zero-sum games by optimizing stepsizes without discretization.
WebCQ applies cooperative MARL with QTRAN and DQN on semantic action vectors to web GUI testing, exploring 33.3% more states and 42.2% more actions than MARG on eight commercial sites.
SOM uses a Structural Causal Model to create an explicit graph of opponent observation-to-action links, allowing LLMs to reason along those paths for more accurate and stable predictions in multi-agent settings.
citing papers explorer
-
On the Geometry of Games and their Solvers
Introduces a structure-aware solver synthesis method with a learned game representation that organizes solvability into a continuous geometry aligned with solver dynamics.
-
Benchmarking the Limits of In-Context Reinforcement Learning for Ad-Hoc Teamwork
New benchmark ICRL4AHT reveals that history-conditioned ICRL methods fail to show robust adaptation in multi-agent Overcooked-V2, underperforming random baselines on unseen teammates and layouts.
-
Understanding Dynamics of Adam in Zero-Sum Games: An ODE Approach
Derives ODE limits of Adam-DA showing that first- and second-order momentum parameters reverse their convergence roles in zero-sum games compared to minimization, validated on GAN experiments.
-
Sample-efficient inductive matrix completion with noise and inexact side-information
A projected gradient descent algorithm for noisy inductive matrix completion achieves linear convergence and stable recovery at sample complexity governed by side-information dimension, extending to inexact side-information with optimal error degradation.
-
Equilibrium and Pricing in Consumer Networks with Nonlinear Utilities: An Online Shape-Constrained Learning Approach
The paper establishes equilibrium existence and uniqueness for nonlinear utility consumer networks under contraction conditions and proposes a shape-constrained isotonic regression approach with strict no-regret convergence for learning utilities in targeted monopoly pricing.
-
Vulnerable Agent Identification in Large-Scale Multi-Agent Reinforcement Learning
Proposes HAD-MFC framework that decouples upper-level vulnerable agent selection from lower-level adversarial policy learning in large-scale MARL using Fenchel-Rockafellar transform and MDP reformulation with provable optimality preservation.
-
Do Not Discretize, Optimize: Almost Greedy Fictitious Play
Almost Greedy Fictitious Play achieves an instance-dependent O(1/T) convergence rate to Nash equilibrium in zero-sum games by optimizing stepsizes without discretization.
-
WebCQ: Cooperative Multi-Agent Deep Reinforcement Learning for Scalable Web GUI Testing
WebCQ applies cooperative MARL with QTRAN and DQN on semantic action vectors to web GUI testing, exploring 33.3% more states and 42.2% more actions than MARG on eight commercial sites.
-
SOM: Structured Opponent Modeling for LLM-based Agents via Structural Causal Model
SOM uses a Structural Causal Model to create an explicit graph of opponent observation-to-action links, allowing LLMs to reason along those paths for more accurate and stable predictions in multi-agent settings.