Neural Information Processing Systems http://nips.cc/

The transition of control from autonomous systems to human drivers is critical in automated driving systems, particularly due to the out-of-the-loop (OOTL) circumstances that reduce driver readiness and increase reaction times. Existing takeover strategies are based on fixed time-based transitions, which fail to account for real-time driver performance variations. This paper proposes an adaptive transition strategy that dynamically adjusts the control authority based on both the time and tracking ability of the driver trajectory. Shared control is modeled as a cooperative differential game, where control authority is modulated through time-varying objective functions instead of blending control torques directly. To ensure a more natural takeover, a driver-specific state-tracking matrix is introduced, allowing the transition to align with individual control preferences. Multiple transition strategies are evaluated using a cumulative trajectory error metric. Human-in-the-loop control scenarios of the standardized ISO lane change maneuvers demonstrate that adaptive transitions reduce trajectory deviations and driver control effort compared to conventional strategies. Experiments also confirm that continuously adjusting control authority based on real-time deviations enhances vehicle stability while reducing driver effort during takeover.

Adversarial environments require agents to navigate a key strategic trade-off: acquiring information enhances situational awareness, but may simultaneously expose them to threats. To investigate this tension, we formulate a PursuitEvasion-Exposure-Concealment Game (PEEC) in which a pursuer agent must decide when to communicate in order to obtain the evader's position. Each communication reveals the pursuer's location, increasing the risk of being targeted. Both agents learn their movement policies via reinforcement learning, while the pursuer additionally learns a communication policy that balances observability and risk. We propose SHADOW (Strategic-communication Hybrid Action Decision-making under partial Observation for Warfare), a multi-headed sequential reinforcement learning framework that integrates continuous navigation control, discrete communication actions, and opponent modeling for behavior prediction. Empirical evaluations show that SHADOW pursuers achieve higher success rates than six competitive baselines. Our ablation study confirms that temporal sequence modeling and opponent modeling are critical for effective decision-making. Finally, our sensitivity analysis reveals that the learned policies generalize well across varying communication risks and physical asymmetries between agents.

Deep learning achieves state-of-the-art results in many tasks in computer vision and natural language processing.

Differential games [4, 6] involve multiple players controlling a dynamical system through their actions, which are described by differential state equations. These games evolve over a continuous-time horizon, where each player seeks to optimize an objective function that depends on the system's state, their own actions, and potentially the actions of others. In this study, we extend the classic differential game model to scenarios involving motion-payoff uncertainty, where players face uncertainties in both the dynamic equations and the payoff functions, and are unaware of certain parameters in the environment or in their opponents' payoff structures. In dynamic games, optimal control techniques are generalized to accommodate multiple players with both shared and conflicting interests. As shown in [9], if a set of interconnected partial differential equations--commonly referred to as the Hamilton-Jacobi-Bellman (HJB) equations--has solutions, then a Nash equilibrium can be achieved. At this equilibrium, no player can improve their outcome by unilaterally changing their strategy. However, traditional dynamic game models often assume that all players possess complete knowledge of the game. In many real-world scenarios, players face rapidly changing and uncertain environments, leading to incomplete information about the system's dynamics and payoffs [22, 3, 15, 1]. To address this uncertainty, we apply Bayesian updating methods, where players update their beliefs about unknown parameters as new information becomes available.

Dynamic game theory offers a toolbox for formalizing and solving for both cooperative and non-cooperative strategies in multi-agent scenarios. However, the optimal configuration of such games remains largely unexplored. While there is existing literature on the parametrization of dynamic games, little research examines this parametrization from a strategic perspective where each agent's configuration choice is influenced by the decisions of others. In this work, we introduce the concept of a game of configuration, providing a framework for the strategic fine-tuning of differential games. We define a game of configuration as a two-stage game within the setting of finite-horizon, affine-quadratic, AQ, differential games. In the first stage, each player chooses their corresponding configuration parameter, which will impact their dynamics and costs in the second stage. We provide the subgame perfect solution concept and a method for computing first stage cost gradients over the configuration space. This then allows us to formulate a gradient-based method for searching for local solutions to the configuration game, as well as provide necessary conditions for equilibrium configurations over their downstream (second stage) trajectories. We conclude by demonstrating the effectiveness of our approach in example AQ systems, both zero-sum and general-sum.

In Section 4.2, we have shown the effectiveness of In Section 3.4, we have analyzed that I2Q can easily solve the task with multiple optimal joint policies. Here, we give another way to solve this problem. D3G cannot obtain a winning rate in SMAC, as shown in Table 1. Although QSS value is a biased estimation in this implementation, the implementation without forward model is practical. The results are shown in Figure 16.

We propose a mesh-free policy iteration framework that combines classical dynamic programming with physics-informed neural networks (PINNs) to solve high-dimensional, nonconvex Hamilton--Jacobi--Isaacs (HJI) equations arising in stochastic differential games and robust control. The method alternates between solving linear second-order PDEs under fixed feedback policies and updating the controls via pointwise minimax optimization using automatic differentiation. Under standard Lipschitz and uniform ellipticity assumptions, we prove that the value function iterates converge locally uniformly to the unique viscosity solution of the HJI equation. The analysis establishes equi-Lipschitz regularity of the iterates, enabling provable stability and convergence without requiring convexity of the Hamiltonian. Numerical experiments demonstrate the accuracy and scalability of the method. In a two-dimensional stochastic path-planning game with a moving obstacle, our method matches finite-difference benchmarks with relative $L^2$-errors below %10^{-2}%. In five- and ten-dimensional publisher-subscriber differential games with anisotropic noise, the proposed approach consistently outperforms direct PINN solvers, yielding smoother value functions and lower residuals. Our results suggest that integrating PINNs with policy iteration is a practical and theoretically grounded method for solving high-dimensional, nonconvex HJI equations, with potential applications in robotics, finance, and multi-agent reinforcement learning.

Modeling vehicle interactions at unsignalized intersections is a challenging task due to the complexity of the underlying game-theoretic processes. Although prior studies have attempted to capture interactive driving behaviors, most approaches relied solely on game-theoretic formulations and did not leverage naturalistic driving datasets. In this study, we learn human-like interactive driving policies at unsignalized intersections using Deep Fictitious Play. Specifically, we first model vehicle interactions as a Differential Game, which is then reformulated as a Potential Differential Game. The weights in the cost function are learned from the dataset and capture diverse driving styles. We also demonstrate that our framework provides a theoretical guarantee of convergence to a Nash equilibrium. To the best of our knowledge, this is the first study to train interactive driving policies using Deep Fictitious Play. We validate the effectiveness of our Deep Fictitious Play-Based Potential Differential Game (DFP-PDG) framework using the INTERACTION dataset. The results demonstrate that the proposed framework achieves satisfactory performance in learning human-like driving policies. The learned individual weights effectively capture variations in driver aggressiveness and preferences. Furthermore, the ablation study highlights the importance of each component within our model.